Chapter 2

Low Temperature Combustion Engines

Abstract Current stringent emission legislation, market and environmental concern

governs the automotive research for developing high-efficiency and low emission
engines. Vehicle manufacturers meet the present emission norms using a combination
of in-cylinder emission reduction methods and exhaust after-treatment devices. Pro-
posed future emission norms are even more stringent; and thus, newer technologies
are required to satisfy emission standards worldwide. Low temperature combustion
(LTC) engines have potential to deliver higher fuel conversion efficiency and simul-
taneous reduction of NOx and soot emissions to an ultralow level. The LTC engines
can also reduce the heavy dependence on NOx and soot after-treatment devices for
meeting the emission norms. In this chapter, LTC principles along with different
proposed LTC strategies are discussed in detail. The LTC strategies are broadly
categorized into homogeneous charge compression ignition (HCCI) and partially
stratified charge compression ignition (SCCI). The SCCI strategy is further classified
as thermal stratification- and fuel concentration stratification-based combustion pro-
cess. The partially premixed compression ignition (PPCI) and reactivity-controlled
compression ignition (RCCI) are the two main engine combustion modes with fuel
concentration stratification are discussed in this chapter. Thermally stratified com-
pression ignition mode is also described in the present chapter.

Keywords LTC • HCCI • Stratification • Combustion • Heat release • LTHR •

HTHR • PCCI • EGR • SACI • Ignition • RCCI • Dual fuel • Direct injection

2.1 Low Temperature Combustion Principle

Presently, internal combustion (IC) engines are the most preferred option for
automotive powertrain. Two commonly used engine types are spark ignition (SI)
and compression ignition (CI) engines, running on gasoline and diesel, respectively.
The CI engines have comparatively higher thermal efficiency due to higher com-
pression ratio and lean (equivalence ratio<1) and unthrottled engine operation.
Therefore, CI engines are preferred choice for medium- and heavy-duty applica-
tions. The in-cylinder NOx, CO and unburned hydrocarbon (HC) formation is
higher in SI engines, and these emissions from tailpipe can be reduced to legislation

© Springer International Publishing AG 2018 31

R.K. Maurya, Characteristics and Control of Low Temperature Combustion Engines,
Mechanical Engineering Series, https://doi.org/10.1007/978-3-319-68508-3_2
32 2 Low Temperature Combustion Engines

limits level by effective operation of three-way catalytic converter (TWC). The

conventional CI engines have mainly high emissions of NOx and particulate matter
(PM), and also suffer from the NOx-PM trade-off [1]. The regulated limits of NOx
and PM emissions are drastically reduced all over the world in the last 20 years (see
Fig. 1.3). The early emission norms were complied by in-cylinder emission reduc-
tion strategies such as higher fuel injection pressure, exhaust gas recirculation
(EGR), optimal fuel injection timings, improved design of combustion chamber,
boosting intake air, etc., without any exhaust after-treatment technologies [2]. The
present and future emission norms are fulfilled or expected to be accomplished by a
combination of in-cylinder strategies as well as exhaust after-treatment devices.
Adaptation of presently available after-treatment technologies faces challenges of
higher cost, fuel economy penalties, durability issues and larger space requirements
in comparison to the engines without exhaust after-treatment technologies
[3]. Additionally, there is a requirement of higher fuel conversion efficiency at
minimal cost to compete in the market. Consequently, in-cylinder combustion
strategies need to be significantly improved for higher thermal efficiency as well
as further reduction in engine-out emissions so that dependency on the after-
treatment technologies can also be reduced.
In-depth understanding of in-cylinder processes is required for the development
of appropriate technology to meet the present and future emission legislation limits
while maintaining higher thermal efficiency [4]. Various studies are conducted
using advanced laser-based combustion diagnostic techniques to understand the
in-cylinder processes during diesel combustion in CI engines. A conceptual model
for quasi-stationary diesel spray combustion was proposed based on laser sheet
imaging (Fig. 2.1a) [5]. This model presented a liquid fuel penetration length much
shorter than penetration length estimated by earlier studies. Furthermore, no drop-
lets are present in the combustion zone; only fully vaporized fuel is present. Just
downstream of the liquid part of the diesel spray, a rich premixed combustion zone
appeared. Soot particles are formed in regions of locally rich mixture, where fuel
and air first reacted, and then this rich mixture burns out in a high temperature
diffusion flame at the jet periphery, leading to NOx formation [10]. Soot formation
starts in the fuel-rich premixed zone (Fig. 2.1a). Rate of soot formation is dependent
on equivalence ratio in this zone, which is determined by flame lift-off length.
Flame lift-off characteristics play an important role in combustion and emission
processes in diesel engines [11]. Flame lift-off length is the distance between the
nozzle hole and the most upstream part of the turbulent diffusion flame. Most of the
surrounding air entrainments into the jet are through this distance (lift-off length).
Air entrained downstream of the lift-off length reacts (oxygen is consumed) in the
diffusion flame, and therefore entrainment into the jet interior is very limited in the
burning portion of the jet (Fig. 2.1a). Equivalence ratio (φ) inside the jet is lower in
case the more air entrained into the spray up to the lift-off length, which leads to
lesser soot formation [12]. Thus, equivalence ratio at lift-off plays an important role
in soot formation during spray combustion. Flame lift-off characteristics depend on
ambient (cylinder gas) temperature and density, injector orifice diameter, injection
pressure and oxygen concentration [13, 14]. Pickett et al. [13] derived an empirical
relation to estimate lift-off length (H) given by Eq. (2.1):
2.1 Low Temperature Combustion Principle 33

Fig. 2.1 (a) Illustration of NOx and soot formation processes in diesel combustion in CI engines
(Adapted from [5–7]). (b) LTC operating region on φ-T map (Adapted from [4, 6, 8, 9])

H ¼ CT 3:74
a ρa0:85 d 0:34 U1 Z1
st ð2:1Þ

where C is a proportionality constant, Ta [K] is ambient gas temperature, ρa [kg/m3]

is ambient gas density, d [μm] is injector tip orifice diameter, U[m/s] is injection
velocity and Zst is stoichiometric mixture fraction, which accounts for effects of
34 2 Low Temperature Combustion Engines

ambient gas oxygen concentration. Mean cross-sectional φ of a model jet as a

function of distance from nozzle of injector can be estimated by Eq. (2.2) (proposed
by Naber and Siebers [15]) for quantitative estimate air entrainment effect in spray:

 2ðA=FÞs
Φ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi ð2:2Þ
1 þ 16ðx=xþ Þ2  1
qffiffiffiffi pffiffiffi  
where xþ ¼
ρf ca d
and tan
α
¼ ac 0:19 qffiffiffiffi
ρa ρa
ρa tan ðαÞ
2
2 1 ρf  C2 ρf

where ρf and ρa are the fuel and ambient densities, respectively. Ca is the area
contraction coefficient of the nozzle, d is the nozzle hole diameter and α is the jet
spreading angle. Actual equivalence ratio (φ) at flame lift-off length can be
estimated using Eqs. (2.1) and (2.2). This estimated φ can give an idea about the
soot formation during spray combustion.
Computational studies analysed the variations of soot formation on the local
equivalence ratio–temperature (φ-T ) map as depicted in Fig. 2.1b [8, 16]. It is found
that for φ < 2, no soot is formed irrespective of the combustion temperature and no
soot formation is observed below 1500 K irrespective of the φ (Fig. 2.1b). The
maximum soot formation tendency is around 1800–2000 K, and at higher temper-
atures, the sooting tendency is reduced for a given equivalence ratio. The φ-T
correlation is applicable to a homogeneous cylinder charge or the locally homoge-
neous regions of a heterogeneous cylinder charge.
Figure 2.1b illustrates the dependency of the NOx-soot formation and soot
oxidation on the combustion temperature and local equivalence ratio in the cylin-
der. Size and shape of the soot formation area in φ-T map are fuel dependent as
presented by Kitamura et al. [8]. However, the NOx formation area is not fuel
dependent in φ-T map. Figure 2.1b depicts that the conventional SI engine operated
on stoichiometric mixture produce significant amount of NOx emissions (shown in
NOx region in φ-T map). In conventional CI engines, combustion reactions start in a
richer mixture (about φ ¼ 4) and complete in a stoichiometric (φ ¼ 1) diffusion
flame [4]. Flame is prone to initialize and propagate towards the locally stoichio-
metric regions. The stoichiometric burning tends to produce very high flame
temperatures. The combustion zones fall in the high soot and NOx formation areas
in φ-T map for conventional CI engines (Fig. 2.1b), which leads to higher NOx and
PM emissions. To reduce the NOx formation, if the combustion temperature is
lowered, then soot formation kinetics is also slowed down [17, 18]. However, at
lower combustion temperature, in-cylinder soot oxidation rate also decreases more
than the formation rates [19], and hence the exhaust gas contains more PM in the
diesel engines. Studies confirm that lowering combustion temperature by using
EGR in CI engines initially increases the emission of PM [20]. For very high EGR
rate operating conditions, soot formation rate can be too low so that PM emission
can decrease even with lower oxidation rates [16, 17, 19, 20]. Another potential
challenge arises by low engine combustion temperature is higher levels of CO and
HC emissions. The CO and HC formations are dependent on temperature and can
2.1 Low Temperature Combustion Principle 35

form in lean as well as rich fuel–air mixtures. Higher combustion temperature leads
to near-complete oxidation of CO and HC in leaner conditions, but richer condi-
tions result in partial oxidation of HC to CO resulting into higher CO emissions. At
lower combustion temperature, richer mixture leads to higher CO and HC emis-
sions (Fig. 2.1b). There exists a region in φ-T map with comparatively lower CO
and HC emissions apart from soot–NOx region called low temperature combustion
(LTC) region (Fig. 2.1b). Thus, both NOx and soot formations can be simulta-
neously avoided in LTC, which is inevitable in conventional high temperature
combustion (HTC) processes in compression ignition engines.
In the pursuit of developing an engine having ability to yield ultralow NOx and
soot emission along with higher thermal efficiency, several new CI combustion
strategies have been demonstrated in the last two decades. A common name for
these combustion strategies is LTC mode. Figure 2.2 presents the evolution of

Fig. 2.2 Evolution of different combustion strategies in IC engines

36 2 Low Temperature Combustion Engines

different combustion strategies for IC engines. In LTC strategy, adequate premixing

between the fuel and air is created to avoid diffusion combustion in a globally dilute
environment. The combustion of lean fuel–air mixture leads to lower combustion
temperatures in the cylinder. In LTC engines, charge dilution (with either air or EGR)
is typically used to avoid high NOx formation by lowering combustion temperature in
the cylinder. In LTC engines, providing more time (3–20 ms) for premixing fuel and
air evades high soot formation rates due to lower local φ achieved by mixing.
However, overmixing of fuel and air results into very low φ, which leads to very
high HC and CO emission due to very low combustion temperature [6]. In-cylinder
control of soot emissions is a very complex process in LTC engines. Soot emission
depends on quality of fuel–air mixing, combustion temperature and availability of
oxygen in the charge. Therefore, these parameters can be utilized for controlling the
in-cylinder soot formation. Lowering combustion temperature by dilution and
increasing premixing by prolonged ignition delay are two typical methods for soot
control. Another strategy to control the soot emission is by controlling the reactivity
of fuel–air mixture by reformulating the fuel composition (oxygen content, C/H ratio,
octane/cetane) because soot region in φ-T map is dependent on fuel also (Fig. 2.1b).
Most of the LTC concepts have utilized all the three parameters with a special focus
on any one parameter in a particular strategy.
Various LTC concepts have been investigated with several acronyms and names
(Fig. 2.2) in published literature. The LTC strategies can be divided into two main
categories, namely, homogeneous charge compression ignition (HCCI) and partially
stratified charge compression ignition (SCCI) based on degree of premixing of fuel–
air mixture. In the HCCI strategy, charge is well mixed before compression stroke
and engine is operated on lean mixture. In the SCCI strategy, possibly two types of
stratification occur, namely, thermal stratification and fuel concentration stratification
in the cylinder. Thermal stratification is used in thermally stratified compression
ignition (TSCI) and spark-assisted compression ignition (SACI) strategy. Fuel con-
centration stratification is possible by single as well as dual fuel. In the fuel stratifi-
cation category, gasoline partially premixed combustion (PPC) and dual fuel
reactivity-controlled compression ignition (RCCI) are the two main combustion
concepts, which are widely investigated due to their potential of better combustion
phasing control and lower emission along with higher thermal efficiency. In the last
few years, PPC and RCCI strategies gained significant advancement such as multiple
injection, dual direct injection fuel stratification (DDFS), etc. Other combustion
strategies investigated with gasoline-like fuels and different injection strategies are
published with different acronyms such as gasoline compression ignition (GCI), low
temperature gasoline combustion (LTGC), gasoline direct-injection compression-
ignition (GDCI) (see Fig. 2.2). Detailed discussion of all the mentioned LTC strat-
egies is presented in the next sections of the present chapter.
2.2 Homogeneous Charge Compression Ignition 37

2.2 Homogeneous Charge Compression Ignition

2.2.1 HCCI Fundamentals

Homogeneous charge compression ignition (HCCI) is one of the engine operating

modes in LTC region (Fig. 2.1b). The HCCI mode means the compression ignition
(auto-ignition/self-ignition) of homogeneous (premixed) fuel and air mixture in the
engine cylinder. Hence, HCCI operating mode exhibits characteristics from both CI
and SI engines. The HCCI operating mode is similar to conventional SI engine in
charge preparation strategy (both modes use premixed charge) and similar to CI
engines in combustion initiation process (in both modes combustion starts with
autoignition). Even though HCCI operation mode called “homogeneous” uses
premixed charge, in real engines, there are always thermal inhomogeneities (due
to wall heat transfer and convection) and fuel concentration inhomogeneities
(difficulties in preparing fully homogeneous charge) [21]. Figure 2.3b shows the
HCCI combustion process vis-
a-vis CI and SI combustion processes. In HCCI
combustion, nearly homogeneous charge (fuel–air mixture) enters in the engine
cylinder during intake stroke (Fig. 2.3a). During compression stroke pressure and
temperature of charge increases, and auto-ignition reactions start when temperature
reaches towards auto-ignition temperature (close to TDC). When auto-ignition
temperature reaches, entire charge in cylinder reacts simultaneously. However,
combustion process takes finite amount of time, even though entire cylinder charge
is active. In-cylinder imaging using laser-induced fluorescence (LIF) of OH radical
(for reaction activity) and planar laser-induced fluorescence (PLIF) of fuel showed
that onset of HCCI combustion has very large local spatial and temporal variations.
These variations are present even with premixed charge created using either
external mixing tank or port fuel injection (PFI) system [22]. Generally, premixed
charge created using PFI has higher inhomogeneity in comparison to premixed
charge created by external mixing tank. Nonoccurrence of simultaneous combus-
tion throughout the combustion chamber is also a result of thermal stratification
caused by wall heat transfer and turbulent mixing during compression stroke, which
is experimentally confirmed by chemiluminescence imaging [23, 24]. HCCI com-
bustion occurs sequentially starting with hottest zone (also referred to as “hot spots”
[25], see Fig. 2.3a) where most favourable conditions for auto-ignition exist,
followed by next hottest zone [21]. Due to the start of exothermic reactions, charge
temperature increases, and hence, oxidation reactions become even faster providing
local positive feedback to the charge temperature [25]. Sufficient time is not
available to distribute all the generated heats (in the hot spot) to surrounding cold
bulk gases (due to fast temperature feedback), which leads to gradual amplification
of small inhomogeneities in the cylinder [24, 25]. The amplification of small
inhomogeneities is confirmed and observed as a large structure in experimental
chemiluminescence images of HCCI combustion. Hot spot size in HCCI combus-
tion is found to be of the same order as the turbulence integral length scale in
combustion chamber [25]. The HCCI combustion changes the combustion
38 2 Low Temperature Combustion Engines

Fig. 2.3 (a) The HCCI combustion process. (b) Combustion process comparison of SI, CI and
HCCI engines in four-stroke cycle

behaviour during the process, and Hultqvist et al. [24] observed three distinct
combustion modes in the HCCI combustion process. First, the initial combustion
process starts with evenly distributed number of regions in the combustion cham-
ber, where small inhomogeneities in fuel or temperature are favourable to the auto-
ignition reactions. Fuel is gradually consumed by local reactions in these regions
without flame propagation because of no sharp borders are observed between
burned and unburned gases, and the fuel concentration gradually decreases with
little or no expansion of the reaction zone. Second, increased chemical activity is
due to the start of exothermic reactions lead to positive feedback to the charge
temperature. The initial ignition kernels grow larger as combustion progresses.
Study also suggested that the number and locations of kernels are reliant on global
parameters like distribution of air, fuel, residual gas and charge temperature. Third,
the formation of new ignition kernels at locations that have become favourable
because of the global temperature and pressure rise is due to combustion progress.
2.2 Homogeneous Charge Compression Ignition 39

These new ignition kernels grow until they merge with other kernels in the
combustion chamber.
Due to combustion progress in entire combustion chamber, combustion duration
is comparatively shorter in HCCI engine, and combustion occurs close to TDC
position. Work is obtained in expansion stroke, and cycle is completed by exhaust
stroke similar to conventional engine by expelling the exhaust gases out of the
cylinder (Fig. 2.3a). Comparison of HCCI, CI and SI combustion in a four-stroke
cycle is presented in Fig. 2.3b. In HCCI operation mode, power output is controlled
by varying the fuel flow rate, and engine is operated unthrottled similar to CI
engines. The HCCI engine is operated on globally leaner mixture in order to control
the combustion rate, which leads to lower combustion temperature. Premixed
charge and lower combustion temperature in HCCI combustion process leads to
ultralow NOx and soot emissions. However, HCCI engines emit comparatively
higher HC and CO emissions, which can be easily mitigated by oxidation catalyst.
There is no direct control on ignition timing and combustion rate in HCCI com-
bustion process similar to direct control of spark timing and injection timing in the
conventional engine combustion processes. In HCCI engines, combustion timing
and combustion rate are governed by chemical kinetics [26, 27]. The HCCI
combustion process does not have either visible flame propagation (like SI engines)
or highly stratified diffusion combustion (like diesel engines). Idea of flame prop-
agation in HCCI combustion process is rejected by analysing the PLIF and chemi-
luminescence images as new “hot spots” generated randomly and the structures
appeared in the images are not moving from image to image [23, 24]. A study
summarized the two main combustion chamber condition requirements for the
HCCI combustion process as follows: (i) auto-ignition temperature of premixed
charge should be reached towards the end of compression stroke, and (ii) the fuel–
air mixture should be diluted enough either by air or residuals to achieve reasonable
HCCI combustion rate [25].
Historically, combustion process similar to HCCI combustion is known from last
more than 100 years [28]. A hot-bulb two-stroke oil engine was designed where
premixed vaporized fuel–air mixture was created in a heated prechamber by
injecting fuel onto the surface of hot bulb. Fuel–air mixture created in prechamber
was autoignited in the main chamber (cylinder) [28]. Another study achieved
controlled autoignition in cylinder by injecting active species created from partially
burned mixture in a prechamber [29]. In the early investigations of HCCI combus-
tion process, the most recognized and first systematic studies were conducted on
two-stroke engines [30, 31]. To overcome the problems of two-stroke engines (high
residuals at low loads and tendency for run-on combustion), Onishi et al. developed
a combustion process named “active thermo-atmospheric combustion (ATAC)”
which depends on high residual levels and high initial charge temperature
[30]. They found significant improvement in fuel economy and significantly
lower emissions in ATAC mode. Concurrent with this study, Nouguchi et al.
conducted spectroscopic analysis of combustion process in opposed piston
two-stroke engine named Toyota-Soken (TS) [31]. They noted the overall fast
combustion rate, excellent fuel economy and lower engine emissions. Study
40 2 Low Temperature Combustion Engines

detected the high levels of intermediate species (CH2O, HO2 and O radicals) well
before auto-ignition in the combustion chamber, which was the result of low
temperature oxidation chemistry of larger paraffinic hydrocarbon fuels. After the
start of combustion, high concentrations of CH, H and OH radicals are found, which
indicates high temperature oxidation chemistry during bulk combustion. After
pioneering the work of Onishi and Noguchi, several follow-up studies were
conducted [32–34], and first production of two-stroke HCCI engine was developed
with combustion name “active radical combustion (ARC)” [35, 36]. Based on
previous investigations on two-stroke engine, Najt and Foster conducted first
study on four-stroke engine to advance the understanding of underlying physics
of HCCI combustion [37]. Another study further extended the HCCI combustion
investigation on four-stroke engine, and first introduced the terminology “homo-
geneous charge compression ignition (HCCI)” [38]. Since the 1990s, the HCCI
combustion had developed into a great research topic worldwide [39–41]. The
HCCI combustion process is also known with another terminology called “con-
trolled auto-ignition (CAI)” process [42]. The CAI process used variable valve
timing (VVT) to trap large amount of residual gases in chamber to achieve auto-
ignition of premixed gasoline in the cylinder. In CAI process, ignition timing is
controlled by amount of residual gases trapped in the cylinder.
The HCCI engine operation has been investigated with significant success in
two-stroke and four-stroke engines using liquid and gaseous fuels [40–43]. Both
gasoline-like fuels and diesel-like fuels can achieve the HCCI combustion using
appropriate strategies for charge preparation (see Chap. 4) and combustion phasing
control strategy (see Chap. 5). Volatile fuels make premixed charge preparation
strategy simple and easier (PFI, fumigation). Less volatile fuels require com-
plex fuel injection strategies (direct injection) for charge preparation. To achieve
auto-ignition in the combustion chamber, several actuation strategies such as
variable compression ratio (VCR), intake heating, VVT, EGR, etc. are used, though
each technique has its own merits and demerits [44]. Auto-ignition characteristics
in HCCI combustion process also depend on fuel, and HCCI combustion charac-
teristics are governed by fuel oxidation chemistry [41]. Auto-ignition of different
fuels is achieved through few categories of reaction pathways characterized by
temperature range over which certain reactions occur.

2.2.2 HCCI Auto-Ignition and Heat Release

Autoignition quality of the fuel is an important factor that influences all the current
HCCI engine designs. Auto-ignition can be defined as the ignition of a fuel–air
mixture because of heat produced from exothermic oxidation reactions without the
involvement of external energy sources such as a spark or a flame [45, 46]. The
HCCI engine operation has been achieved with fuels having a wide range of auto-
ignition properties ranging from higher octane gasoline-like fuels to higher cetane
diesel-like fuels. Several studies used primary reference fuels (PRFs) containing a
2.2 Homogeneous Charge Compression Ignition 41

Fig. 2.4 Illustration of single- and two-stage heat release in HCCI combustion for two different
fuels (Adapted from [50])

blend of n-heptane and iso-octane for investigation of fundamental HCCI combus-

tion characteristics [47, 48]. Generally, iso-octane is used as a surrogate for
gasoline and n-heptane is used as surrogate fuel for diesel. Auto-ignition charac-
teristics of fuel or fuel constituent in HCCI engine operation can be typically
categorized into single-stage ignition and two-stage ignition process [49]. Figure 2.4
demonstrates the two different types of ignition in HCCI combustion for two
different auto-ignition characteristics of fuels. Two-stage ignition fuel has separate
region with a small heat release in low temperature regime around 10–20 CAD
before the main heat release in high temperature regime. The time delay between
two stages of heat release is known as “negative temperature coefficient (NTC)”
regime. However, single-stage ignition fuels do not have early heat release region.
Typically, lower octane fuels (less auto-ignition resistant) such as diesel, n-heptane,
PRF80 (80% iso-octane and 20% n-heptane) and DME exhibit two-stage ignition,
while higher octane fuels (high auto-ignition resistant) such as ethanol, methanol,
butanol, iso-octane and natural gas typically exhibit single-stage ignition [27, 39,
51]. It is important to note that the main autoignition process does not take place
instantaneously for both types of fuels. In single-stage auto-ignition fuels, oxidation
reactions start progressively once charge temperature reaches around 950–1050 K
due to compression of the charge by piston [49, 50]. The charge temperature
initially increases slowly because of intermediate temperature heat release
(ITHR) reactions [49, 52], and after attaining the thermal runaway point, charge
temperature increases quickly because of the high temperature heat release (HTHR)
reactions [49]. In two-stage auto-ignition fuels, low temperature heat release
42 2 Low Temperature Combustion Engines

(LTHR) or “cool-flame” oxidation reactions take place at the start of ignition when
temperature is in the range of 760–880 K before attaining the ITHR phase
[53]. After LTHR, similar to single-stage heat release, ITHR takes place in
two-stage ignition fuels also. The HCCI engine operation by two-stage ignition
fuels needs lower charge preheating or lower compression ratios because LTHR
raises the charge temperature in the cylinder [49].
Combustion/oxidation of hydrocarbon fuels occurs through a classical chain
reaction mechanism consisting of several individual reactions with many interme-
diate species created during oxidation reaction, which converts fuel into combus-
tion products. Individual reactions during combustion are also known as elementary
reactions. Typically, four fundamental types of elementary reactions, namely,
initiation, propagation, branching and termination reactions, occur during combus-
tion. Elementary initiation reactions produce unstable radical species from the
breakdown of stable species. The step in which free radical produces another
radical by reaction is called chain propagation reaction. There is no change in the
radical concentration in chain propagation reactions. Reactions producing more
free radicals than consumption of radicals are known as chain branching reactions,
and this type of reactions results in very high combustion rates. When entire
reactants are consumed, free radicals combine with one another to produce stable
species by terminating the reactions known as chain termination reactions [54].
During the compression stroke, temperature inside the cylinder increases grad-
ually, and no significant reaction occurs until the charge temperatures reach about
550 K [51]. As piston further moves towards TDC (charge temperature further
increases), first initiation reactions create small amount of radical species, which
abstract hydrogen atom from fuel molecule at temperature below 850 K leading to
production of alkyl radicals (R) by reaction R1 [27]:

RH þ OH ! R  þ H2 O ðR1Þ

where radical species are denoted as the “” symbol next to character. Different
molecular structures of paraffin lead to the different types of alky radicals by
hydrogen abstraction. The alky radicals are consumed in two parallel paths
shown by reactions R2 and R3 [55]. One pathway is to produce a conjugate olefin
and a hydroperoxy radical (from reaction R2), and an alternative other pathway is to
form alkylperoxy radical (RO2) by reaction R3.
R  þO2 ! olefin þ HO2  ðR2Þ
R  þO2 $ RO2  ðR3Þ

The reaction R3 is a reversible reaction, and very important reaction for the
HCCI auto-ignition process. Reaction R3 has different reaction rates constant in
reverse and forward directions. Reverse reaction requires high activation energy
because the oxygen–alkyl radical bond must be broken to proceed the reaction.
However, forward reaction needs lower activation energy to proceed the reaction,
and hence its rate is affected only by the concentration of alkyl radical and oxygen
[51]. Equilibrium of the reversible reaction R3 depends on both the temperature and
2.2 Homogeneous Charge Compression Ignition 43

pressure of cylinder. At higher temperature, equilibrium shifts towards reac-

tants side because reverse reaction is faster than forward reaction. In this condition,
concentration of alkylperoxy radical decreases drastically, and low temperature
chain branching reactions are quenched leading to very low combustion rate
[51]. This phenomenon is responsible for NTC regime in HCCI combustion using
two stage heat release fuel.
The alkylperoxy radical undergoes internal isomerization by the oxygen
abstracting a hydrogen atom from a C–H bond elsewhere within the oxygen
molecule to form a hydroperoxyalkyl radical (QOOH) [55]:

RO2  ! QOOH ðR4Þ

The succeeding reactions of the QOOH depend on the structure of the QOOH,
and several possibilities exit for the next reaction. The QOOH radical is consumed
through two basic routes [55]. In the first route, an OH radical is produced when
QOOH radical either spontaneously decomposes into a lower-molecular-weight
alkene or forms a cyclic ether and OH radical (R5):

QOOH ! CARBONYL þ R0  þ OH  ðR5Þ

In the other route, the QOOH radical undergoes a second oxygen addition to
form a hydroperoxyalkylperoxy radical (OOQOOH), followed by an internal
hydrogen atom abstraction to produce an alkyl hydroperoxide which decomposes
into an aldehyde and two OH radicals (R6–R7):

QOOH þ O2 $ O2 QOOH ðR6Þ

O2 QOOH ! HOOROOH ðR7aÞ
0
HOOROOH ! RCHO þ R O þ OH  þ OH  ðR7bÞ

It can be noticed that reaction R6 is also a reversible reaction and has similar
characteristics of reaction R3. Equilibrium of these reactions depends on pressure
and temperature of the combustion chamber, and when charge temperature reaches
about 800–850 K, the net addition reactions of oxygen molecule to radicals
completely stop. Silke et al. also confirmed that reactions R3 and R6 had the largest
effect on the required BDC temperature for constant combustion phasing at differ-
ent boost pressure conditions because LTHR affects the requirement of inlet
temperature for the auto-ignition of fuels in HCCI engines [56]. Reactions R1–R7
define the low temperature oxidation mechanism of a large paraffinic fuel.
Figure 2.5 summarized the reaction pathway at different temperature ranges for
oxidation of n-heptane [57, 58].
Due to further increase in compression temperature (above 850 K) intermediate
temperature regime reaction R2 produces more olefinic hydrocarbons and
hydroperoxy radicals (HO2). Hydrogen peroxide (H2O2) formation becomes sig-
nificant by HO2 radicals by abstracting hydrogen atom from fuel (by reaction R8),
and concentration of H2O2 increases gradually until charge temperature reaches to
1000 K [51]:
44 2 Low Temperature Combustion Engines

Fig. 2.5 Reaction pathways of n-heptane oxidation at various temperature ranges [57, 58]

RH þ HO2  $ R  þH2 O2 ðR8Þ

Further increase in the charge temperature (above 1000 K) leads to the decom-
position of H2O2 into hydroxyl radical by reaction R9, and this process quickly
releases amount of hydroxyl radical resulting into higher overall reaction rate and
heat release:

H2 O2 þ M ! OH  þOH  þM ðR9Þ

The decomposition of H2O2 into OH radicals and their consequent reactions

leads to the increase in charge temperature, which further accelerates the decom-
position of H2O2. This phenomenon leads to an auto-ignition process, which
increases the temperature above 1200 K in the high temperature reaction regime
[27]. Figure 2.6 presents the experimental and simulated distribution of H2O2 and
HO2 in the cylinder with respect to crank angle (CA) position in HCCI engines
[59]. It is observed that concentration of H2O2 is higher during intermediate heat
release stage. Another consequence of the hydrogen peroxide branching reaction is
that the intermediate and hot ignition temperatures are comparatively independent
of the fuel characteristics [27].
The high temperature oxidation of hydrocarbons can be described in sequential
three steps [55] shown in reaction R10. First parent fuel is converted into lower
hydrocarbon and then, intermediate species further converted to produce CO and
water. Lastly, CO is oxidized to CO2 releasing large fraction of heat energy:

RH þ OH ! R0  R0  þH2 O ! CO þ H2 O ! CO2 þ heat ðR10Þ

2.2 Homogeneous Charge Compression Ignition 45

Fig. 2.6 Variation of measured and simulated H2O2 and HO2 with crank shaft position in HCCI
engine [59]

The important chain carrying radicals in high temperature regime are OH, H and
O [55]. These radicals are very reactive, and hence, hydrocarbon oxidation rates are
extremely fast. The important chain branching reactions in the high temperature
regime are R11, R12 and R13:

H  þO2 ! OH  O  ðR11Þ
O  þH2 ! OH  þH  ðR12Þ
H2 O þ O ! OH  þOH  ðR13Þ

The final stage of the energy release at the high temperature is the oxidation of
CO, and principal reactions involved in CO oxidation are R14, R15 and R16 [55]:

CO þ OH ! CO2 þ H  ðR14Þ

CO þ HO2  ! CO2 þ OH  ðR15Þ
CO þ O  þM ! CO2 þ M ðR16Þ

At high temperature, OH (R14) mainly consumes the entire CO and other
reactions have only small contribution in CO oxidation.
Different chemical kinetic mechanisms are used to analyse combustion charac-
teristics and auto-ignition timings for various fuels. Chemical kinetic mechanisms
can be grouped into four general classes: (i) single-step mechanism (no elementary
reactions), (ii) generalized mechanisms (no track of individual reactive species),
(iii) reduced mechanisms and (iv) detailed mechanisms. First two types of mech-
anisms are empirical approaches. Reduced mechanisms are derived from detailed
mechanisms by conducting the sensitivity analysis. Sensitivity analysis determines
46 2 Low Temperature Combustion Engines

Fig. 2.7 Reduced oxidation mechanism of n-heptane and iso-octane mixtures in HCCI combus-
tion [48]

the most important reactions for a given process. Detailed mechanisms are the most
complex reaction mechanism. Detailed oxidation mechanism attempt to incorpo-
rate all the chemical details consisting hundreds of species, well over a thousands of
elementary reactions [55]. To understand the heat release process or predict the
emissions from HCCI combustion, detailed mechanism or appropriately reduced
mechanism is essential. Mostly surrogate fuels are used for chemical kinetic
analysis consisting of n-heptane and/or iso-octane. Figure 2.7 presents a proposed
reduced oxidation reaction mechanism for mixtures of n-heptane, and iso-octane in
HCCI combustion consists of 32 species and 55 reactions [48]. The reaction
mechanisms for both paraffins are quite similar. The figure shows that the HCCI
combustion occurs, when H2O2 starts to decompose at a significant rate (which
triggers ignition) and produce OH radicals.
The high octane industrial gasoline exhibits different characteristics than PRFs
in HCCI ignition process. An industry average gasoline consists of roughly 60%
paraffins, 10% olefins and 30% aromatics [60]. Aromatic reactions play an impor-
tant role in gasoline auto-ignition process. Figure 2.8a depicts the heat release curve
for gasoline at three different equivalence ratios in HCCI engine operated at an inlet
temperature of 70  C and compression ratio of 13.5 [61]. Figure depicts three
different stages of heat release: (i) weak cool flame heat release (denoted as C),
2.2 Homogeneous Charge Compression Ignition 47

(ii) in-between pre-ignition (denoted as P) and (iii) final ignition (denoted as F). To
explain this phenomenon, chemical kinetic simulation is conducted using gasoline
95 (gasoline octane number 95), PRF95 (contains 95% iso-octane and 5%
n-heptane) and a surrogate fuel containing 11 vol% n-heptane, 59 vol%
iso-octane and 30 vol% toluene. It is found that gasoline 95 and surrogate fuel
show similar three-stage heat release which is not there in PRF95 [61]. A phenom-
enon called “obstructed pre-ignition (OPI)” is defined, which is responsible for
delay in the final ignition of gasoline and surrogate. Typically, HTHR occurs due to
decomposition of H2O2 leading to formation of large amount of OH radicals. This
process is obstructed by consuming the OH radicals in formation of relatively stable
benzyl radical by toluene present in the fuel. The benzyl radical formation competes
with the reaction of OH radicals with the other alkanes and olefins. Therefore,
benzyl radical acts as a sink for the OH radicals formed in the cylinder, which leads
to reduction in the combustion rate that delayed the final heat release [61].
Figure 2.8b summarized the autoignition process of surrogate fuel consisting of
n-heptane, iso-octane and toluene. The figure depicts the influence of the obstructed
pre-ignition phenomenon with the most important reactions and species (shown
within a dashed square). The obstructed pre-ignition heat release occurs between
the cool flame and final ignition stages in gasoline as well as three-component
surrogate fuel of gasoline (containing toluene). This type of heat release is not
observed in the fuel PRF95 (containing n-heptane and iso-octane) having same

Fig. 2.8 (a) Comparison of the heat release rates at three equivalence ratios for gasoline 95 in
HCCI engine [61]. (b) Summary of the reaction pathway showing the interaction of the
“obstructed pre-ignition” phenomenon in HCCI combustion [61]
48 2 Low Temperature Combustion Engines

Fig. 2.8 (continued)

octane number. This observation indicates that toluene has an important role in the
chemical kinetics of the gasoline autoignition.
Understanding of ignition characteristics and heat release process is important
for effective control and operation of HCCI engine. The ignition characteristics can
also be altered by changing the fuel composition. For example, the ITHR is induced
2.2 Homogeneous Charge Compression Ignition 49

by adding small amount of n-heptane in ethanol mixtures [62]. The ITHR can play a
key role in sustaining high load HCCI engine operation at retarded combustion
phasing [62, 63]. To achieve higher engine load in HCCI engine, a large amount of
fuel needs to be burned in cylinder, which results into very high pressure rise rate
and excessive noise. To avoid the very high combustion rate, combustion phasing
needs to be delayed after TDC. The cylinder volume expansion after TDC tries to
cool down the charge before HTHR. This charge cooling can be avoided by ITHR,
which can allow charge temperature to keep increasing shortly after the TDC.
Hence, HCCI combustion can be achieved at delayed combustion phasing for
fuels with ITHR. A study investigated the ITHR using ethanol and n-heptane
mixture in HCCI combustion, depicts that ITHR can be in the range of 2–5% of
total heat release. Figure 2.9 shows the variations of ITHR with the n-heptane
percentage in fuel–air mixture for different boost pressures in HCCI engine
[62]. The ITHR is found minimal at pure ethanol case, and moderate (2–3%) up
to 35% n-heptane in the mixture. When n-heptane percentage increases more than
35, ITHR rapidly increases to 5% of total heat release.
Reactions responsible for ITHR are investigated by chemical kinetic analysis.
Figure 2.10 exhibits the contribution of different reaction groups in heat release rate
with respect to crank shaft position for different compositions of n-heptane at intake
pressure of 2.2 bar and φ ¼ 0.4 [62]. The reactions are grouped according to the
number of carbon atoms in the oxidized species, namely, C1–C2, c1–c2, C3–C4 and
C7. Again, “C1–C2” denotes species with carbon derived from n-heptane, and “c1–
c2” is for species with carbon derived from ethanol. Another study also suggested that

Fig. 2.9 ITHR as a function of n-heptane percentage in fuel and intake pressures (φ ¼ 0.4 and
CA50 ¼ 10  aTDC) in HCCI combustion [62]
50 2 Low Temperature Combustion Engines

Fig. 2.10 Variation of heat release rates in HCCI combustion for (a) 100% ethanol, (b) 25%
n-heptane, (c) 45% n-heptane and (d) 55% n-heptane from various reaction groups [62]

the number of carbons in the hydrocarbon species undergoing partial oxidation is

correlated with primary heat-releasing reactions. Reactions consisting of small mol-
ecules (zero, one and two carbon atoms) produce heat at different crank angle
positions and magnitudes, than species with three and four carbon atoms, and they
also show different trends than larger hydrocarbons (five or more carbon atoms) [49].
In case of pure ethanol, very low ITHR observed (Fig. 2.10a) and ethanol
reactions c1–c2 are the largest single groups of heat-producing reactions in 25%
n-heptane case (Fig. 2.10b). At the 45% n-heptane, a more prominent ITHR
characteristic is noticed, which is correlated with an increase of both heptane
oxidation reactions and mostly ethanol oxidation reactions. A shift from ITHR to
LTHR characteristics is noticed at 55% the n-heptane fraction, and this character-
istic corresponds to an increase in importance of C1–C2 reactions (derived from
n-heptane). However, heat release in ethanol reaction groups also increases. It is
concluded that the blending of the highly reactive fuel increases the ITHR and at
very high blend percentage, LTHR can also be triggered. The ITHR can be
generated by adding small amount of reactive fuel, in the parent fuel, which does
not exhibit ITHR characteristics in HCCI combustion [62].
2.2 Homogeneous Charge Compression Ignition 51

2.2.3 HCCI Advantages and Challenges

The HCCI combustion process attracted the attention of researchers from the last
few decades due to their benefits of higher thermal efficiency along with simulta-
neous reduction of NOx and PM to a ultralow levels. The level of benefits gained
from HCCI combustion process depends on the fuel and strategy used to achieve the
auto-ignition in the engine cylinder. Figure 2.11 presents the main advantages and
challenges of HCCI combustion process in IC engines. One of the main strengths
and benefits of HCCI engine is the higher thermal efficiency. The main factors
affecting the thermal efficiency of engine are compression ratio, pumping loss,
specific heat ratio, heat transfer from combustion chamber, gas molecular dissoci-
ation, combustion phasing and mechanical friction [64]. The first three factors
(higher compression ratio, lower pumping loss, higher specific heat ratio) leading
to higher thermal efficiency are common to HCCI and conventional CI engines. The
conventional CI engines have higher thermal efficiency because of engine operation
at higher compression ratio, lower pumping loss (due to unthrottled operation) and
higher ratio of specific heat (due to lean engine operation). The HCCI engines are
operated at lower combustion temperature leading to lower heat transfer loss and
less dissociation of gas molecules. Additionally, premixed charge ignition leads to

Fig. 2.11 The HCCI engine advantages, major challenges and their proposed solutions
52 2 Low Temperature Combustion Engines

sootless (smokeless) combustion, which also reduces the radiation heat loss in
HCCI engines. Combustion phasing dictates the thermal efficiency by its effect
on combustion timing loss and heat transfer [64]. The HCCI combustion has
smaller combustion duration because of simultaneous ignition of entire charge in
the cylinder, which leads to near constant volume combustion cycle. Constant
volume combustion in HCCI engine also contributes to higher thermal efficiency.
Mechanical friction leads to reduction in brake thermal efficiency of engine.
Mechanical friction mainly depends on the peak cylinder pressure and engine
speeds. Generally, advantage of lower mechanical friction depends on compression
ratio of engine used for HCCI combustion process.
Another major strength of HCCI combustion engine is simultaneous reduction of
NOx and PM emission to ultralow levels due to lower combustion temperature
(lower NOx formation reactions) and premixed charge (lower soot formation due to
absence of local rich zone in charge) operation in LTC region (Fig. 2.1). The HCCI
combustion process has fuel flexibility to operate on both gasoline-like and diesel-
like fuels. To operate on particular fuel, suitable auto-ignition control strategy
needs to be used. In advanced compression ignition engines such as HCCI and
extended HCCI, required fuel reactivity is converging to in-between reactivity of
gasoline and diesel (see Fig. 1.13). Higher fuel economy is achieved in HCCI
engines due to higher thermal efficiency. The HCCI engines have higher part load
efficiency and limited operating range. This characteristic makes HCCI engine well
suited for hybrid electric vehicles (HEVs), where lower-power engine is required
and engine can avoid the transient operation. Therefore, HCCI can offer higher fuel
economy and is a potential candidate for HEVs [65].
Although HCCI combustion has advantages over conventional engines, it also
has several technical challenges (Fig. 2.11) that need to be addressed before
utilizing it for production engines. One of the main challenges in HCCI combustion
is to control the combustion phasing, which governs the combustion rate and peak
pressure in the cylinder. As discussed (Sect. 2.2.2), HCCI combustion timing is
governed by chemical kinetics, and there is no direct control over the combustion
phasing. Hence, simultaneous ignition of entire charge in the cylinder leads to very
high combustion rate especially at higher engine loads, where large amount of fuel
needs to be burned. The main factors affecting HCCI ignition timings and combus-
tion rate are mixture auto-ignition properties and mixture temperature history in the
cylinder. Mixture auto-ignition properties are affected by type of fuel, fuel concen-
tration, fuel additives and reforming, blending of two or more fuels, residual (and/or
EGR) rate and reactivity of residuals [55]. Similarly, mixture temperature history in
the cylinder is affected by intake conditions (temperature and pressure), EGR and/
or residual, thermal barriers, boosting, in-cylinder fuel injection timings, water
injection, compression ratio, latent heat of vaporization of the fuel, coolant tem-
perature, heat transfer from engine components, engine speed and load conditions
[55]. Hence, actuation methods that affect the stated parameters can be used to
control the combustion timings and combustion rate in HCCI engine. The proposed
actuation strategies for controlling the HCCI combustion phasing are VVT and
residual/exhaust gas trapping, VCR, variable EGR, modulation of intake
2.2 Homogeneous Charge Compression Ignition 53

temperature, water injection, in-cylinder injection timings, variable coolant tem-

perature and modulating two or more fuels [66]. Although there exists the sufficient
number of proposed strategies to control the combustion phasing in HCCI com-
bustion, every strategy has its own operating limits due to thier response time,
on-board control on engine, etc.
The other major challenge is to extend the limited HCCI operating range, while
maintaining the full advantages of HCCI combustion. To achieve higher engine
load operation, comparatively larger amount of fuel needs to be burned at appro-
priate combustion timings. Due to oxidation reactions in entire cylinder simulta-
neously, very high pressure rise rate is observed at higher engine load, which also
leads to higher pressure oscillations, combustion noise and higher NOx emissions.
High pressure rise rate limits the power output and defines high load boundary
(upper operating limit) of HCCI engine operation. Additionally, lower load oper-
ation is also limited, which defines lower operating limit. At low engine load
operation, insufficient temperature is achieved to activate auto-ignition of the
mixture, which results into emissions of higher CO and HC [40]. Parameters
affecting high load limit are systematically investigated. Several strategies are
proposed to extend the upper limit of HCCI engine operation. Fundamental param-
eters affecting high load HCCI operation are fuel properties and chemical kinetics
(LTHR, ITHR and HTHR), in-cylinder charge conditions (residuals, charge and
thermal stratification), combustion timing and heat transfer [27]. Solutions pro-
posed to extend HCCI operating range are boosting intake air (turbocharging/
supercharging), engine operation in dual mode (SI–HCCI or CI–HCCI), spark-
assisted HCCI, charge stratification and late combustion phasing engine operation.
Another major issue with HCCI engines is the higher CO and unburned HC
emissions especially at lower and moderate engine loads. Higher unburned HC and
CO emissions mainly occur because of lower combustion temperature due to lean–
burn or highly diluted charge combustion. The fuel conversion efficiency is also
hampered by higher HC and CO emissions at low engine load conditions. The
CO-to-CO2 conversion reactions are sensitive to temperature and oxygen availabil-
ity in the cylinder (Fig. 2.1), and high temperature is required to oxidize CO in to
CO2 (Sect. 2.2.2). Exhaust after-treatment by oxidation catalyst is one of the best
proposed methods for reduction of HC and CO from HCCI engines. However,
oxidation catalyst with low catalyst light-off temperature is required for HCCI
engines due to lower exhaust gas temperatures particularly at lower engine loads.
There exist methods to reduce catalyst light-off time, which can be used to achieve
the required operating temperature of catalytic converter [67–69]. Other strategies
proposed for reduction of HC emissions are wall insulation, catalyst coating and
reduced piston topland crevices [55].
In HCCI engine, the problem of weak cold start capability exists because auto-
ignition process is very sensitive to intake charge temperature. Initial charge
temperature needed for auto-ignition is dependent on fuel properties and engine
operation conditions. At cold start conditions, charge temperature is very low, and
heat transfer loss from compressed charge is also higher due to colder cylinder
walls [40]. Without any compensating mechanism for low lower charge
54 2 Low Temperature Combustion Engines

temperature, initiation of combustion is very difficult, and it can lead to misfire.

Several strategies are proposed to address the cold starting in HCCI engines such as
using glow plugs, increasing the compression ratio using VCR or VVT and using a
different fuel or fuel additive. The most practical approach to address the cold start
in HCCI engine is to start the engine in conventional SI or CI mode, and switch to
HCCI mode after warm-up [40, 66]. Effective premixed charge preparation and
avoiding fuel–wall interactions are essential in HCCI engine for higher fuel con-
version efficiency and lower emissions. This issue typically comes up for poor
volatility diesel-like fuels in HCCI engine because charge inhomogeneity affects
the auto-ignition process and governs the ignition timing in HCCI engine. To
prepare sufficiently premixed charge in HCCI engine, port injection of volatile
and gaseous fuel, and early in-cylinder injection with sophisticated fuel injector for
less volatile diesel-like fuels are proposed [66].

2.2.4 Parameters Influencing HCCI Combustion

2.2.4.1 Equivalence Ratio

The amount of fuel burned in the cylinder mainly controls the power obtained from
HCCI engine. In HCCI engine, throttle is not used to control the inducted air
quantity; therefore changing the quantity of fuel at different engine loads leads to
the change in equivalence ratio. To represent the quality of fuel–air mixture with
respect to stoichiometric air–fuel ratio, two metrics, namely, equivalence ratio (φ)
and relative air–fuel ratio (λ), are typically used. These two parameters are
inversely correlated (φ ¼ 1/λ) with each other. Figure 2.12 presents the effect of
λ on in-cylinder pressure and heat release rate (HRR) in HCCI combustion at
constant intake air temperature. Figure shows that maximum cylinder pressure
(Pmax) and maximum HRR (HRRmax) increase as mixture becomes richer (lower
λ values). Maximum pressure rise rate (PRRmax) is also very high for richer mixture
operation of HCCI combustion because higher quantity of premixed fuel is burned
simultaneously in the cylinder. Combustion duration also decreases with richer
fuel–air mixture, and lowest combustion duration (<10 CAD) observed for richest
mixture (λ ¼ 2.0). Combustion phasing becomes very late for leaner mixture, which
leads to lower HRR. Chances of misfire increases for very lean conditions in HCCI
combustion. The combustion phasing advances with richer fuel–air mixture, and
increasing λ from 2.0 to 2.6 (towards leaner mixture) leads to the decrease in
HRRmax by an order of magnitude (Fig. 2.12).
The main factors affected by a change in equivalence ratio (quantity of fuel per
cycle) are the temperature and composition of the residuals, wall temperatures and
related heat transfer rates, heating/cooling during induction, burn duration and
combination of chemical–kinetic and thermodynamic properties of the charge
mixture [71]. During HCCI combustion operation, increase in fuelling rate (or φ)
also leads to increase in the residual and cylinder wall temperatures. To understand
2.2 Homogeneous Charge Compression Ignition 55

Fig. 2.12 Variations of in-cylinder pressure and HRR at constant intake air temperature using
gasoline HCCI combustion (Adapted from [70])

Fig. 2.13 Variations in CA10 for iso-octane with fuelling for the continuously fired, fire 19/1 and
fire 18/2 schemes with a constant Tin ¼ 155  C, and for gasoline and PRF80 with fuelling fire 19/1
scheme (Adapted from [71])

the individual effect of these parameters, studies used (i) alternate firing operation
to remove changes in wall temperature and residuals [71–73] and (ii) a method for
determining the effective intake temperature to remove the effect of heating/
cooling during induction of charge [71]. Figure 2.13 presents variations in crank
56 2 Low Temperature Combustion Engines

angle position for 10% heat release (CA10) with alternate firing schemes for
different fuels at constant intake temperature (Tin) [71]. The key advantages of
conducting experiments at constant Tin are as follows: (i) it shows the response of
HCCI combustion to rapid change in equivalence ratio without Tin adjustment;
(ii) heat transfer during induction does not vary with φ for constant wall temper-
ature case. In alternative firing schemes (19/1 or 18/2), engine was operated on
19 or 18 cycles on constant equivalence ratio (φ ¼ 0.2), and then remaining 1 or
2 cycles (where measurements are conducted) at equivalence ratio of interest and
sequence repeats after 20th cycle. More details of alternate firing strategy can be
found in original study [71]. Variations in wall temperature (Twall) and residuals are
removed together by using the 19/1 scheme. In this scheme, 20th cycle has the same
wall temperature and residual for all the tested φ. In 18/2 firing scheme (two cycles
are fired at the desired φ and data is taken for 20th cycle), Twall can be kept
approximately constant, and the acquired cycle has residuals of 19th cycle which
is typically present in continuous firing at the desired φ for which measurement is
taken. Figure 2.13 shows that the start of combustion (CA10) for iso-octane
becomes more retarded almost linearly as φ is increased at constant Tin, when the
effect of variables Twall and residuals are removed using the 19/1 firing scheme. It
means the higher requirement of Tin with increase in φ for constant CA10 phas-
ing using iso-octane fuel because of lower combustion temperature due to the
reduction in ratio of specific heat (γ) as more fuel is added at higher φ. Gasoline
data (19/1 firing) is also fairly linear but the slope is less than iso-octane as γ is
similar. For PRF80 in 19/1 firing, effect of γ is similar to iso-octane, however strong
cool flame kinetics advances the CA10 phasing with φ, particularly φ>0.16
(Fig. 2.13). The 18/2 firing scheme data (for iso-octane) trend follows closely
19/1 scheme data than to the continuously fired data. It means that under normal
continuous fire conditions, the changes in Twall with φ have a larger effect on CA10
than the effects of residuals [71]. This can be justified for typical low residual
engines, as generally residual mass fraction is low (5.5% for the test case).
The CA10 phasing is a good measure of start of combustion and hence useful in
isolating the effect of fuel chemistry (auto-ignition chemistry and thermodynamic
properties) on ignition. However, the effect of changing equivalence ratio on CA50
(crank angle position corresponding to 50% heat release) phasing is important from
an engine performance perspective. Figure 2.14 shows the variations in CA50
phasing for continuously fired and constant Twall and residual (19/1 scheme)
using three different fuels in HCCI combustion. The 19/1 scheme data at constant
Tin presents the immediate response of combustion phasing to a rapid change in φ.
This observation is useful for interpretation of fast transient operation where Tin and
Twall not necessarily vary at that pace. The shift in CA50 phasing with φ can be taken
as a measure of the amount of compensation immediately required for constant
phasing operation of engine with a change in load/fuelling. For each fuel, combus-
tion phasing advances with increase in φ with different sensitivities for each fuel.
Figure 2.14 presents the magnitude of the effects of fuel type on CA50 (or the
required compensation) during a fuelling transient. It was found that iso-octane
needs essentially no immediate compensation to maintain the CA50 at TDC, and
2.2 Homogeneous Charge Compression Ignition 57

Fig. 2.14 Variations in CA50 phasing for continuous and 19/1 firing scheme for iso-octane,
gasoline and PRF80 (Adapted from [71])

only a relatively small amount of compensation after Twall equilibrates. Gasoline

also shows similar trend with little more variation in CA50 with φ than iso-octane as
both fuels do not show cool flame chemistry [71]. This observation indicates that
controlling combustion phasing during fuelling transients should be relatively easy
and straightforward for iso-octane or gasoline. The PRF80 data exhibits substantial
changes in the CA50 with φ (Fig. 2.14) due to the significant cool flame chemistry.
This means substantially more temperature compensation would be required to
maintain constant CA50 with this fuel, and controlling fuelling transients could be
more difficult with these kinds of fuels. Since iso-octane exhibits less than a 1.2
CA variation in CA50 from 0.1  φ 0.3 (Fig. 2.14), fuel stratification offers little
potential for adjusting the CA50 phasing. However, fuel stratification in the charge
has the potential to shift combustion phasing by several degrees in PRF80 because
it is more reactive and φ-sensitive fuel.
The φ-sensitivity of fuels also depends on the intake air pressure. Figure 2.15
presents the φ-sensitivity of three different fuels (gasoline, iso-octane and PRF73)
for naturally aspirated (NA) and boosted (high intake pressure) conditions. High
negative slope in the curves indicate higher fuel reactivity (and earlier ignition) and
more sensitive fuel. In naturally aspirated case, PRF73 shows significant advance-
ment in CA10 with increased φ. Therefore, PRF73 is a highly φ-sensitive fuel
because of the strong dependence of its robust pre-ignition reactions (reactions
responsible for the LTHR and ITHR) on the fuel concentration [72]. However,
iso-octane and gasoline show comparatively less sensitivity at naturally aspirated
case due to weak pre-ignition reactions of both fuels. In contrast to PRF73, gasoline
and iso-octane show delayed CA10 timing with increasing φ because of lower
compressed gas temperature achieved with higher fuel quantity due to decrease in
58 2 Low Temperature Combustion Engines

Fig. 2.15 φ-Sensitivity of different fuels for naturally aspirated engine operation and at a range of
boost pressures (Adapted from [72, 73])

γ (cp/cv) which dominates over pre-ignition heat release [63, 72]. Intake pressure
boosting significantly increases the auto-ignition reactivity of gasoline [63]. The
φ-sensitivity of gasoline increases (even stronger than PRF73) at Pin ¼ 2 bar
(Fig. 2.15) due to large increase in pre-ignition reactions in terms of increased
ITHR, while gasoline remains as a single-stage heat release fuel. High φ-sensitivity
of gasoline suggests that partial fuel stratification (PFS) should be effective for
reducing the HRR of boosted HCCI operation. The PFS can allow higher φ without
excessive PRR (knock) and/or more advanced combustion phasing for higher
efficiency. Additionally, these results also suggest that the ITHR pre-ignition
reactions are as important (sometimes more important as case of PRF73) for
determining φ-sensitivity as the LTHR reactions [72–74].

2.2.4.2 Fuel Properties

In general, several physiochemical and combustion properties are important for

automotive fuels (see Fig. 1.11). Volatility and auto-ignition reactivity are the two
essential fuel properties required for HCCI combustion. To prepare well-premixed
homogeneous charge, volatile fuel is required. Sufficient fuel reactivity is required
for obtaining auto-ignition at appropriate combustion phasing in the cylinder within
control parameters of the HCCI engine. In case of very reactive fuels, limited
operating range is obtained, and fuel reactivity also limits the engine compression
ratio. For advance LTC engines, fuel reactivity requirement is merging in between
diesel and gasoline (see Fig. 1.13). Gasoline-like fuels (high volatility fuels) are
more suitable for well-premixed HCCI combustion engines. Fuel volatility is
controlled by intermolecular forces depending on molecular size and structure.
2.2 Homogeneous Charge Compression Ignition 59

There is no significant electrostatic attraction in hydrocarbon molecules as they are

not positively or negatively charged molecules. The dispersion forces govern the
molecular attraction in hydrocarbons generated by the change in electron density in
the molecular periphery [27]. The strength of attraction forces depends on the
molecular surface area and intermolecular forces. Attraction forces increase with
higher surface area, which requires comparatively large energy to reach boiling
point. Therefore, volatility of fuel decreases with increasing number of carbon atom
in straight-chained hydrocarbons (such as diesel consists of straight-chained and
lightly branched 10–19 C atom hydrocarbons), and these kinds of fuels create
difficulty in premixed charge preparation in HCCI engines. The chain branching
increases the volatility by reducing molecular surface area leading to lower
intermolecular forces [27].
Auto-ignition reactivity of HCCI fuels depends on the intensity of oxidation
reactions taking place before the start of HTHR (hot ignition). Two main categories
of single- and two-stage heat release fuels in HCCI combustion are illustrated in
Fig. 2.4. Two-stage ignition fuels exhibit cool flame reaction (LTHR) followed by
ITHR, which leads to HTHR reactions (detailed discussion in Sect. 2.2.2). However
typical gasoline-like fuels exhibit single-stage heat release, where no LTHR is
observed, and HTHR is governed by heat release during ITHR regime in naturally
aspirated conditions. At higher intake pressure, fuel such as gasoline changes the
auto-ignition reactivity due to higher ITHR (Fig. 2.15). Typically, fuels having
larger amount of heat release during LTHR (if present) and ITHR before onset of
HTHR auto-ignite easily. This type of fuels requires less temperature and pressure of
intake charge at intake valve closing position to achieve auto-ignition in the cylin-
der. As discussed in Sect. 2.2.2, hydrogen abstraction and isomerization reactions
are the important reactions during LTHR for exhibiting two-stage heat release for a
particular fuel. The molecular structure of the fuel and type of “C–H” bond (primary,
secondary or tertiary bond) control the occurrence of the hydrogen abstraction and
isomerization reactions [27, 75, 76]. Primary “C–H” bond is the strongest bond
(among all C–H bonds), and for hydrogen abstraction requires highest activation
energy [75]. Therefore, fuel molecule having more number of secondary C–H bond
(long straight-chain molecules) is more prone to hydrogen abstraction during LTHR
reactions. During isomerization reaction also, secondary hydrogen removal is easy
(due to comparatively weaker bond) from long straight-chain molecule as isomer-
ization reaction takes place when the O2 chain reaches around to remove a hydrogen
atom from another site [27]. Additionally, due to long chain length and flexibility of
the fuel molecule, hydrogen abstraction by O2 is easy and more probable in long-
chain molecules [77]. Hence, long straight-chain alkanes (like n-heptane) are more
prone to LTHR in comparison to branched alkane (like iso-octane).
The temperatures characterizing the LTHR and ITHR zone depend on particular
fuel and engine operating conditions like boost pressure and engine speed
[21, 49]. Figure 2.16a shows the comparative in-cylinder pressure and mean
cylinder charge temperature for two different kinds of fuels (iso-octane, single-
stage heat release, and PRF80, two-stage heat release) [50]. The LTHR is observed
at 340 CAD in PRF80 fuel while no significant heat release in iso-octane until
60 2 Low Temperature Combustion Engines

Fig. 2.16 (a) Variation of in-cylinder pressure and temperature with crank angle for PRF80 and
iso-octane in HCCI engine [50]. (b) Standard deviation of IMEPg as a function of CA50 for
iso-octane, PRF80 and ethanol (Adapted from [78, 79])

HTHR at 367 CAD. The cylinder charge temperature is comparatively lower for
PRF80 before LTHR. The LTHR increases the temperature as well as pressure
before HTHR of PRF80 at about same 367 CAD (Fig. 2.16a). The cylinder charge
temperature is lower in PRF80 for same combustion phasing because of lower
intake temperature requirement due to LTHR and strong ITHR. In case of PRF80,
strong ITHR is observed because of increased intermediate temperature kinetics
due to heat released during LTHR. The lower charge temperature in case of PRF80
can take the advantage of having higher charge density by inducting more charge in
the combustion chamber, which leads to higher work output for same equivalence
ratio.
2.2 Homogeneous Charge Compression Ignition 61

The cycle-to-cycle variations in start of combustion (CA10) are reduced in

PRF80 due to higher rate of temperature rise before HTHR [50]. The cyclic
variations are associated with small random charge temperature variations, and
these temperature variations are dampened out because of heat release during
LTHR in the cylinder. Figure 2.16b shows the variation in indicated mean effective
pressure (IMEP) with CA50 for iso-octane, PRF80 and ethanol. The figure shows
that PRF80 can tolerate higher CA50 retard with lower cyclic variations (acceptable
standard deviation) in IMEP because of lower cyclic temperature fluctuations and
higher temperature rise rate before HTHR [4, 21, 50]. Increased combustion
phasing retards (which allows higher φ without knock), and higher inducted mass
leads to achieve higher engine load operation in HCCI combustion using two-stage
fuels like PRF80. In richer mixture condition (φ¼0.38 to 0.42, Fig. 2.16b), engine
operation with more retarded CA50 is possible for the same limit of IMEP
variations.
Fuel auto-ignition reactivity also affects the high load limit in HCCI combustion
engines. Figure 2.17 demonstrates the effect of auto-ignition reactivity on maxi-
mum achievable high load in naturally aspirated HCCI engine using iso-octane,
PRF80 and PRF60 fuels. Figure shows that maximum IMEP (6.5 bar) is obtained
for higher auto-ignition reactivity fuels (PRF60 and PRF80). Comparatively lower
IMEP is obtained using lower reactivity fuel (iso-octane). The different fuels have
different mechanisms of limiting high load boundary in HCCI operation. Iso-octane
has two limiting factors (excessive NOx and wandering CA50) for high load
boundary at compression ratio (CR) 14 and 3bar/CAD pressure rise rate (PRR)
(Fig. 2.17). With increase in fuelling rate, IMEP and combustion temperature
increases, which leads to excessive NOx emission beyond certain limit of IMEP.

Fig. 2.17 Illustration of high load limits for different fuels for naturally aspirated conditions in
HCCI combustion [80]
62 2 Low Temperature Combustion Engines

On further increase of fuelling rate, steady-state condition cannot be established and

constant combustion phasing cannot be maintained (i.e., wandering CA50).
Unsteadiness is caused by high sensitivity of the ignition timings to random
temperature fluctuations rather than the large changes in wall temperature with
CA50 [80]. At higher compression ratio (CR ¼ 18), wall heating runaway occurs
beyond 5.2 bar IMEP. Cylinder walls heat up at advanced combustion phasing than
average, due to higher heat transfer to the wall. The higher wall temperature (Twall)
leads to even more advanced combustion causing even higher heat transfer and so
on. Small random variations in the charge temperature (at retarded CA50 and low
ITHR) cause relatively large variation in CA50 which leads to higher Twall varia-
tions in comparison to normal non-limiting cycles [21]. This higher thermal sensi-
tivity leads to runaway at this condition. At lower PRR (3 bar/CAD) and CR (14),
variations in Twall are not adequate to initiate runaway to knock but cause wander-
ing of CA50. PRFs are less sensitive to variations in temperature due to LTHR
followed by increased ITHR. PRF60 is a higher reactivity fuel and requires large
amount of EGR to maintain acceptable PRR limit by retarding CA50. In this case,
high load boundary is limited by availability of oxygen, which decreases with
increase in EGR fraction. At higher loads, faster runaway to knock occurs in
PRF80 in comparison to iso-octane. Due to two-stage heat release in PRF80,
CA50 is less sensitive to the increase in inlet charge temperature. Therefore, in
this case, wall heating is not mainly responsible for fast runaway to knock. Study
proposed NOx-induced runaway caused by chemically enhanced auto-ignition due
to NOx present in the residuals (EGR) [80]. For detailed explanation and justifica-
tion of data, refer to original study [80].

2.2.4.3 Intake Temperature and Pressure

The intake charge conditions (equivalence ratio, temperature and pressure) and fuel
composition in charge have significant effect on combustion phasing, heat release
and other performance parameters in HCCI engine. The intake temperature mod-
ulation is one of the most common strategies to control the HCCI combustion
phasing. In the beginning of the compression stroke, temperature of fuel–air
mixture increases to speed up the reaction kinetics and achieve the auto-ignition
at desired crank angle position. This higher temperature can be achieved by intake
air preheating or retaining part of hot residuals inside the cylinder. Fast thermal
management (FTM) is an another technique for fast control of intake temperature.
In FTM, intake air temperature is controlled by mixing of cold and hot air coming
from different sources [27, 81]. Figure 2.18 shows the variations of cylinder
pressure and HRR at constant relative air–fuel ratio (λ) for different intake air
temperatures [70]. Increasing intake air temperature increases the cylinder pressure
and HRR very rapidly (Fig. 2.18). Pressure rise rate also increases drastically with
increase in intake temperature. Figure also depicts that start of combustion
advances with the increase in intake air temperature. Higher intake air temperature
increases the charge temperature in the cylinder, which increases the reaction rate
2.2 Homogeneous Charge Compression Ignition 63

Fig. 2.18 Variation of in-cylinder pressure and HRR at 1200 rpm and constant λ for gasoline
HCCI combustion (Adapted from [70])

kinetics resulting into advance combustion and faster HRR. Thus, intake air
preheating significantly affects the combustion phasing in HCCI combustion [70].
The requirement of particular intake air temperature depends on the engine
operating parameters (λ, speed, coolant temperature) and fuel and engine design
parameters such as compression ratio. By selecting appropriate parameters, optimal
combustion phasing can be obtained. At particular compression ratio and engine
speed, thermal conditions of charge in the cylinder are determined by intake charge
temperature and wall temperature. The intake charge temperature has direct impact
on the thermal state of the bulk gas in the core of cylinder, and the coolant (wall)
temperature creates more substantial effect on the charge in the thermal boundary
layer. Thus, the wall temperature affects the thermal stratification in the combustion
chamber near-wall regions [82]. The study also found that variations in combustion
phasing and HRR created by change in wall temperature can be compensated by
varying intake temperature in opposite direction with factor of 1.11 [82]. The
decrease of coolant temperature has strong effect on combustion instability in
comparison to the decrease in intake charge temperature, and hence combustion
stability limits are more dependent on wall temperature in HCCI combustion
engine. The requirement of intake air temperature is affected by compression
ratio of engine and octane number of fuel. Compression ratio has strong influence
on intake air temperature. Study shows a reduction in compression ratio from 21.4
to 17 results into increased requirement of intake air temperature from 30  C to
130  C for iso-octane fuel [39]. Increase in octane number from 0 to 100 requires
increment compression ratio by a factor of two in HCCI engine [39]. An increased
compression ratio of engine increases HRR and advances the combustion phasing
[83]. Increase in boost pressure resulted into strong decrease in inlet temperature
requirement [84].
64 2 Low Temperature Combustion Engines

Intake pressure boost is typically used to increase the power output of engine,
and it significantly extends the high load limit of HCCI engines. At higher intake
pressures, same equivalence ratio and dilution level are achieved by inducting
higher amount of fuel in every cycle, which leads to higher engine loads in the
engine. The application of high intake pressure (boost) is limited by knocking (very
high PRR). The knocking propensity of HCCI engine increases with increase in
intake pressure mainly by two reasons: (i) higher amount of fuel inducted generates
high PRR in combustion for the same combustion duration and phasing, and
(ii) higher intake pressure increases the auto-ignition reactivity by an amount
depending on the fuel [21]. The φ-sensitivity of fuels can also increase at higher
intake air pressure (Fig. 2.15), which can be utilized for increasing the higher load
range using fuel stratification. Higher intake pressure in combination with external
exhaust gas recirculation can be used to attain higher engine load, and a study
achieved IMEPg of ~16.3 bar at 3.25 bar intake pressure with gasoline fuel in HCCI
combustion [63].

2.2.4.4 Engine Speed

The engine speed and fuel type affect the start of bulk gas reactions in HCCI
combustion. There exists a strong coupling between fuel auto-ignition quality and
required intake charge temperature to have an optimal combustion phasing. The
intake charge temperature directly affects the completeness of combustion in the
cycle because CO-to-CO2 conversion reactions are highly sensitive to the maxi-
mum combustion temperatures [85]. The onset of bulk quenching of CO-to-CO2
conversion reactions can be used as fuelling rate limits at particular operating
condition, and below this fuelling rate, combustion is incomplete. Figure 2.19
shows the variation of required intake charge temperature with engine speed for
different fuelling rates for PRF80 (two-stage heat release) and gasoline (single-
stage heat release) fuel. Typically, higher intake charge is required at higher engine
speed for maintaining the combustion phasing because at higher engine speed, there
is less time available for chemical reactions to set off the auto-ignition process
[70, 85, 86]. Additionally, the small inlet valve diameter, the short cam duration and
the pressure drop over the intake air heater (if present) lead to the decrease in
volumetric efficiency (ηv) of engine as engine speed increases. Lower volumetric
efficiency results into lower pressure during intake stroke. The higher charge
temperature is required to compensate for pressure drop (due to lower ηv) and to
accelerate the chemical kinetics of reactions due to less availability of time at
higher engine speed [70, 86]. Gasoline requires 20 C higher intake temperature as
engine speed increases 600–1200 rpm. However, between engine speeds
1200–2400 rpm, intake charge requirement is almost constant (Fig. 2.19). Several
possible reasons for this inconsistency are as follows [85]: (i) lower heat transfer
from gas to cylinder due to less time available at higher engine speed provides
higher compression temperature at constant Tin; (ii) cylinder surfaces are signifi-
cantly hotter at higher engine speed which contributes to higher charge
2.2 Homogeneous Charge Compression Ignition 65

Fig. 2.19 Variations of required intake charge temperature with engine speed for PRF80 and
gasoline in HCCI combustion (Adapted from [85])

temperatures; (iii) the “ram” effect is significant above 1800 rpm engine speed,
which effectively compresses and heats the intake charge towards the end of the
intake stroke; and (iv) charge heating due to turbulent dissipation of kinetic energy
during the charge induction increases with engine speed.
In case of PRF80, intake charge requirement is highly dependent on engine
speed and equivalence ratio (Fig. 2.19). The intake charge temperature for PRF80
has a much stronger dependence on φ at 1200 rpm and lower engine speeds. The
required intake charge temperature drops quickly with increased fuelling (φ). There
is a vast difference in necessary intake temperature for maintaining the combustion
phasing at high equivalence ratios. As the φ decreased, the two curves appear to be
merging [85]. This trend can be understood in terms of LTHR (cool flame activity)
variations of PRF80 as it is a two-stage heat release fuel. The cool flame activity
increases with lower engine speed, higher equivalence ratio and higher n-heptane
content [85]. The heat release during LTHR duration leads to the reduction in the
intake charge temperature requirement.
Fuels with two-stage heat release have some significant advantage to overcome
the major challenge of extending high load boundary and improving combustion
phasing control. Fast combustion phasing control can be accomplished with LTHR
fuels because of the φ-dependence of the LTHR. The combustion phasing can be
controlled by varying the amount of fuel stratification for a fixed φ because the
amount of LTHR depends on the actual mixture undergoing combustion [53, 71,
87]. The φ-dependence of the LTHR in two-stage fuels also offers a means for
lowering the maximum HRR by partial stratification using dual injection [78]. The
cycle-to-cycle variations are also lower at extensively retarded combustion phasing
operation of engine using LTHR fuels [50]. The LTHR also have higher power
66 2 Low Temperature Combustion Engines

Fig. 2.20 Variations of HRR for PRF80 at φ ¼ 0.38 for different engine speeds at constant
combustion phasing in HCCI combustion (Adapted from [53])

density (Fig. 2.17). The LTHR significantly reduces the required Tin, which
increases the charge density for a fixed intake pressure. The amount of LTHR is
highly dependent on engine speed. Figure 2.20 shows the variations of HRR with
engine speed for PRF80 at constant combustion phasing. The amount of LTHR
decreases drastically with increasing engine speed (Fig. 2.20). There are two main
reasons for this observation. First, the LTHR is fairly slow, oxidation reactions
occur actively in 760–880 K temperature range, and there is less time for the LTHR
to take place before 880 K as the engine speed is increased [53]. The faster
progression of charge temperature from lower to higher temperatures at higher
engine speed is due to the combination of LTHR and compression heating. Second,
engine is operated at higher charge temperature and at higher engine speeds to
maintain the combustion phasing constant, which leads to reduction in cylinder
pressure, while the charge passes through the 760–880 K range. The low temper-
ature heat release is sensitive to pressure (Sect. 2.2.2), which further decreases the
amount of LTHR. The combination of these two effects results into the rapid
reduction of the amount of LTHR with increasing engine speed. Figure 2.20 also
shows that the LTHR disappears completely in engine speed range of 1400–1500-
rpm, where the Tin requirement increases most rapidly [53].
Fuel composition has strong effect on the engine speed beyond which LTHR
disappears. Figure 2.21 shows the variations of required intake temperature to
maintain the constant combustion phasing for different fuels at constant equiva-
lence ratio in HCCI combustion. It can be noticed that the change of n-heptane/iso-
octane blending ratio influences the autoignition reactivity of the fuel, which affects
the Tin required to maintain combustion phasing. Figure shows that the LTHR is
2.2 Homogeneous Charge Compression Ignition 67

Fig. 2.21 Variations of intake temperature requirement with engine speed for different fuels at
Φ ¼ 0.38 (Adapted from [53])

present in the ranges of 660–870 rpm, 1050–1440 rpm and 1640–2000 rpm for
PRF90, PRF80 and PRF70, respectively. Figure 2.21 also illustrates that the toluene
addition reduces the engine speed sensitivity of the PRF blends. This happens
because the LTHR-suppression effect of toluene is roughly proportional to the
amount of LTHR that the n-heptane would have produced and had toluene not
been present [53].
Figure 2.21 also demonstrates the several issues related to LTHR fuels: (i) the
very reactive fuel (PRF70) generates too much LTHR at lower engine speed, and
the required Tin drops below the ambient temperature, which is impractical; (ii) the
disappearance of LTHR at higher engine speeds renders the partial fuel stratifica-
tion ineffective, which is used to lower the PRR to achieve higher engine loads, and
it also limits the potential for combustion phasing control; and (iii) the engine speed
range where LTHR is present imposes high demands on the thermal management as
Tin varies with engine speed [53].

2.2.4.5 Exhaust Gas Recirculation

Exhaust gas recirculation (EGR) is a practical means to control the charge temper-
ature in HCCI engine by adding high levels of exhaust gases into the intake
manifold of the engine. The EGR also provides the dilution of charge to control
the excessive HRR of HCCI combustion. Typical composition of the EGR is CO2,
H2O, N2, O2, CO, unburned HC, particulate matter, NOx and some combustion
reaction intermediates. Two different EGR strategies are used in HCCI engine,
namely, internal EGR and external EGR. In the internal EGR strategy, hot
68 2 Low Temperature Combustion Engines

combustion products are captured in the cylinder by early closing the exhaust valve
(EVC) towards the end of the exhaust stroke. The EVC timing decides the amount
of retained EGR in the cylinder. The early EVC is realized using VVT system of
engine. Earlier EVC timing leads to more hot residuals in the cylinder that raises the
charge temperature at BDC (bottom dead center) of intake stroke. The higher
residual mass fraction in the charge has higher heat capacity, which tries to delay
the ignition timing. The higher residual mass captured in the cylinder also has
higher temperature that tries to advance the ignition timings. Generally, the higher
temperature effect of retained residuals in the cylinder dominates in the auto-
ignition process [27]. A study investigated the charge heating, dilution (oxygen
reduction), heat capacity, chemical (only CO2 and H2O) and stratification effects of
internal EGR in CAI engines [87]. It is concluded that charge heating is mainly
responsible for advanced ignition timings, dilution and heat capacity have no
significant effects on ignition timing but decrease the HRR and increase the
combustion duration of CAI engine. Stratified residual gases facilitate the CAI
combustion. In the external EGR strategy, desired amount of exhaust gas is
recirculated back in the intake manifold from the exhaust manifold through appro-
priate cooling system. The amount of EGR can be varied using exhaust valve in the
exhaust line and changing the exhaust back pressure. The effects of combined
internal and external EGR on gasoline CAI have been used to increase the attain-
able load in a multi-cylinder engine [88]. Study found that highest output achieved
using internal EGR is limited by excessive PRR, and additional external EGR
retarded the ignition, which lowers the HRR and PRR.
The use of substantial EGR/residuals is central to the most gasoline HCCI
applications for maintaining the combustion phasing at desired position. The
EGR is also desirable for diluting the charge for operating the engine on stoichio-
metric conditions, so that three-way catalytic converter can be used to mitigate the
undesired NOx at higher engine load conditions [89]. The diesel-like fuels easily
auto-ignite by the rise in charge temperature during compression stroke. In diesel
HCCI, substantial quantities of EGR are required for reducing the auto-ignition
propensity of the fuel. Typically, substantial EGR addition is used to increase the
ignition delay for getting sufficient time for fuel/charge mixing in the cylinder. The
use of EGR also sufficiently reduces the oxygen concentration in the cylinder.
Therefore, the peak combustion temperatures in the fuel-rich regions (due to
inhomogeneity) do not reach to a level required for significant NOx formation
[90]. Hence, the use of EGR is also central to diesel HCCI combustion. The EGR
can also be used to manage LTHR at higher engine speeds, where LTHR vanishes
(Fig. 2.21). Therefore, a potential exists to use the benefits of LTHR for combustion
phasing and HRR control over a wide range of speeds and boost
conditions using EGR.
The addition of EGR typically retards the ignition timings for all the fuels. The
amount of retard in ignition timings by EGR depends on the type of fuel and
quantity of EGR used. The different mechanisms exist to produce thermodynamic
and chemical effect of EGR during combustion. A study identified the various
mechanisms of EGR as [89] (i) thermodynamic cooling effect due to increased
2.2 Homogeneous Charge Compression Ignition 69

Fig. 2.22 Effect of EGR

and its constituents on
ignition timing for (a)
gasoline and (b) PRF80 in
HCCI combustion (Adapted
from [89])

specific heat capacity, (ii) O2 concentration reduction effect, (iii) enhancement of

auto-ignition due to the presence of H2O and (iv) enhancement or suppression of
auto-ignition due to the presence of trace species (CO and NO). Figure 2.22 shows
the effect of EGR and its constituents on ignition timing retard for gasoline and
PRF80 fuels. To separate the effect of trace species present in real EGR due to
incomplete combustion, the complete stoichiometric products (CSP) containing
CO2, H2O and N2 are used for the study. Addition of CSP) in intake has strong
effect for both gasoline and PRF80 (Fig. 2.22). A detailed study conducted using
gasoline, iso-octane, PRF80 and PRF60 shows that CSP has strongest effect on pure
iso-octane and lowest sensitivity to gasoline, while PRF80 and PRF60 have
in-between sensitivity [89]. Retarding effect of real EGR is higher for PRF80
than gasoline and iso-octane. In practical EGR system, water content in EGR
may condense on cold surfaces (mainly at the time of cold starting and/or cold
weather), which will change the composition of real EGR. This effect is illustrated
by the trend between dry CSP and CSP. Loss of water makes less effective in
iso-octane while more effective in gasoline and PRF80 (Fig. 2.22).
70 2 Low Temperature Combustion Engines

The strongest cooling effect is found for CO2 for both gasoline and
PRF80 (Fig. 2.22). The cooling effect of H2O is in between CO2 and CSP. This
observation is also experimentally confirmed by measuring motored compression
temperature data at 350 CAD by varying the EGR and its constituents [89]. In the
same experiment, it is found that N2 has opposite effect and it slightly increases the
compression temperature. However, using N2 as diluent retarded the combustion
phasing due to oxygen sensitivity of auto-ignition reactions not by thermal effect.
Therefore, N2 is used to separate the O2 reduction effect of EGR in engine
combustion. In all other cases, combustion phasing retard is a combination of
thermodynamic cooling and O2 reduction effect. Detailed description with exper-
imental observation is provided in original study [89]. The O2 reduction effect is
relatively weak compared to thermodynamic cooling (seen by comparing the slope
of curves) in gasoline but has more pronounced effects in PRF80 (Fig. 2.22). The
higher O2 sensitivity (higher slope of N2 curve in Fig. 2.22b) is attributed to the
LTHR in PRF80. Trend in CSP and dry CSP indicates that the dry CSP has slightly
stronger retarding effect in gasoline and PRF80 (Fig. 2.22). This observation
suggests that H2O has a chemical enhancing effect on hot ignition more than
thermodynamic cooling particularly at lower concentration. However, in case of
gasoline, adding water increases both chemical and thermodynamic effects, and
CA10 becomes significantly more retarded.
The difference between CSP and real EGR is only by the presence of trace
species produced from incomplete combustion. The concentrations of these trace
species in the intake manifold increase rapidly with increasing EGR rate. Real EGR
has less retarding effect on ignition timing for gasoline and comparatively more
retarding effect on PRF80 (Fig. 2.22). The gasoline has an enhancement of the auto-
ignition reactions due to trace species from real EGR. Gasoline is a single-stage less
reactive fuel, and addition of whole range of different hydrocarbons might lead to
new pathways of oxidation reactions during auto-ignition process. PRF80 is a
reactive species, and most of the molecules produced by incomplete combustion
might be comparatively less reactive [89]. This explains the behaviour of gasoline
and PRF80 with real EGR. The CO is a major specie present in the EGR. Study
conducted on pure n-heptane and two surrogate fuels (80% n-heptane–20% toluene,
75% n-heptane–25% iso-octane) showed that CO addition in the intake manifold
has no significant effect on auto-ignition process for up to 2000 ppm concentrations
[91]. Other studies also experimentally as well as numerically confirmed that CO
addition up to 170 ppm does not have significant effect on auto-ignition process
[47, 92]. However, a numerical study showed that the CO additions of 1000 ppm or
more leads to delayed the auto-ignition timings. The same study also showed that
adding more than 10,000 ppm CO in the manifold, the chemical energy supplied by
CO starts to increase the peak pressure, which is less likely a situation in HCCI
engine [92].
Formaldehyde is also emitted in significant quantity in HCCI engines
[93, 94]. Adding formaldehyde in intake manifold delays the ignition timings
because formaldehyde consumes the hydroxyl radicals and decreases overall reac-
tivity [92]. Although HCCI engine is known for ultralow NOx emission but at
2.2 Homogeneous Charge Compression Ignition 71

certain engine operating conditions, NO emission can be in significant amount. The

effect of NO in the intake during EGR is systematically investigated [95, 96]. Study
conducted using PRF, toluene reference fuel (TRF) and gasoline showed that sharp
advance of LTHR as well as the main heat release at low concentrations (~15 ppm)
of NO is added to the intake air. The rate of this change decreases with increasing
NO concentration. At higher concentrations of NO, heat release is retarded with
higher NO concentration) mainly for PRFs [95]. Study also concluded that com-
bustion phasing in the HCCI engine is more sensitive to variations of charge NO
concentration compared to the SI engine [95]. Another study also concluded that
NO addition in intake up to 45 ppm advances the auto-ignition process, and beyond
this value, the auto-ignition is retarded for PRF [96]. This observation indicates that
the NO can be used to control HCCI combustion phasing by varying the amount of
EGR (containing adequate NO concentration). Table 2.1 presents the summary of
effect of EGR constituents on ignition timings for different autoignition character-
istics of fuels as discussed in the current section.
In naturally aspirated operating conditions without EGR, the amount of LTHR is
dependent on fuel type. Reactive fuels show LTHR in a very narrow range of
engine speed and intake temperature (Tin) (Fig. 2.21). The EGR can be used very
effectively to operate with a constant amount of LTHR at fixed Tin, across wide
ranges of engine speeds because EGR counteracts autoignition, especially the
amount of LTHR [53]. The EGR can also be used to extend the high load boundary
in combination with intake boost. Figure 2.23 shows the maximum achieved IMEP
(with ringing intensity limit < 5 MW/m2) with and without EGR at CR ¼ 14 and
1200 rpm. Figure shows that the engine load is significantly increased with the use
of cooled EGR and high boost pressure. The maximum load line with EGR
(Fig. 2.23) shows that the IMEP increases almost linearly with boost up to
Pin ¼ 200 kPa. Above 200 kPa boost pressure, the slope of the maximum load
line becomes less steep. Main mechanism suggested by authors for this reduction in
the slope is as follows [63]. Initially, 200 kPa Pin  240 kPa, reduction in slope
occurs because CA50 could not be retarded beyond the amount used for
Pin ¼ 200 kPa with acceptable combustion stability. Hence, allowable fuelling

Table 2.1 Effect of EGR constituents on auto-ignition for single- and two-stage heat release fuels [97]
Single-stage fuels Two-stage fuels
(gasoline and iso-octane) (PRF60 and PRF80)
Effect on auto- Effect on auto-
Effect items ignition Sensitivity ignition Sensitivity
Thermodynamic Retarding High Retarding Low
cooling
[O2] reduction Retarding Low Retarding High
Presence of H2O Enhancement Low Enhancement High
Presence of trace Enhancement – Suppression –
species
72 2 Low Temperature Combustion Engines

Fig. 2.23 Variation of maximum IMEP obtained with EGR and without EGR using gasoline in
HCCI combustion at 1200 rpm (Adapted from [63])

rate cannot be increased to achieve higher IMEP. For higher intake pressures Pin 
260 kPa, the maximum load is no longer limited by the ability to retard CA50
because fuelling is limited by the oxygen content of the charge due to higher
amount of EGR used to retard the combustion phasing. Therefore, maximum
achievable load (IMEP) has different limiting mechanisms at different operating
conditions.
The maximum achievable steady IMEP depends on the fuel type, selected PRR
(or knock metric), EGR quality and EGR/fresh gas unmixedness [98]. As the
fuelling is increased beyond the maximum achievable steady condition, it is
increasingly difficult to maintain sufficiently retarded combustion) to prevent
runaway knock without having poor combustion stability that can also drift into
misfire. The five load limiting factors/mechanisms which are identified leading to
runaway knock are [98] (i) residual NOx-induced runaway advancement of the
combustion phasing, (ii) EGR-NOx/wall heating-induced runaway, (iii) EGR-NOx-
induced runaway, (iv) EGR-induced oxygen deprivation and (v) excessive partial-
burn occurrence due to EGR unmixedness.
Figure 2.24a summarized the main runaway mechanism for PRF80 and
PRF60 at different PRR in combination of the type of EGR (well mixed or
hetrogeneous, simulated or real EGR). In PRF80 fuel, main limiting mechanisms
are residual NOx-induced runaway and EGR-NOx-induced runaway. Using simu-
lated EGR (complete stoichiometric products), very fast knock runaway is found in
comparison to wall heating-induced runaway. Authors suggested that the runaway
is so rapid because of auto-ignition enhancement) due to residual NOx [98]. When
2.2 Homogeneous Charge Compression Ignition 73

Fig. 2.24 (a) Main knock runaway mechanism for PRF80 and PRF60 at different PRR in HCCI
combustion (Adapted from [98]). (b) Illustration of stable and unstable CA50 with well-mixed real
EGR and heterogeneous EGR at PRR 3 bar/CAD (Adapted from [98])

some part of NOx generated in one cycle is retained within the cylinder as a part of
the residual gases, the autoignition enhancement comes into effect immediately
during the following cycle. With real EGR in PRF80, runaway is found 21 times
slower than for the residual NOx-induced runaway observed for the PRF80, and
4 times faster than the wall heating-induced runaway observed for the iso-octane. It
is concluded that largest contributor to this moderately fast runaway with real EGR
is the enhancement of the autoignition by NOx recirculated to the intake with the
EGR gases. Wall heating-induced runaways also have small contribution to this
74 2 Low Temperature Combustion Engines

process. The load limiting mechanisms are different for the two fuels (PRF80 and
PRF60). With simulated EGR (CSP) in PRF60, IMEP is limited by the available
oxygen because PRF60 is more reactive than PRF80, therefore requiring the use of
higher amount of EGR. In PRF60, comparatively slow runaway is observed using
real EGR. In PRF60, lower NOx formation due to the high EGR rate leads to a
substantial reduction of the contribution from EGR-NOx enhancement of the
autoignition, and wall heating mechanism also almost equally contributes to run-
away process with real EGR. A more complete discussion with suitable justification
can be found in original study for all the mentioned mechanisms [98].
Effect of well-mixed and heterogeneous EGR needs to be investigated
(Fig. 2.24b). The stringent requirements for uniform cylinder-to-cylinder distribu-
tion of EGR are challenging. Additionally, due to the pulsating nature of the flow in
the exhaust, EGR and intake systems, there is an inherent risk of cycle-to-cycle
variations of the amount of EGR gases inducted into each of the cylinders. A study
investigated the well-mixed and heterogeneous EGR using two different routes of
EGR supplied [98]. The effects of a slight EGR unmixedness are clearly noticeable
at more retarded CA50 position used to achieve a lower PRR of 3 bar/ CA
(Fig. 2.24a). In this type of engine operations, partial-burn cycles are likely to
appear. The data in Fig. 2.24a used strict limit of maximum one partial-burn cycle
which is allowed in 1000 cycles [98]. Partial-burn cycle is defined as having an
IMEP lower than 95% of the average IMEP. If engine is operated with heteroge-
neous EGR above this load limit, an unacceptable frequency of partial-burn cycles
appears (Fig. 2.24b). Moving average of CA50 is presented in Fig. 2.24b for
heterogeneous and well-mixed EGR operation of HCCI engine. It is found that
with well-mixed EGR), the HCCI combustion phasing is comparatively well stable.

2.2.4.6 Inhomogeneities

Although HCCI combustion is termed as “homogeneous,” however in real HCCI

engines, there are always inhomogeneities mainly due to fuel stratification and/or
thermal stratifications. Investigations are conducted to prove the presence of inho-
mogeneity by comparing the simulated homogeneous combustion (single zone) and
real experimental data at same operating conditions. Figure 2.25 demonstrates the
effect of inhomogeneities in real engine by comparing PRR at different equivalence
ratios in fully homogeneous and real engines. Single-zone homogeneous charge
model is used for simulation of pure “homogeneous” case because it is not possible
to create a charge in a real engine, which is fully homogeneous in both mixture and
temperature. Figure 2.25a shows that PRR increases very fast with increase of
equivalence ratio in case of homogeneous model than experimental HCCI engine.
For the same knocking limit, it is possible to operate real engines at higher
equivalence ratio at both combustion phasing (Fig. 2.25a). Higher equivalence
ratio operation leads to higher engine loads. It is believed that the lower PRR in
real engine is because of naturally occurring charge stratification due to either
mixture or thermal inhomogeneities [23]. Inhomogeneities in the charge can appear
2.2 Homogeneous Charge Compression Ignition 75

Fig. 2.25 (a) Variation of

PRR with φ at different
CA50 for simulated fully
homogeneous and real
HCCI engine (Adapted
from [23]). (b) Combustion
duration (CA90-10) vs. CA10
for homogeneous and real
experimental data at
φ ¼ 0.367 (Adapted from
[99])

due to various reasons such as incomplete mixing of fresh charge and residuals,
incomplete fuel/air mixing or heat transfer, non-isothermal intake conditions and
turbulent mixing during the compression stroke [23]. Thermal and/or mixture
stratification in HCCI engine leads to spatially and temporally distributed combus-
tion in the cylinder rather than simultaneous combustion in purely homogeneous
charge. Therefore, the natural stratification in the cylinder plays an important role in
controlling the high load limit of an HCCI engine. This benefit of natural stratifi-
cation is increased at retarded combustion phasing because with same knock limit,
it is possible to operate the engine at even higher φ (Fig. 2.25a). Due to this reason,
it is concluded that even without enhancement of naturally occurring thermal
stratification, combustion phasing retard has potential to operate the engine at
higher fuelling rate and higher engine speeds [100]. At retarded combustion
phasing, charge temperature is lower, and combustion duration increases rapidly
with delayed ignition timings (CA10). The change in combustion duration is mainly
affected by rapidly changing thermal stratification of the charge around TDC and
less affected by slowing chemical kinetics at particular ignition timing
76 2 Low Temperature Combustion Engines

Fig. 2.26 (a) Illustration of inhomogeneities in HCCI combustion using chemiluminescence

image sequences (Adapted from [23]). (b) Illustration of evolution of thermal stratification in
HCCI combustion using T map images (Adapted from [101])

[99]. Figure 2.25b shows that combustion duration for experimental engine is
changing rapidly with ignition timings (CA10) than single-zone homogeneous
model. The single-zone model represents the effect of combustion phasing on the
chemical kinetics alone. A more complete discussion on the effect of combustion
retard and thermal stratification on PRR is present in the reference [99].
Various imaging studies have indicated the significant inhomogeneities in the
HCCI combustion [23, 24, 101, 102]. Figure 2.26 illustrates the inhomogeneities in
the HCCI combustion, which uses premixed charge. Typically, heat transfer
through cylinder wall, which affects the thermal boundary layers, is considered
responsible for thermal stratification. Chemiluminescence imaging and PLIF imag-
ing studies showed that thermal stratification extends in entire bulk charge in the
cylinder [23, 101, 102]. Figure 2.26a shows the chemiluminescence images of
HCCI combustion using iso-octane at φ ¼ 0.24 for CA50 at TDC (360 CAD).
Strong turbulent structures are noticed from figures, which indicate that the com-
bustion is not homogeneous. Initially, the chemiluminescence images suggest that
combustion starts at multiple locations throughout the cylinder with no specific
pattern. The images clearly show that several regions have ignited (bright spots)
while other regions have not started to ignite (Fig. 2.26a). As combustion pro-
gresses (during 358 –360 CAD), more regions start to react resulting into much
brighter chemiluminescence images. At 360 CAD position, most regions are
producing strong chemiluminescence, indicating intense reactions (Fig. 2.26a).
2.2 Homogeneous Charge Compression Ignition 77

After this position, intensity starts decreasing due to burn-out of first ignited
regions, while lately ignited regions are burning. Images shown in Fig. 2.26a are
of not the same engine combustion cycle (aquired at same setting in different
cycles). In the same study, authors also acquired the sequential images of one
particular combustion event, and images show that autoignition begins with the
hottest zone and progresses towards comparatively cooler zones in a sequential
manner. It is also speculated that turbulence is important in mixing the cooler gas
near the cylinder wall with the hotter gas in the bulk of the cylinder [23].
Study conducted to investigate the primary cause of inhomogeneities concluded
that the inhomogeneities are caused primarily by thermal stratification due to heat
transfer during compression, combined with turbulent transport [23]. The incom-
plete mixing of premixed charge and incomplete mixing during direct fuel injection
have insignificant role because images for three different kinds of injection strate-
gies are giving essentially the same appearance and the same development through
the engine cycle [23]. Another study investigated the role of the thermal stratifica-
tion produced during the intake stroke on the temperature distribution at TDC of
compression stroke [103]. It is concluded that the variations in the initial temper-
ature distributions have almost no effect on the temperature distribution and the
cylinder wall heat transfer development during compression stroke. The convective
transport of cold boundary-layer gases towards the cylinder centre is found as the
dominant mechanism. Therefore, thermal distribution at BDC has no effect on
evolution of temperature field during compression stroke. However, very small
variations in the initial flow field amplified in the compression stroke, and the
resulting differences close to TDC have significant effect on temperature and the
wall heat flux distribution [103]. The thermal stratification created by different
intake temperatures for each port in HCCI combustion has limited effect in reduc-
ing PRR [104–106]. This observation is clearly explained by the fact that thermal
distribution at BDC has minor effect on the temperature distribution at TDC.
Majority of combustion takes place in the central part of cylinder (between the
cylinder head and piston crown surfaces) from the start of combustion until peak
PRR. Separate boundary-layer combustion along the cylinder head and piston
crown surfaces takes place during the later part of combustion (after peak PRR).
Therefore, thermal stratification between the bulk gases and the boundary layer has
less contribution in controlling the peak PRR. The peak PRR is mainly controlled
by thermal stratification within bulk gases [23]. To get the information about the
magnitude of the thermal stratification and its spatial distribution in the entire
chamber and boundary layer, PLIF imaging study is conducted. Chemilumines-
cence imaging provides valuable information on the effect of thermal stratifica-
tion, but do not show the magnitude of the thermal stratification and its spatial
distribution. [101]. Figure 2.26b shows the evolution of thermal stratification in the
cycle using T maps created by PLIF image in motored cycle. The colour mapping in
the figure shows the stratification between the coolest and hottest regions can be in
the order of 100 K. Natural thermal stratification occurs mostly from wall heat
transfer and turbulent convection [23]; therefore thermal stratification is not
expected to be significant until the charge temperature has increased well above
78 2 Low Temperature Combustion Engines

the wall temperature. Additionally, thermal stratification of the bulk gas needs time
for convection to transport the cooler near-wall gases into the bulk of charge. There
is no noticeable stratification at 305 CA (Fig. 2.26b) as expected, and some distinct
thermal stratification appears at 340 CA. The thermal stratification increases until
piston reaches TDC and then starts decreasing.
Researchers explored various strategies to create the thermal stratification arti-
ficially in the cylinder to increase engine load because natural stratification controls
the maximum PRR. Few studies tried to increase the wall heat transfer by lowering
coolant temperature or increasing the turbulence or swirl (by changing geometrical
design) in the cylinder to create thermal stratification [27, 107]. However, increased
heat transfer can decrease the thermal efficiency also. The combustion phasing
delay provides more time for the evolution of naturally occurring thermal stratifi-
cations, and hence delayed combustion phasing is the most effective strategy to
control the maximum PRR in HCCI combustion [27].
Strictly pure HCCI combustion means that it uses completely homogeneous
charge, and there are no inhomogeneities. As discussed, inhomogeneities are
present even with externally fully premixed charge (using mixing tank) or PFI in
intake manifold [22, 23]. Naturally occurring fuel stratification in premixed charge
does not have significant effect on combustion development [23]. Artificially
created fuel stratification using direct injection has the potential to increase the
high load limit and control the peak PRR [108]. It is very difficult to control the bulk
thermal stratification in real engine to control the PRR; therefore intentional fuel
stratification strategy is explored to increase the high engine load limit. This
strategy is commonly referred to as partial fuel stratification (PFS) strategy, and
their extension is typically known as partially premixed compression ignition
(PPCI). To control the maximum HRR and PRR, the PFS strategy needs to create
sequential auto-ignition (similar to thermal stratification), which requires local
φ-sensitive fuel (Fig. 2.15). Detailed discussion on the determination of
φ-sensitivity is provided in Sect. 2.2.4.1. In PFS strategy, majority of fuel
(80-90%) is injected in the manifold, and small amount of fuel is injected directly
in the cylinder to create fuel stratification in the cylinder. Figure 2.27 shows the
effect of fuel stratification on cylinder pressure and heat release curve. Figure 2.27a
shows that HRR decreases significantly for split injection (stratified) case in
comparison to single PFI injection at same overall φ. The sequential auto-ignition
(because of φ-sensitivity of fuel) due to fuel stratification leads to the increase in the
combustion duration and lowers the HRR in case of split injection strategy.
Figure 2.27b shows the effect of direct injection (DI) quantity on pressure and
HRR using gasoline at intake pressure (Pin) 2 bar as gasoline is φ-sensitive at higher
pressure (Fig. 2.15). Figure 2.27b shows that increasing DI fraction has significant
effect on HRR and cylinder pressure. Zones with higher φ increases with higher DI
quantity and ignite early. Early autoignition of rich zone charge advances auto-
ignition for same CA50, which effectively increases combustion duration
[72]. Higher combustion duration lowers the peak pressure as well as peak HRR
(Fig. 2.27b). Fuel stratification in the cylinder is also used for improving the low
2.2 Homogeneous Charge Compression Ignition 79

Fig. 2.27 (a) Variation of pressure and HRR for homogeneous (premixed) and fuel-stratified
operation (Adapted from [78]). (b) Effect of DI quantity on pressure and HRR using gasoline at
Pin ¼ 2 bar (Adapted from [72])

load combustion efficiency (by advancing the combustion phasing) [109], and
therefore this method can be used to increase the lower load limit of HCCI engine.
In PFS strategy, the partial mixture stratification also forms an associated
thermal stratification (in addition to the typical thermal stratification created by
heat transfer and convection) that tends to partially counteract the ϕ-sensitivity
effect of the fuel on autoignition timing. This thermal stratification is produced by
two reasons. First, rich regions (containing more directly injected fuels) have
relatively greater cooling due to fuel vaporization of the DI fuel. Second, the richer
regions have a lower γ, therefore their temperature increase with compression is
relatively lower. Both effects act in the same direction, causing temperatures at the
time of autoignition to be lowest in the richest (highest ϕm) regions and to be
progressively hotter as the local ϕm decreases. Thus, this temperature distribution
acts opposite of the chemical effect for ϕ-sensitive fuels that causes the richest
80 2 Low Temperature Combustion Engines

regions to autoignite first, followed by sequentially leaner regions [110]. Fortu-

nately, the ϕ-sensitivity effect is sufficiently strong for many fuel-type/operating
condition combinations such that it is more than to overcome the temperature
effects associated with mixture stratification. For these conditions, the richest
regions autoignite first followed by sequentially leaner regions, allowing PFS to
reduce the peak HRR and provide significant improvements in thermal efficiency as
well as higher engine load [110].

2.3 Spark-Assisted HCCI Engine

Spark-assisted compression ignition (SACI) is a technique to set off HCCI com-

bustion in existing engine using an external ignition source [111]. This technique is
a key enabler for SI–HCCI hybrid system [21]. The engine is started in SI mode to
tackle the cold start problem of HCCI and then shift to HCCI mode for lower engine
load operations. At higher engine load requirements, engine can be shifted back to
SI operation mode. Frequent SI–HCCI transition is required due to load variations
during real engine operation. This transition is achieved with SACI by beginning a
flame by spark and progressively varying the quantity of residuals using valve
timing, which leads to combustion of fraction of charge in SI or HCCI mode
[21, 112]. Although SACI is considered mainly as combustion mode transition
method in the beginning [112], further in-depth investigation demonstrated that
SACI strategy can be used to increase high and low load limits of pure HCCI
combustion mode [21].
The SACI process can be divided into four different combustion phases (shown
in Fig. 2.28a), namely, the spark discharge (region A), the early kernel growth
(EKG) (region B), the flame propagation (region C) and the compression ignition
phase (region D) [113]. In SACI, dilute homogeneous mixture is prepared similar to
HCCI engine in the manifold or early intake stroke. However, a small amount of
fuel is injected late in compression stroke to achieve a fuel-rich mixture near the
spark plug for promoting SI and initiate flame [113]. Another study achieved SACI
by direct injection of small amount of fuel (additional to PFI charge) near the spark
plug, facilitating a propagating flame which increases charge pressure and temper-
ature in the cylinder, leading to auto-ignition in the surrounding homogeneous
diluted charge [114]. After the spark discharge, during EKG period (during CA
position shown as B), a small reaction kernel formed starts growing slowly with
very little heat release. During EKG period, flame is too weak for strong propaga-
tion. The end of EKR period is defined as point where second derivative of the
flame area takes a maximum value and eventually rapid heat release takes place
[113]. From this point flame propagation starts in the engine cylinder. Flame
propagation releases additional heat by fuel burning, leading to further compression
heating which drives the compression ignition in the remaining charge. The end of
the flame propagation period is defined as a point, where the combustion transi-
tion takes place to CI combustion. The compression ignition HRR is faster than
2.3 Spark-Assisted HCCI Engine 81

Fig. 2.28 (a) Illustration of different combustion phases in SACI process (Adapted from [113]).
(b) Multimode combustion diagram presenting SI, SACI and HCCI combustion zones in terms of
burned and unburned gas temperature for iso-octant [111]

turbulent flame propagation (due to combustion occurrence over reaction time

scale). Therefore, maxima of second derivative of HRR curve can be used as
transition point to CI combustion phase (Fig. 2.28a) [113, 115]. The CI combustion
phase is comparatively shorter, and the EKG phase is a primary source of combus-
tion phasing cyclic variability in SACI engine [113].
82 2 Low Temperature Combustion Engines

Figure 2.28b shows a conceptual diagram defining the HCCI, SACI and SI
combustion zones in terms of unburned gas temperature (TU) near TDC and burned
gas temperature (TB), which is the corresponding constant pressure adiabatic
temperature. The SACI region is enclosed by operating constraint lines of limits
for excessive knock, flame propagation, auto-ignition, bulk quenching and exces-
sive NOx emissions [111]. The shape of region or temperature range depends on
fuel and engine operating conditions. The parameter φ0 (fuel to charge equivalence
ratio [111]) is used as a primary load metric for air and/or EGR diluted mixtures.
The fuel to charge equivalence ratio (φ0 ) is approximately related to fuel–air
equivalence ratio (φ) by relation φ0  φ (1 residual gas fraction). The most
successful demonstrations of SACI take place at higher equivalence ratios (φ0 ⩾
0.4) and above the proposed turbulent SI flame propagation limit line. For φ0 ⩾ 0.4,
spark can be used to advance and modify the HRR significantly [111, 116,
117]. The spark near TDC is used to improve the combustion stability for lower
equivalence ratios near φ0 ¼ 0.3, but it is not effective in advancing HRR. Analysis
of SACI data in terms of the multimode combustion diagram suggests that spark
assist is mainly useful at moderate and high engine loads for ignition and HRR
control [111]. The two basic requirements for SACI are as follows: (i) the charge
equivalence ratio and temperature must be adequately high to sustain flame prop-
agation, which is fast enough to consume fraction of charge, and (ii) the compressed
gas pressure and temperature must be sufficiently low such that the charge does not
auto-ignite until it is further compressed by flame combustion and heat release [21].
The SACI combustion has the potential of partially decoupling the combustion
phasing from peak HRR, which is one of the major technical challenges in HCCI
engines. Maximum HRR can be controlled by varying the fraction of fuel con-
sumed by flame propagation/auto-ignition at particular combustion phasing of
SACI [118]. The simulations of SACI combustion predicted 43% reduction in the
maximum HRR with increasing spark advance while only 23% reduction in
unburned charge at the time of autoignition. The variation in spark advance changes
fraction of charge consumed by flame leading to variation in the end-gas thermal
and compositional distributions prior to auto-ignition [118]. Figure 2.29 shows the
flame front location and spatial distribution of end-gas temperature in the cylinder
at 5 aTDC in SACI. Temperature is shown by colour and flame front location by

Fig. 2.29 Distribution of unburned gas temperature and flame front location at 5 aTDC for two
spark advanced timings as viewed from the above (a, b) and side (c, d) [118]
2.3 Spark-Assisted HCCI Engine 83

grey surface (Fig. 2.29). More charge is consumed by flame for advanced spark
timings. Therefore, the reduction in maximum HRR during CI regime is dependent
on both the mass of the end-gas charge and the end-gas charge reactivity [118].
An experimental study conducted at constant equivalence ratio by independently
varying spark timing and unburned temperature (by internal and external EGR) in
SACI regime concluded that at constant CA50, a 40% reduction in maximum HRR
and 75% reduction in ringing intensity can be achieved while no penalty on
thermal efficiency [119]. Longer combustion duration and a larger portion of
flame-based heat release are found with decrease in unburned gas temperature and
spark advance. The study also found that bulk end-gas reactions start at a similar
unburned zone temperature (1040 K), regardless of the temperature at spark timing
[119]. Another experimental study concluded that flames could be used to consume a
substantial portion (20–30%) of the charge mass to modulate HRR in SACI com-
bustion [120]. The effect of spark assist on HCCI combustion varies with equiva-
lence ratio for mixtures diluted with preheated intake air. At higher equivalence ratio
(φ > 0.62), spark assist has much larger effect on combustion phasing, and lower
equivalence ratio (φ < 0.38) spark assist does not have significant effect on HCCI
combustion behaviour [121]. Flame propagation cannot occur, or the flame is very
weak to have significant effect on the combustion phasing at excessively dilute
(70%) low load engine operating conditions [111]. In the SACI process, the
deflagrative reaction fronts are sustained in ultra-dilute mixtures (30–60% dilution
[122]) when unburned mixture temperature is increased using charge preheating
(e.g. with intake heaters and/or hot residual gases) [116, 123]. Typically, unheated
conventional SI combustion cannot use such highly dilute mixtures, and there exists
a lean limit (~30% air dilution) for flame initiation/propagation [124]. The elevated
unburned gas temperatures are purposefully used to achieve flame propagation
within a highly dilute mixture and to ensure end-gas auto-ignition in SACI.
The dilution method affects the combustion process in LTC engine, and a variety
of dilution methods (air, EGR, etc.) are used. In HCCI combustion, EGR is more
effective to reduce peak HRR and increases the combustion duration for HCCI than
air at constant combustion phasing. A computation study showed that laminar
burning velocities decrease with increase in EGR dilution (vs. air dilution) due to
thermal and chemical effects of EGR [125]. Study found that the flame speed is
most sensitive to the high temperature chain branching and chain-terminating
reaction rates, where O2 concentration plays an important role. Thus, EGR diluted
charge is less affected by spark assist than air dilute charge for same energy content.
An experimental study reported that early HRR consistent with flame propagation
are more rapid for air dilute SACI condition than EGR dilute engine operating
conditions at constant equivalence ratio, spark timing and combustion
phasing [120].
One of the main advantages of SACI is comparatively higher engine load
operation without excessive peak PRR/knock. In SACI engine, higher load opera-
tion is possible due to mainly three reasons [21]. First, SACI operates at higher
charge density than HCCI by inducting more charge in the cylinder due to com-
paratively lower intake heating/induction of hot residuals. Higher intake charge
84 2 Low Temperature Combustion Engines

Fig. 2.30 (a) The HCCI

operating range extension
by SACI using NVO in
naturally aspirated engine
(Adapted from [122]) and
(b) extension of operating
load regime by SACI using
fixed camshafts and VVT
with asymmetric NVO
(Adapted from [127])

requirement is compensated by heat release during flame propagation. Second,

overall lower HRR leads to higher fuelling rate and/or comparatively more
advanced combustion phasing without excessive PRR. The overall HRR is lower
because significant amount of charge is burned by flame propagation, which is
generally slower than bulk auto-ignition; overall burn duration increases. Typical
flame combustion portions are 30–40% for negative valve overlap (NVO) and
40–55% for positive valve overlap (PVO) [126]. Third, the SACI can be operated
on more retarded combustion phasing, which reduces the HRR during HCCI burn.
Retarded HCCI phasing can allow higher fuelling rates leading to higher engine
load. Figure 2.30 demonstrates the operating range extension of HCCI combustion
by SACI combustion.
2.4 Thermally Stratified Compression Ignition 85

Figure 2.30a shows that the engine load can be well extended above the usable
range of HCCI using SACI flame-based heat release by controlling the overall
HRR. This is achieved in naturally aspirated automotive engine with CR 12.5 using
NVO for gasoline. It is reported that ringing intensity increases with CA50 advance
and combustion stability decreases with retarded combustion phasing. The engine
operating load constraints also depend on control strategy used to achieve SACI.
Figure 2.30b shows the operating regime for HCCI and SACI with fixed cam engine
and with VVT along with asymmetric NVO. Fixed cam operating range is found
comparatively lower in SACI combustion [127]. A study achieved nearly 10 bar
IMEP by continuously retarding spark timing and combustion phasing using large
PVO to control residual concentration [126]. Additionally, SACI offers a more
robust level of cycle-to-cycle control on ignition timings and increases the appli-
cability of advanced LTC in transient multimode engine operation. The spark
assistance can improve the combustion stability near misfire limit [128]. The
spark assistance has been also demonstrated as valuable misfire prevention mech-
anism during rapid combustion mode transition between HCCI and conventional
SI [129].

2.4 Thermally Stratified Compression Ignition

As discussed in Sect. 2.2.4.6, thermal stratification in HCCI combustion signifi-

cantly affects the HRR and ringing intensity. It is also discussed that cylinder wall
condition has a small influence on the thermal stratification in combustion chamber.
Instead, the TDC temperature, combustion phasing and swirl all have a significant
effect on the level of thermal stratification before auto-ignition [130]. Due to ease of
control, several LTC strategies (Sects. 2.5 and 2.6) use intentional fuel stratification
created by direct fuel injection to control the peak HRR and PRR in premixed
compression ignition. However, fuel stratification leads to a risk of higher PM and
NOx emissions due to the locally rich zones present in the cylinder. A new strategy
called thermally stratified compression ignition (TSCI) is proposed to artificially
create the thermal stratification in the cylinder [131]. In this combustion mode,
water is directly injected in the cylinder to control the mean charge temperature
along with temperature distribution in the cylinder, which offers cycle-to-cycle
control of start of combustion and HRR in premixed compression ignition. Water
injection in the port is also investigated in HCCI engine, and study indicated that
evaporative cooling in the intake manifold offers the control of ignition timings by
controlling the intake valve closing temperature [132]. Another study also used
direct water injection using diesel fuel injector to extend the load range premixed
compression ignition [133]. The study showed that the combustion suppression
effect increases with increase in water injection quantity and excessive water
injection leads to higher THC emission along with lower thermal efficiency.
Direct water injection significantly retards the combustion phasing in HCCI
engines [131]. Thermal stratification analysis is used to determine the unburned
86 2 Low Temperature Combustion Engines

temperature distribution before the ignition process. The results indicate that the
temperature distribution is widened by direct water injection because evaporation
of water in the spray locally cools down the areas targeted by the spray. The areas in
the chamber not targeted by water spray remain unaffected. The effectiveness of
cooling is affected by injection timing of water [131]. The study showed that early
injection timing (at or earlier than 80 CA bTDC) is not effective in spreading the
temperature distribution because local cooling achieved by water spray got mix-out
with charge due to larger available time before ignition. For late injection timing
(at 10 CA bTDC or later), water spray not gets enough timing for evaporation, and
effects on temperature distribution are minimal. It is found that approximately
20 CAD at 2000 rpm are required to water spray for breakup and evaporation.
Therefore direct water injection can be used for cycle-to-cycle control of HRR in
LTC engines. The study showed TSCI has a 370% larger operating range than pure
HCCI combustion at particular operating conditions [131]. The TSCI strategy is
less explored in published literature in comparison to fuel stratification strategies
and needs further investigations to be practically implemented. Additionally, this
strategy might have also corrosive effect on contained components, which further
needs to be investigated.

2.5 Partially Premixed Compression Ignition

The LTC strategies can be divided into two broad categories, namely, HCCI and
SCCI, based on degree of charge premixing. In the HCCI engines, fuel vapour is
normally well mixed with the air and residual gases before compression, and charge
is over all lean (Sect. 2.2.1). Laser-based combustion diagnostics and modelling
investigations indicate that HCCI combustion is far from homogeneous even
though charge is well-mixed (Sect. 2.2.4.6). The brake thermal efficiency of
combustion engine is a product of combustion, thermodynamic, gas exchange and
mechanical efficiencies. Higher brake thermal efficiency in HCCI engines is
achieved due to lower pumping losses and higher thermodynamic efficiency
because of higher compression ratio and higher γ of working fluid in the cylinder
due to lean mixture engine operation. A study showed that the thermodynamic
efficiency of HCCI can reach up to 50% or even slightly higher, but there are some
issues with combustion, gas exchange and mechanical losses [134]. The combus-
tion efficiency is around 90% for most of the operating conditions. The HCCI
engine is mainly operated in low load range, and hence parameters such as friction
losses and cooling losses have a large impact on the achieved brake efficiency. The
overall brake thermal efficiency is found in the range of 32–42% [134]. The other
novel strategies investigated to overcome the limitation of HCCI mode are SACI
and SCCI modes. Several studies are conducted to investigate various aspects of
SACI (see Sect. 2.3), but the applicability is limited because it is very hard to
master, and it has higher cyclic variations, higher pumping loss and relatively
higher NOx emissions.
2.5 Partially Premixed Compression Ignition 87

In SCCI mode, intentional thermal or fuel stratification is used to overcome the

limitations of HCCI mode. The HCCI load control by creating bulk thermal
stratification is still a challenge, and research efforts continue (Sect. 2.4). Fuel
stratification creation in the combustion chamber is easy using direct fuel injection,
and presently all required technologies to achieve fuel stratification are available.
This category of combustion mode is generally termed as partially premixed
compression ignition (PPCI) or sometimes abbreviated as partially premixed com-
bustion (PPC), which uses direct fuel injection to modulate mixing time. The PPCI
can be considered as an intermediate process, which is a mix of conventional diesel
combustion and HCCI combustion [135]. In PPCI fuel is injected during the
compression stroke to allow sufficient premixing before ignition. Ideally, all the
fuels should be injected prior to start of ignition (positive ignition dwell), but at the
same time, fuel injection should not be too early to produce excessive HRR once
ignition starts. The ignition dwell is the time duration from end of injection to start
of ignition. Typically, conventional diesel engine has negative ignition dwell. The
positive ignition dwell in PPCI mode provides time for some degree of premixing to
all the injected fuels. Generally, positive ignition dwell is obtained by increasing
ignition delay using EGR [17]. Shorter fuel injection durations also help to obtain
positive ignition dwell in PPCI mode. The PPCI strategy is investigated for diesel-
like (high cetane fuels) and gasoline-like (low cetane fuels) fuels. First diesel PPCI
and then gasoline PPCI mode are discussed in the following sections.

2.5.1 Diesel PPCI

Conventional diesel engine combustion has a soot–NOx trade-off, and generally,

methods of reducing soot lead to increases in NOx emissions and vice versa. To
obtain simultaneous reduction in soot and NOx emissions, charge must be well
mixed (to avoid regions with unfavourable carbon–oxygen ratios, charge must also
be rich enough to prevent misfire), and combustion results into lower combustion
temperatures (below NOx formation temperature). To achieve lower combustion
temperature, most of the strategies used cooled EGR. To achieve premixed charge
combustion, high degree of fuel–air mixing is required before ignition. The forma-
tion of premixed mixture is mainly determined by location and timing of fuel
injection. Since diesel is a non-volatile fuel, direct injection is preferred for
premixed combustion. In diesel PPCI mode, two different strategies of direct
injection are used, namely, early direct injection and late direct injection with
respect to conventional diesel combustion (CDC) injection timings. In early direct
injection strategy, fuel is injected directly into the cylinder during the compression
stroke, well before TDC to allow time for sufficient fuel-air mixing. The generic
combustion with early direct injection is typically known as premixed charge
compression ignition (PCCI) [136]. In late direct injections, fuel is injected just
before TDC or towards the start of expansion stroke. This injection concept is the
basis of the modulated kinetics (MK) and highly premixed late injection (HPLI)
88 2 Low Temperature Combustion Engines

combustion strategies in LTC regime [137, 138]. In MK combustion mode, high

swirl ratio is used to improve the mixture formation; and to increase the ignition
delay, a reduction in compression ratio and cooled EGR is used. High injection
pressure in combination with higher nozzle orifice diameter is used to reduce
the fuel injection duration. The combination of lower oxygen concentration
(by EGR) and retarded fuel injection timings leads to simultaneous reduction in
NOx and soot emissions in MK concept.
In early direct injection PCCI combustion, various strategies exist such as
premixed lean diesel combustion (PREDIC), multiple stage diesel combustion
(MULDIC), homogeneous charge intelligent multiple injection combustion
(HiMICS), premixed compression ignition (PCI), uniform bulky combustion sys-
tem (UNIBUS) and narrow angle direct injection (NADI). These strategies use one
or more injectors, and single or multiple injections in an engine combustion cycle.
The start of fuel injection is limited by misfiring due to overmixing (in very early
injection) in most of the early injection systems. In the first study of PREDIC
strategy, different injector configurations and positions are investigated in combi-
nation with variation of injection timings for mixing the fuel [139]. Fuel is injected
simultaneously with two side injectors in order to avoid collision of the fuel spray
with the cylinder liner. The mixing of air and fuel is promoted by using long
ignition delay. Low NOx and soot emission is achieved because of the good fuel-
air mixing and lower combustion temperature. However, results are mainly affected
by the fuel injection timings, geometrical arrangement of injectors and charge
dilution. Another study presented several improvements in PREDIC strategy and
shows that maximum achievable load in PREDIC mode is 50% of naturally
aspirated conventional diesel combustion [140]. This limit comes from the limit
in excess air ratio to maintain low NOx and from maximum cylinder pressure
limitations (knocking). To extend the PREDIC operation range, a second injection
around TDC is introduced in the MULDIC system, which combines a lean
premixed combustion (PREDIC) with a more conventional diffusion combustion
separated in time [141]. Similarly, the PCI combustion concept uses a conventional
direct injection diesel injector, an early start of injection and a reduced compression
ratio to achieve low NOx emission [142]. The UNIBUS system used a fuel injector
with a pintle-type nozzle featuring a large hole and a bulbous protrusion to reduce
penetration and keep the fuel mixture in the centre of the cylinder away from the
cylinder wall [143]. In NADI combustion strategy, the combustion chamber design
and spray formation are optimized for early injection by using nozzles with a
narrow spray include angle (around 80 ) such that fuel can be injected in early
compression stroke at relatively lower charge density and temperature environment
[144]. In this system, engine can operate in conventional mode at full load with a
careful design of the piston bowl geometry. Most of the early direct injection
systems can achieve lower NOx and soot emissions only at lower engine load
conditions.
The ignition characteristics and heat release in PPCI combustion mode are
significantly different from conventional diffusion-controlled diesel combustion.
In LTC mode, initial ignition reactions occur at slower rate than conventional diesel
2.5 Partially Premixed Compression Ignition 89

Fig. 2.31 Variations of HRR for early and late direct injection PPCI and conventional diesel
engine at similar engine load conditions (Adapted from [5, 145, 146])

combustion due to higher dilution and premixing along with relatively cooled
cylinder gases (due to EGR). Figure 2.31 shows the heat release characteristics in
early and late direct injection PPCI vis-
a-vis conventional diesel engine at similar
engine load conditions (~4 bar IMEP). In conventional diesel combustion, two
distinct combustion phases of premixed combustion and mixing-controlled
90 2 Low Temperature Combustion Engines

combustion are observed after the ignition delay period. In the conventional diesel
engine, fuel injection starts near TDC (~11 bTDC) in compressed hot (~990 K) and
high density (16 kg/m3) air [5]. The fuel spray mixes with hot air, and vaporization
starts by taking thermal energy from surrounding hot air. The decrease in sensible
thermal energy due to vaporization manifests as decrease in HRR during ignition
delay period. The fuel–air mixture prepared during ignition delay period auto-
ignites simultaneously during premixed phase combustion leading to high peak
HRR. The HRR during this period is mainly controlled by chemical kinetics of the
mixture formed during the ignition delay period. The HRR decreases significantly
after premixed combustion phase due to limited charge preparation by diffusion-
controlled process. In mixing-controlled combustion phase, combustion rate is
mainly controlled by evaporation and mixing process in the spray instead of
chemical kinetics. In typical PPCI combustion mode, the HRR exhibits additional
characteristics (two-stage heat release), which is not present in conventional diesel
HRR. In the early injection PPCI, fuel injection starts comparatively early (~22
bTDC), where cylinder charge is relatively cooler (~770 K) and less dense (11 kg/m3)
[17]. The first-stage heat release begins near 15 bTDC and ends near 11 bTDC
(Fig. 2.31). The second-stage heat release rate is very high (nearly three times of
conventional HRR) and is observed in early PPCI mode. In this case, no significant
mixing-controlled heat release is observed. The late injection PPCI strategy has
similar characteristics to early PPCI with a comparatively lower peak HRR and
more significant mixing-controlled combustion phase.
Large fraction of diesel fuel typically consists of long straight-chain alkanes,
which mainly governs the two-stage heat release in LTC regime. Although diesel is
a mixture of hundreds of chemical species, significant insight into kinetics of diesel
combustion can be obtained by investigation of single straight-chain alkane surro-
gate. The n-heptane is typically considered as surrogate fuel for diesel as its ignition
delay and HRR characteristics in engine are similar to diesel [17]. Figure 2.32
shows the ignition behaviour and temperature evolution in early and late injection
PPCI mode using n-heptane as fuel. Important insight into heat release process in
PPCI mode can be obtained by analysing the evolution of few important species
during combustion. Additionally, the behaviour of species like formaldehyde
(HCHO) and hydroxyl radical assists in the analysis of laser diagnostic data as
these species are accessible in optical diagnostic techniques [17]. Figure shows that
a pool of ketohydroperoxide (KPH) species is formed in pre-ignition reaction just
before first-stage heat release. The KPH species plays an important role in first-
stage ignition in PPCI combustion. The sequence of reaction starts with O2, and
subsequently, hydrogen atom abstraction from parent fuel molecule occurs by
hydroxyl and hydroperoxy radicals that results into alkyl radical formation. The
alkyl radical reacts with oxygen molecule to form alkylperoxy radicals (RO2). After
several isomerization reactions, a relatively stable KPH is formed. The decompo-
sition of the RO2 and KPH molecules generates OH and other radicals, which
further reacts with the parent fuel molecules. This process accelerates the ignition
reactions. At higher temperatures (above 850), formation of RO2 is curtailed
(reaction R3, Sect. 2.2.2) due to higher reverse reaction rate, and this slows down
2.5 Partially Premixed Compression Ignition 91

Fig. 2.32 Ignition

behaviour of vapour-phase
n-heptane for (a) early
injection (initial
temperature 802 K) and (b)
late injection (initial
temperature 870 K) PPCI
conditions [17]

the reactions building KHP pool (Fig. 2.32). Three important species (formalde-
hyde, hydrogen peroxide and CO) are formed in first-stage heat release [17]. The
KHP decomposition leads to the formation of oxygenated hydrocarbons, which
forms formaldehyde after decomposition. A pool of H2O2 is formed and it has an
92 2 Low Temperature Combustion Engines

important role in HTHR (Sect. 2.2.2). The H2O2 decomposition exceeds produc-
tion, as charge temperature rises above 1000 K leading to formation of pool of
OH radicals, which accelerates the overall oxidation rate. The CO is also formed
during the first-stage heat release. However, CO is mostly formed in the early part
of the second-stage heat release (Fig. 2.32). Figure shows that second-stage heat
release in late injection PPCI has comparatively longer span than early injection
PPCI, which is also shown in Fig. 2.31. However first-stage heat release is com-
paratively smaller. Early injection PPCI mode does not suffer from incomplete
combustion in leaner zones to the same degree as late injection PPCI mode. In the
late injection PPCI conditions, the charge cools during the expansion stroke that
prevents the progress of second-stage ignition. The more complete discussion on
combustion reactions in PPCI can be found in original study [17].
Progress of combustion process in PPCI mode is significantly different from
conventional diesel combustion mode as illustrated by conceptual model presented
in Fig. 2.33. In the early injection conditions, fuel is injected at low temperature and
less dense environment leading to comparatively higher liquid length than conven-
tional combustion mode because of fast jet penetration. The instantaneous liquid
length decreases by vaporization of downstream portion of fuel jet due to mixing by
propagation of entrainment wave (indicated by vertical lines) around 7 ASI after
peak injection rate. The fuel is fully vaporized within 1 after the end of injection
(EOI) in the PPCI condition investigated in the study [17]. Ignition processes for the
PPCI jet diverge from the conventional diesel jet after the initial jet penetration. In
conventional diesel combustion (CDC), first fuel-rich portions of the spray jet
ignite, and soot precursors formed during the premixed combustion, followed by
the soot formation. In PPCI mode, fuel injection ends before second-stage heat
release, and OH radical formed near the peak of the second-stage heat release. The
OH and formaldehyde distributions do not overlap spatially in the chamber. Some
fuel-rich zones are apparent in the downstream jet, where formaldehyde is con-
sumed, but OH does not appear due to the too rich mixtures for accumulating
significant OH radical. After few CAD, soot precursors and soot appear within the
rich pockets, typically surrounded by OH radical. The soot pockets in PPCI jets are
formed in far downstream, near the head vortex, while in conventional diesel jets,
soot fills the jet cross-section. Refer to the original study for detailed description of
the PPCI model in LTC regime [17].
The main disadvantage of diesel PPCI mode is requirement of very high EGR
(upto 80%) for achieving simultaneously low soot and NOx at higher engine load
conditions [147]. The combustion efficiency is very low (~85%) at this very high
EGR operating conditions. The compression ratio also needs to be reduced to have
lower soot concentration, which limits the expansion ratio. Figure 2.34 shows the
variations of NOx and soot with EGR at different compression ratios. The figure
shows that at higher compression ratio, soot formation starts increasing for EGR
greater than 50% (Fig. 2.34b) and below 50% EGR, NOx emission is higher. For
simultaneous reduction of the NOx and soot emissions, lower compression ratio and
higher EGR are required in diesel partially premixed combustion. The use of
thermodynamically better compression ratio similar to present diesel engines
2.5 Partially Premixed Compression Ignition 93

Fig. 2.33 Conceptual model for conventional and LTC (single injection EGR diluted) diesel
combustion for heavy-duty engine [17]

(~CR ¼ 18) significantly lowers the ignition delay due to the increase in temper-
ature and pressure in the cylinder. Lower ignition delay limits the premixing of fuel,
and this will limit the achievable high load with PPCI using diesel.
94 2 Low Temperature Combustion Engines

Fig. 2.34 Variation of NOx and soot emissions with EGR at 8 bar IMEP for (a) CR ¼ 12.4 and (b)
CR ¼ 17.1 (Adapted from [147])

To overcome the problem of shorter ignition delay, low cetane (high octane) fuel
is used in PPCI mode, and very low NOx and soot are obtained at moderate EGR
(50%) at a 15 bar net IMEP [148]. The combustion efficiency approaches to 98%
even at very close to stoichiometric engine operation with 50% of EGR. This result
leads to the further research in low cetane (high octane, gasoline-like) fuels in PPCI
engines. The gasoline PPCI is discussed in the next section.
2.5 Partially Premixed Compression Ignition 95

2.5.2 Gasoline PPCI

Inhomogeneities (thermal and fuel stratifications) play a key role in controlling PRR
in premixed compression ignition engines (Sect. 2.2.4.6). Thermal stratification has
a potential to control HRR in premixed compression ignition engines. Natural
thermal stratification exists in cylinder, but it is difficult to create or control thermal
stratification in the bulk charge of cylinder. The very unusual methods along with
sophisticated control are required to achieve variations in the bulk temperature
distribution that utilizes the thermal stratification strategy well (Sect. 2.4). Fuel
stratification is typically achieved by varying the fuel injection timings or by using
multiple injection strategies. This strategy creates zones of different fuel concentra-
tions in the combustion chamber, which leads to sequential auto-ignition. The zones
with higher fuel concentration auto-ignite first and comparatively lower concentra-
tion zone next. Sequential auto-ignition in the distribution of equivalence ratio leads
to the reduction in peak HRR and extends the combustion duration. The advantage of
fuel stratification depends on the φ-sensitivity of fuel (Fig. 2.15). The more
φ-sensitive fuel has higher potential to utilize the fuel stratification strategy. Fuels
like gasoline show higher φ-sensitivity at higher intake pressure in comparison to
naturally aspirated conditions. Therefore, gasoline can be used in PPCI combustion
at higher intake pressure conditions more effectively depending on the stratification
levels in the cylinder. As discussed in Sect. 2.5.1, diesel PPCI has limitations on
higher load limits. To overcome the limitations of diesel PPCI combustion, gasoline
PPCI combustion is proposed. This strategy has various names by different groups of
researchers such as PPCI [149], gasoline compression ignition (GCI) [150, 151],
partially premixed combustion (PPC) [135], low temperature gasoline combustion
(LTGC) [110] and gasoline direct injection compression ignition (GDCI) [152]. All
these strategies used gasoline-like single fuel in the cylinder. Dual fuel-stratified
PPCI is known as reactivity-controlled compression ignition (RCCI), which uses
two different reactivity fuels for combustion (discussed in Sect. 2.6).
Figure 2.35 demonstrates the LTC strategies based on the level of fuel stratifi-
cation in the cylinder from fully premixed to heterogeneous conventional diesel
combustion. Using diesel, two different LTC strategies exist, namely, RCCI and
PCCI. In RCCI combustion, low reactivity (gasoline-like) fuel is premixed, and
high reactivity (diesel-like) fuel is directly injected to create reactivity stratification
in the cylinder. Partially premixed combustion using diesel is known as PCCI (see
Sect. 2.5.1). The gasoline PPCI strategies can be divided into three main categories:
(i) partial fuel stratification (PFS), (ii) moderate fuel stratification (MFS) and (iii)
heavy fuel stratification (HFS) (Fig. 2.35) based on degree of fuel stratification in
the combustion chamber [6]. The gasoline PPCI is first introduced and experimen-
tally demonstrated in 2005 [148, 153]. Consequently several studies are conducted
in PFS [21, 71–73, 78, 154–157], MFS [152, 158–163] and HFS [164–172] regime.
In PFS strategy, homogeneous part of charge is prepared by either port fuel
injection or very early direct fuel injection, and stratification is created using
subsequent direct injection of fuel in small quantity. Stratification is created in
such a way that sequential auto-ignition takes place while keeping ultralow NOx
and soot emissions.
96 2 Low Temperature Combustion Engines

Fig. 2.35 Illustration of LTC strategies in advanced CI engines using gasoline and diesel based on
level of stratification in the cylinder [6]

In the MFS strategy, stratification level is higher in comparison to PFS, and it is

increased by reducing the amount of premixed fuel. Generally in MFS strategy, all
the fuels are injected directly in the cylinder during compression stroke and a fuel
injection event near TDC (around 15% of total fuel) to trigger combustion [6]. The
HFS strategy has highest level of fuel stratification and it typically has very less
premixed fuel. Multiple direct injection events relatively close to TDC is used for
charge preparation in HFS strategy. The HFS strategy uses higher fuel injection
pressure in comparison to PFS and MFS for completing the injection events before
the start of ignition.
The PFS and MFS strategies have comparatively lower combustion efficiency
especially at lower engine loads due to lower combustion temperature achieved
using leaner mixture. However, reduced heat loss and higher ratio of specific heat
(due to leaner mixture) of charge lead to higher thermal efficiency. The NOx
emissions are strongly connected with the level of fuel stratification and decrease
as mixture becomes more homogeneous [6]. Therefore, higher amount of EGR is
required in HFS strategy for LTC operation and lower NOx emissions. A study
conducted for a variety of gasoline fuels (in PPC combustion mode with HFS-like
strategy) shows that around 50% EGR is required to keep NOx emission at low level
[166]. This EGR percentage is still below than the EGR required for diesel PPCI.
This requirement puts pressure to have efficient air handling system to supply high
level of EGR and overall lean mixture especially at higher engine load operations.
In HFS strategy, ignition timing is governed by injection timings; therefore com-
bustion phasing can be controlled on a cycle-to-cycle basis. The MFS strategy has
2.5 Partially Premixed Compression Ignition 97

comparatively lower PRR and longer combustion duration than HFS strategy. In
MFS strategy, the combustion phasing is typically controlled by the last fuel
injection near TDC.
In the gasoline PPCI mode, fuel injection strategy is central to overall gasoline
compression ignition concept. A study suggested the multiple late injections of
gasoline (RON91) in GDCI strategy (in MFS regime). In GDCI mode, directly
injected gasoline vaporizes and mixes rapidly at low injection pressure typical of
gasoline direct injection (GDI) engines [152]. Injection strategy used in GDCI
mode is shown in Fig. 2.36a on φ-T map. The colour contours in φ-T map show
simulated CO emissions. The fuel injection process can have one, two or three
injections per cycle in cylinder during compression stroke (indicated as quantities

Fig. 2.36 (a) Illustration of GDCI fuel injection strategy. Reprinted with permission Copyright
© 2014 SAE International [161]. (b) Variation of combustion phasing, COV of IMEP and ISNOx
as function of last injection timing at moderate EGR (Adapted from [152])
98 2 Low Temperature Combustion Engines

Q1, Q2 and Q3 in φ-T map). Every injection event starts in the upper left of the φ-T
map (liquid), which rapidly vaporizes and mixes to equivalence ratio less than two
prior to the start of ignition. At the beginning of ignition, fuel–air mixture is
stratified to obtain controlled heat release and stable ignition [161]. The ideal fuel
injection process is represented by ideal “stratification line” in Fig. 2.36a. The
combustion process must occur in the rectangular region (Fig. 2.36) to achieve low
NOx and PM emissions simultaneously. To minimize CO emissions also, combus-
tion must occur in the region 0 < φ < 1.2 and temperature 1300 < T < 2200
K. Therefore, at the beginning of combustion (around TDC), all parcels in the
cylinder should be no richer than φ of approximately 1.2 [152]. To achieve the
GDCI combustion mode, study used centrally mounted injector, a shallow pent roof
combustion chamber, and 15:1 compression ratio engine. To support fuel injection-
controlled mixture stratification, a quiescent open chamber design is used. Swirl,
tumble and squish motions are minimized because excessive charge motion may
destroy the fuel stratification created during the fuel injection events [161]. The
piston bowl shape and injector spray characteristics are also matched for typical
GDCI injection timings for significant reduction in piston surface area and lower
propensity for cylinder wall wetting.
Figure 2.36b illustrates the GDCI combustion process using a double injection
strategy and moderate EGR for 6 bar IMEP at 1500 rpm. A characteristic U-shaped
response of CA50 is obtained by varying the start of injection (SOI) for the last fuel
injection even in the cycle, and hence, combustion phasing to be controlled in one
of two regimes (GDCI early and late injection). In GDCI late injection regime,
combustion phasing retards as injection timing is delayed similar to the response of
a CDC system. In this regime, ignition dwell (IDW) is relatively low. In the GDCI
early injection regime, combustion phasing retards as injection timing advances.
Authors of the study attributed this observation to the significant increase in IDW
due to lower cylinder pressure and temperature [152]. Combustion stability is
measured by coefficient of variation (COV) of IMEP, and combustion stability
decreases as combustion phasing retards (Fig. 2.36b).
Another study investigated the nature of heat release in gasoline PPCI engines
using PFI/DI split injection (60/40) strategy at 1200 rpm, 15.1 compression ratio
and 1.8 bar intake pressure [149]. The PFI is used for preparing the premixed
charge, and DI is used to create stratification which controls the combustion
phasing. Figure 2.37 shows the variation of maximum pressure rise rate (MPRR),
ignition timing (CA10) and combustion phasing (CA50) with DI timings at 15%
EGR. Three regimes (A, B and C) are found in behaviour of maximum PRR
(marked in Fig. 2.37).
In regime A, as injection is retarded from 30 aBDC (intake) to 105 aBDC, the
maximum PRR reduces significantly (from 9 MPa/ms to 1.8 MPa/ms), and the
corresponding retard is also observed in the combustion phasing (CA10 and CA50).
However, the combustion becomes erratic and ignition fails with further injection
retard (beyond 105 aBDC) but retarded injection timings beyond 140 aBDC; the
combustion stabilizes again [149]. In the regime A, ignition delay is longer due to
injection in low temperature and pressure charge. With SOI retard, ignition delay
2.5 Partially Premixed Compression Ignition 99

Fig. 2.37 Maximum PRR, CA10 and CA50 versus DI timings in PPCI engine (Adapted from [149])

decreases, but it is not enough to compensate for the retard of SOI and hence CA10
delays with retarded SOI. In regime B, further injection retard leads to the increase in
maximum PRR, and it goes up to very high peak of 37 MPa/ms. In this regime, fuel
stratification is also higher, and ignition delay is shorter due to injection at rela-
tively higher cylinder pressure and temperature. The fuel-rich zones in the high
temperature environment have a shorter ignition delay, which compensates for the
retarded SOI. Therefore, retarded injection timing results in advancement of ignition
(CA10). In regime C, maximum PRR decreases with further retard of DI injection
timing, and with late DI timings, the heat release becomes more a two-part process,
in which the mass fraction burned curve first shows a slower burn followed by a
faster burn [149]. In this regime, ignition delay is very short due to fuel injection in
higher temperature and pressure conditions. The time needed to physically evapo-
rate and mix fuel overpowers the gain in shortening the delay at this stratification
level. Therefore, trend reverses again and the combustion retards with the retard of
SOI. This study further concluded that maximum PRR correlates mostly with the
combustion phasing that changes the mixture temperature at point of ignition. It is
also suggested that the main effect of SOI is to change the combustion phasing, and
the MPRR responds to the combusting phasing change [149].
A recent study developed a reduced chemical mechanism for five-component
gasoline surrogate and investigated the heat release characteristics in GCI engine
[173]. This study concluded that with comparatively early direct injection timings
(52 aTDC), the GCI combustion process can be understood as a partially
sequential auto-ignition by the competition between the charge cooling effect and
the ϕ-sensitivity effect for the stratified mixture. The partial sequential autoignition
lowers the peak PRR. In case of late DI (5 aTDC) timing, the GCI combustion
duration is increased due to the mixing process between the late injected fuel and
100 2 Low Temperature Combustion Engines

the well-mixed mixture. Thus, the GCI combustion can be decoupled into a
partially sequential auto-ignition and a subsequent non-premixed combustion.
A study investigated the degree of stratification at low load engine conditions for
different injection strategies using OH chemiluminescence imaging at constant NVO,
inlet pressure (1.3) and λ (3.5) and inlet temperature (80) [162]. Figure 2.38 shows the
sequential combustion images and HRR for single injection. In this case, there is a

Fig. 2.38 OH – chemiluminescence and heat release rate for single, double and triple injection in
PPC combustion (Adapted from [162])
2.6 Reactivity-Controlled Compression Ignition 101

zone with rich mixture starting auto-ignition at 4 aTDC, and consequently five
different combustion zones are appearing at around 7 aTDC. There is no significant
difference observed in HRR from single and double injection. The early first injection
in the compression stroke gives sufficient time for the stratified charge to mix with the
surroundings before start of combustion [162]. In case of three injection, peak HRR is
much lower, and combustion duration is comparatively longer. In this case, auto-
ignition initiates at the outer rim of the combustion chamber. The first fuel injection
pulse creates a rather homogeneous charge, and second injection pulse produces a
stratification in the combustion chamber and generating zones with higher reactivity.
The fuel injection in third pulse burns high reactivity zones first, and then the reaction
extends to the relatively leaner zones [162].
Fuel effects on ignition delay and low temperature reactions (LTR) during PPC
are analysed using different research octane number (RON) surrogate fuels
[174]. The study found that the increase in ignition delay as RON value increases.
Ignition delay in PPC engine is more influenced by alcohol than aromatics due to
increases in RON value. A proportional correlation between LTR and ignition delay
is observed. Another interesting finding of the study is that the ethanol and toluene
amplified the LTR phase, while n-heptane suppressed it in PPC mode which is in
contradiction to what was reported in HCCI mode [174]. Another study investi-
gated nine fuels in the boiling point range of gasoline but with an octane number
spanning from 69 to 99 in PPC mode in the range of 5 and 25 bar gross IMEP at
1250 rpm [166]. The study concluded that a fuel in the boiling point range of
gasoline with an octane number in the range of 70 is the best fuel for PPC engine
from maximum load to idle condition. The study conducted in HFS regime used
high boost level combined with 50% of EGR and 1.4 of λ, which allowed
around 53.5% gross indicated efficiency throughout the entire test load sweep. The
boost pressure, λ, injection pressure and main injection timings used for different
loads in PPC are illustrated in Fig. 2.39 for RON70 fuel. Figure shows that very
high boost pressure along with higher EGR rate is required for PPC mode combus-
tion at higher engine loads. Fuel injection pressure used is also very high
(~2400 bar). Therefore, to run in PPC mode at higher engine load, very efficient
air handling and fuel injection system is required.

2.6 Reactivity-Controlled Compression Ignition

2.6.1 RCCI Fundamentals

Reactivity-controlled compression ignition (RCCI) strategy is also in the category

of partially premixed compression ignition, and it uses two fuels of different
reactivities. Thus, RCCI can be named as dual fuel partially premixed combustion.
The premixed compression ignition modes have two major challenges of combus-
tion phasing and combustion rate control. To have better control on combustion
102 2 Low Temperature Combustion Engines

Fig. 2.39 EGR, λ, absolute inlet pressure, common rail pressure and start of main injection as a
function of load in PPC engine (Adapted from [162])

phasing, direct fuel injection strategy is typically used, but sufficiently long ignition
delay is provided for mixing. Due to positive ignition dwell in these strategies,
injection and ignition process is decoupled. Thus, chemistry time scales should be
controlled to achieve appropriate combustion phasing. The GCI strategies (Sect.
2.5.2) have shown significant potential to achieve higher engine load at higher
efficiency as well as lower NOx and PM emissions. In GCI strategies, achieving
combustion at lower engine load is difficult due to lower auto-ignition reactivity of
gasoline. Diesel has higher reactivity but it has difficulty of controlling combustion
phasing at higher engine loads in PPCI mode (Sect. 2.5.1). Therefore, studies
explored the premixed compression ignition using blends of different fuel
2.6 Reactivity-Controlled Compression Ignition 103

reactivities [175–177]. A study concluded that the best fuel for HCCI combustion
has autoignition reactivity between diesel and gasoline fuels [176]. Studies
suggested that different fuel blends are required at different engine operating
conditions such as a high reactivity fuel at lower engine loads and a low reactivity
fuel at higher engine loads. Thus, in-cylinder blending of two different reactivities
can be beneficial in PPCI combustion.
Thermal and fuel stratification in the cylinder is a promising strategy to control
the HRR in premixed charge CI engines (Sect. 2.2.4.6). Partial fuel stratification
and thermal stratification introduce the gradients into the auto-ignition characteris-
tics (i.e. the ignition delay) of the charge [175]. The HRR depends on speed of
propagation of multiple ignition fronts in the chamber. Speed of a propagating
subsonic spontaneous ignition front (Sig) is related to spatial gradient in ignition
delay (∇τ) by inverse relation (Sig ¼ 1/j∇τj) [175, 178, 179]. Inverse relation of
ignition delay gradient to ignition front propagation shows that controlling the
gradients in ignition delay is critical to control the HRR of premixed combustion.
Ignition delay gradient depends on state of mixture and the state of surrounding
mixtures that interact through transport and compression heating. Ignition delay is a
function of temperature, pressure, composition, equivalence ratio and fuel type. The
dependency of ignition delay and associated HRR on fuel type suggests that
in-cylinder fuel blending can be beneficial to control the HRR also. Thus,
in-cylinder fuel blending of two fuels with different reactivities has the potential
for simultaneous control of the HRR and combustion phasing, and hence, RCCI
combustion is investigated in detail [175].
The RCCI combustion is a dual fuel LTC strategy. In this strategy, two fuels of
different auto-ignition reactivities (high and low reactivity) are used for premixed
charge preparation by in-cylinder fuel blending. This strategy adopts multiple
injection strategies and appropriate EGR level for controlling the in-cylinder
reactivity to optimize the combustion phasing and combustion duration, which
leads to higher thermal efficiency along with simultaneous reduction of NOx and
soot emissions to ultralow level [180, 181]. Figure 2.40 schematically illustrates
RCCI combustion concept. The low reactivity fuels (gasoline-like fuels) are
injected in intake manifold using port fuel injection (PFI) system, which premixes
with air during intake stroke, due to their volatile nature. The high reactivity fuel
(diesel-like fuels) is directly injected in the cylinder using DI system with a single,
double or triple injection strategy during compression stroke. The early injected
high reactivity fuel targets the squish region of the combustion chamber, whereas
the relatively late injected high reactivity fuel acts as an ignition source
[181, 182]. The dilute gasoline–air mixture does not auto-ignite without diesel
injection in cylinder. The ratio of low and high reactivity fuel and injection timings
of high reactivity fuel can be used to control the RCCI combustion process.
Injection of two different auto-ignition reactivity fuels creates both the reactivity
and equivalence ratio (φ) stratifications in the entire combustion chamber. Reac-
tivity stratification adds an additional fast-response control parameter in the form of
fuel chemistry, enabling the global fuel properties to change across the operating
map.
104 2 Low Temperature Combustion Engines

Fig. 2.40 Schematic diagram of RCCI combustion mode (Adapted from [182])

Effect of fuel distribution due to start of injection (SOI) timings of high reac-
tivity fuel in RCCI combustion (with iso-octane as the premixed low reactivity fuel
and n-heptane as the direct injected high reactivity fuel) is demonstrated by optical
imaging study [183]. The high-speed combustion luminosity imaging in Fig. 2.41
shows distinctly different combustion events for the early, mid and late SOI
timings. Figure 2.41 shows the effect of injection timing on the combustion process
as the combustion event progresses in the cylinder due to φ- and reactivity strati-
fication. The differences in combustion characteristics are due to differences in the
mixing times of the three cases. The early SOI timing results in rapid energy release
due to an under-stratified (overmixed) charge, and the late SOI timing results in
rapid energy release due to an over-stratified charge (under-mixed) [183]. The mid
SOI timing (50 ATDC) is a representative of typical RCCI injection timings and
demonstrated a long heat release duration due to the stratified nature of the
moderately premixed charge.
Figure 2.42 depicts the PRF (primary reference fuel), equivalence ratio and
ignition delay distributions near the cylinder wall from the optical study
2.6 Reactivity-Controlled Compression Ignition 105

Fig. 2.41 High-speed movie sequence showing liquid fuel sprays and natural luminosity for the
cases with common rail SOI timings of 155 , 50 and 15 ATDC. The number in the upper
left-hand corner of each image shows the crank angle at which the image is acquired. Reprinted
with permission Copyright © 2012 SAE International [183]. Regions coloured red indicate
saturation

[183]. The numbers shown agaist PRF are octane number of PRF fuel (Fig. 2.42).
Large variations in PRF distribution are observed for 15 SOI, and very rich
regions exist for this injection timing. Study concluded that the high equivalence
ratio regions (φ > 0.6) are accountable for the high HRR (for SOI 15 ) because of
the almost same ignition delay in these regions (Fig. 2.42b, rightmost image). The
ignition delay sensitivity to φ decreases with increased φ beyond about φ ¼ 0.6.
Due to lack of ignition delay sensitivity, almost entire mixture ignited in a very
small combustion duration [183, 184]. Combustion timing and duration are depen-
dent on the ignition delay gradients in RCCI combustion. It is well known that short
ignition delays lead to advanced combustion phasing, while long ignition delays
result in retarded combustion phasing. The smaller ignition delay gradients result in
short combustion durations and large ignition delay gradients result in long com-
bustion durations in RCCI engine [184]. The ignition delay gradient is very small
106 2 Low Temperature Combustion Engines

Fig. 2.42 (a) Ensemble-averaged PRF and equivalence ratio distributions and (b) corresponding
ignition delay (calculated from constant volume adiabatic simulations) at 5 CA ATDC for
different SOI timings. Reprinted with permission Copyright © 2012 SAE International [183]

for overmixed case (145 SOI case) leading to short combustion duration and
high HRR. The relatively larger ignition delay gradient (produced by both the
reactivity and equivalence ratio distributions (Fig. 2.42)) is responsible for the
longer combustion duration and lower HRR for the 50 SOI condition.
The evolution of PRF distribution with respect to crank angle position is shown
in Fig. 2.43 for RCCI combustion with two direct injections. The figure shows that
at 50 ATDC (shortly after the first injection), vapour fuel has penetrated near to
the cylinder walls. As piston moves by 45 ATDC position, the fuel vapour starts
spreading along the cylinder liner. The PRF number in the upstream zone rapidly
starts decreasing due to the further penetration of vapour fuel and entraining
ambient gas (air and iso-octane) [185]. At 45 ATDC position, second injection
occurs and mixes with cylinder charge with slower penetration because of the
higher ambient densities at time of second injection. The ensemble-averaged PRF
number in the head of the jet is approximately 40 (i.e. a 40–60 blend of iso-octane
and n-heptane) at 25 ATDC position. It can be noticed that much of the fuel from
the first injection has mixed to produce a relatively uniform mixture in the squish
region of the chamber with a PRF number ranging from 55 to 60, outside region is
influenced by the second fuel injection [185]. Figure 2.43 also shows that the charge
continues to mix in the duration between 21 and 5 ATDC. Highest reactivity
region remains positioned in the downstream portion of the jet, near the cylinder
2.6 Reactivity-Controlled Compression Ignition 107

Fig. 2.43 Sequence of ensemble-averaged PRF maps at different crank angle positions during the
common rail injection event and prior to ignition event in RCCI engine [185]

wall. The fuel reactivity reduces with decreasing distance from the cylinder centre.
A more complete discussion can be found on original study [185].
The reactivity in RCCI combustion chamber can be characterized into global
reactivity and reactivity gradient/distribution [181]. Global reactivity of charge is
purely estimated by the amount of each fuel and the auto-ignition reactivity indices
of fuels. The reactivity gradient is associated with the spray penetration of directly
108 2 Low Temperature Combustion Engines

Fig. 2.43 (continued)

injected high reactivity fuel and the entrainment of surrounding charge. A study
investigated the individual effect of reactivity, equivalence ratio stratification and
temperature stratifications using combination of optical diagnostics and chemical
kinetic modelling [185]. Figure 2.44a depicts the predicted ignition delays (esti-
mated using chemical kinetic simulations in constant volume chamber) as a func-
tion of distance from the cylinder centre to various combinations of φ-, PRF and
temperature stratification. Figure 2.44 depicts that temperature stratification com-
petes with the φ- and PRF stratification (i.e. zones having high φ and low PRF
numbers achieve lower temperatures due to evaporative cooling). Thus, the absence
of the temperature stratification increases the spread between the least reactive and
most reactive regions (Fig. 2.44a). Temperature stratification alone has a small
2.6 Reactivity-Controlled Compression Ignition 109

Fig. 2.44 (a) Effects of equivalence ratio (φ), PRF and temperature stratification on radial
variation of the predicted ignition delay [185] and (b) calculated cylinder and HRR with and
without fuel reactivity gradient (Adapted from [186])

effect on the ignition delay (Fig. 2.44a, case7), and PRF stratification has most
significant effects (Fig. 2.44a, case4). The study concluded that PRF number
stratification is the dominant factor for controlling the ignition location and growth
rate of the reaction region [185]. The equivalence ratio (φ) has comparatively
smaller effect but it is significant.
Figure 2.44b demonstrates the effect of fuel reactivity gradient on HRR and
PRR. In case of only φ-stratification, every zone ignites almost simultaneously,
which leads to shorter combustion duration and higher PPR. In case of φ- and PRF
110 2 Low Temperature Combustion Engines

stratification, zones that are more reactive ignite significantly early in the cycle, and
energy release between zones is staged such that the combustion duration is
considerably increased leading to lower PRR (Fig. 2.44b). Therefore, effects of
φ- and reactivity stratifications must be considered simultaneously in the RCCI
combustion mode [187].

2.6.2 RCCI Fuel Management

Fuel reactivity gradient has a crucial role in RCCI combustion characteristics.

Researchers have extensively investigated different types of low and high reactivity
fuels with their different combinations in RCCI combustion. The RCCI combustion
fuelling strategy is of mainly two types: dual fuel strategy and single fuel strategy
(along with cetane improver). In dual fuel strategy, two different reactivity fuels are
used to create the reactivity gradient in the combustion chamber, and therefore two
fuel tanks are required in this particular strategy [181]. This requirement to refuel
two fuel systems can result in decreased market acceptance. An alternative to dual
fuel strategy is suggested using a single fuel stock with the same fuel delivery
system but with the addition of a highly reactive cetane improver in small concen-
tration to the DI fuelling system [187]. High reactivity cetane improver creates
reactivity gradient in the cylinder, and φ-gradients are created by direct injection
similar to dual fuel strategy. Two common cetane number (CN) improvers are
2-ethylhexyl nitrate (2-EHN) and ditertiary butyl peroxide (DTBP). Significant
increase in CN is achieved by using only a small percentage of either of these
chemicals [188, 189].
In the dual fuel strategy, gasoline and natural gas are most widely used as low
reactivity fuel in RCCI combustion with diesel as high reactivity fuel [180, 181,
190]. Renewable fuels such as methanol, ethanol and iso-butanol are also consid-
ered as substitute of gasoline for low reactivity fuel in RCCI combustion [191–
194]. A study compared the combinations of methanol/diesel and gasoline/diesel
and indicated that the use of methanol could extend the RCCI engine operating load
with relatively lower requirement of EGR [193]. Another study investigated dif-
ferent types of low reactivity fuels (various blends of ethanol/gasoline) in a RCCI
engine and found that lower premixed ratio of E85 (85% of ethanol in volume) is
required to obtain a stable combustion [195]. Typically, diesel and biodiesel are
used as high reactivity fuels in RCCI engine [181, 196]. The RCCI engine fuelled
by gasoline/ULSD (ultralow sulphur diesel) and gasoline/biodiesel showed that
gasoline/biodiesel has comparatively lower NOx and HC emission in comparison to
gasoline/ULSD at the selected operating conditions [196]. Another study showed
that the more stability to the cycle-to-cycle transitions could be achieved in the
tested NG/biodiesel fuelled RCCI engine due to higher cetane and oxygen
containing biodiesel fuel [197].
Similarly, several studies investigated the effect of cetane improvers in single
fuel strategy of RCCI combustion. The effect of CN improvers in RCCI engine is
2.6 Reactivity-Controlled Compression Ignition 111

investigated by port fuel injection (PFI) of gasoline and direct injection (DI) of
gasoline mixed with small amounts of DTBP [198]. Study showed that the RCCI
combustion could be realized by adding as little as 2% by volume DTBP into the
direct injected gasoline. Another study showed similar trend with EHN
[199]. Although there is availability of DTBP, 2-EHN is a more common additive,
and it is used in refineries to increase CN of commercial ULSD blends
[187]. Although EHN is a more effective CN improver than DTBP, the fuel-
bound nitrogen in EHN can increase engine-out NOx emissions [200]. Detailed
discussion on fuel management strategies (dual and single fuel) can be found in
review studies [180, 181, 187].

2.6.3 RCCI Engine Management

Fuel injection strategy plays an important role in combustion and performance

characteristics of RCCI combustion engine. The main controllable parameter is the
injection parameters of high reactivity fuel such as number of injection (single,
double and triple injection events), injection timings, injection pressure and quan-
tity of injection in each event. The effect of SOI timing of high reactivity fuel on
combustion characteristics and charge quality before ignition is discussed in Sect.
2.6.1. The number of injection events affects the reactivity gradient distribution in
the combustion chamber (Fig. 2.43). The number of pulses in the fuel injection
strategy can also affect the engine load. A study investigated three different
injection strategies, early single injection, early double injection and late double
injection, in RCCI engine at constant combustion phasing (CA50 ¼ 6 ATDC)
[201]. This study found that late double injection could be used to extend operating
load range while maintaining acceptable emission levels and peak pressure rise rate
(PPRR) values. A numerical study showed that the split injection fraction of high
reactivity fuel could affect the originating region (bowl or squish) of combustion
and consequently the PPRR and ringing intensity [190]. However, study found
there is less effect of split fuel fraction on engine performance and emissions.
A study used three different fuel injection strategies to achieve the dual fuel
operation in the entire engine operating map with constraint on maximum PRR and
peak cylinder pressure below 15 bar/CAD and 190 bar, respectively [202]. Fig-
ure 2.45 demonstrates the proposed fuel injection strategies and typical HRR from
each strategy at 1500 rpm. Based on the level of charge stratification, three regimes
are characterized as fully premixed RCCI, highly premixed RCCI and dual fuel
diffusion combustion. At higher engine load, diffusion-controlled HRR can be
noticed (Fig. 2.45). In fully premixed RCCI (around 8 bar IMEP), a double diesel
injection pulse with highly advanced injection timings is used. Ultralow NOx and
soot emissions obtained in fully premixed RCCI regime due to the high mixing time
available before start of combustion (SOC). In highly premixed strategy (40% up to
75% load), a double diesel injection pulse is used with the second injection event
occur close to TDC. The first injection is used to improve the reactivity in the
112 2 Low Temperature Combustion Engines

Fig. 2.45 Fuel injection strategies on the engine map (left) and typical HRR curves (right) from
each injection strategy at 1500 rpm [202]

crevice zone. The second diesel injection acts as an ignition source as injection
occurs at the higher pressure and temperature in the cylinder [202]. Diffusive fuel
strategy (from 75% up to full load) used a single diesel injection with a retarded fuel
injection timing that permits injecting higher amount of diesel to achieve the
required load.
Most of the RCCI studies used higher fuel injection pressure (greater than
500 bar) using common rail injection (CRI) hardware for injection of high reactivity
fuels [180]. The feasibility of using low-pressure fuelling strategy by commercial
gasoline direct injector (GDI) hardware is investigated [203]. Low-pressure fuel-
ling strategy has capability to provide cost savings. Study showed that with
conventional diesel combustion (CDC) piston, low-pressure direct injection of
high reactivity fuels lowers the combustion efficiency by 1% while maintaining
benefits of NOx and soot emission at 6 bar IMEP in RCCI engine. Study indicated
that low-pressure direct injection of high reactivity fuel is not able to provide
sufficient mixing to offer equivalent engine performance as with the use of high-
pressure direct injection with CDC piston [203]. Same study optimized the RCCI
piston geometry and found comparable engine performance for both low-pressure
direct injection and high-pressure direct injection RCCI engine operations. Thermal
efficiencies are increased 5% absolute in comparison to CDC piston with 500 bar
CRI operation by using optimized piston with 150 bar GDI operation. This gain in
thermal efficiency is attributed to significant reductions in heat transfer via lower
piston surface area coupled with reduced and elongated heat release rates for the
peak efficiency cases [203].
Appropriate combustion chamber design can be effective in reduction of emis-
sion formation, without affecting the RCCI engine performance. The piston bowl
geometry affects the charge mixing process, which can influence the reactivity, and
φ-stratification in RCCI engine that results into different combustion
2.6 Reactivity-Controlled Compression Ignition 113

characteristics. Piston surface area also affects the heat transfer process, which has
an impact on combustion temperature and engine efficiency. A study optimized the
piston shape for premixed fuel by reducing surface to volume ratio for lower heat
transfer [204]. Decreased surface to volume ratio of RCCI piston bowl can increase
the throat diameter of the piston bowl which is subsequently beneficial to the early
SOI timing of RCCI injection strategy [205]. A study investigated RCCI combus-
tion in a light duty multi-cylinder engine at three operating points according to the
US-FTP cycle by comparing OEM (original equipment manufacturer) piston and
custom-machined pistons designed for RCCI operation [206]. The piston bowl
profile was optimized to reduce unburned fuel emissions and piston bowl surface
area using genetic algorithm optimization. The custom RCCI piston was designed
for lower HC emissions by reducing the squish area and crevice volume. Heat
transfer losses are reduced by reducing the surface area and by increasing the piston
bowl radius and decreasing the bowl depth (for constant compression ratio (CR)).
Reduction in heat transfer loss leads to higher thermal efficiency. Additionally, the
high levels of piston-induced mixing produced by narrow or deep piston bowl
increases the heat transfer coefficient and further add the heat transfer losses.
Therefore, the more quiescent combustion chamber created by the wide or shallow
bowl is beneficial for premixed charge compression ignition (PCI) strategies [187].
The RCCI mode was proposed to reduce the requirement of EGR level in
premixed compression ignition engines [177]. However, the assistance of EGR is
required mainly at higher engine load operation, to mitigate the high PRR in RCCI
engines [181]. Studies demonstrated that PPRR, NOx and soot emissions are
reduced by increasing EGR rate [207, 208]. The unburned hydrocarbons (UHC)
and CO emissions are increased with the use of EGR in most of the studies due to
the formation of local fuel-rich region resulted from the reduced the oxygen
concentration in the cylinder by increasing EGR rate [181]. A study numerically
evaluated the necessity of EGR utilization in RCCI engine fuelled with methanol
and diesel at medium loads [209]. Study suggested that the EGR requirement
depends on the initial cylinder temperature and initial temperature below a critical
value (380 K in this case); EGR can be unemployed and the methanol fraction can
be varied to maintain the optimal performance.
Another important issue in RCCI engine management is cylinder balancing. In a
multi-cylinder engine, there exist significant cylinder-to-cylinder variations in the
initial conditions such as EGR, cylinder temperature, fuel rail pressure, trapped
mass, etc. [180]. These imbalances between cylinders lead to comparatively poor
performance due to excessively delayed or advanced combustion timings in every
cylinder because RCCI combustion is a kinetically controlled process. The varia-
tions in the combustion phasing results in significant variation of exhaust emissions
in each cylinder. To regulate these variations in RCCI engine, the total amount of
fuel and ratio of low to high reactivity fuel for each cylinder can be varied to match
the combustion phasing and IMEP over all the cylinders [180]. A study used an
“adjustment factor” multiplier to the PFI and DI duration commands for each
cylinder to achieve the cylinder balancing in RCCI engine [206].
114 2 Low Temperature Combustion Engines

2.6.4 Direct Injection Dual Fuel Stratification

A variety of PPCI strategies are developed to increase the level of control on

ignition timing and HRR while maintaining the benefits of HCCI combustion.
Typically, PPCI strategies produce fuel stratification in cylinder by direct injection
of either a portion or all of the fuels, in order to increase the operating range
and combustion stability. Fuel stratification leads to φ- and temperature stratifica-
tion in the cylinder resulting into variation in local ignition delay in entire cylinder.
The variation in local ignition delay increases the combustion duration and retard
the combustion phasing, which leads to the reduction in HRR and combustion
noise. There exists a continuum of stratification levels between PFS (mostly
premixed) and PPC (partially premixed), and each strategy has their relative merits
(Sects. 2.2.4.6 and 2.5.2). Dual fuel RCCI combustion has shown high thermal
efficiency and superior controllability with low NOx and soot emissions. The RCCI
strategy addresses the combustion control by extending the concept of control by
additional fuel reactivity stratification. The RCCI combustion has demonstrated
gross indicated thermal efficiencies over 56% and levels of soot and NOx below the
EPA 2010 heavy-duty emission regulation limits [204]. The RCCI combustion
emits relatively high levels of HC and CO, due to low wall temperatures and the
inability for complete oxidation of fuel in crevice regions. In RCCI strategy, shape
and phasing of heat release can be adjusted with much greater fidelity (than is
possible in single fuel strategies), by adjusting the ratio of high and low reactivity
fuel. At particular engine operating condition, the combustion phasing can be
controlled by the fuel ratio, and to a lesser degree, by the timing of the high
reactivity fuel injections [210]. However, the reactivity range of the fuels being
used limits the operating range of RCCI combustion. The amount of high reactivity
fuel required tends to decrease towards zero as engine load increases (well before
full load), and there is no longer any direct control over the combustion process
[211]. To extend the high load operation of RCCI combustion, high octane fuels
such as E85 (an 85% ethanol and 15% gasoline mixture) [212] and methane [190]
are used as low reactivity fuels to have longer ignition delay in comparison to
gasoline. Engine operation at lower compression ratio is another strategy to
increase the load of RCCI engine at the expense of fuel conversion efficiency
[204, 213]. However, stratification of both (high and low reactivity) fuels is one
possible path to achieve additional control on RCCI combustion for engine oper-
ation with conventional fuels and compression ratios of typical diesel engines. This
strategy is known as direct injection dual fuel stratification (DDFS) in published
literature [210, 211].
The DDFS strategy combines the benefits of RCCI and PPC by injecting both
gasoline and diesel directly, enabling control over the in-cylinder distribution of
both fuels [211]. In this strategy, gasoline is injected during the intake stroke and it
is effectively premixed. Similar to RCCI, diesel is injected 40–60 before TDC to
create a reactivity gradient that allows for precise control over the LTHR. After
LTHR is complete, and just as the main heat release begins (0–10 BTDC), gasoline
2.6 Reactivity-Controlled Compression Ignition 115

Fig. 2.46 (a) Drawing of

cylinder showing the
injector locations and
orientations in typical
DDFS [214]. (b)
Conceptual illustration of
injection strategies in DDFS
(Adapted from [210])

is injected a second time and burns in a diffusion-controlled manner, allowing for

considerable increase in load without increasing peak HRR and combustion noise
[211]. Figure 2.46 illustrates the injector locations and orientations along with fuel
injection strategy for implementation of DDFS strategy.
In DDFS strategy, the timing and quantity of the diesel injection afford a large
degree of control over the early heat release. The reactivity and equivalence ratio
stratification created by direct diesel injection extends the heat release and gives
control over when combustion begins. The relative impact of reactivity and equiv-
alence ratio on second injection is evaluated by injecting gasoline and diesel as both
have direct injection arrangement [210]. Figure 2.47a shows the effect of injecting
gasoline instead of diesel on cylinder pressure and HRR. Retarded combustion
phasing and high instability of the gasoline operation are observed when second
116 2 Low Temperature Combustion Engines

Fig. 2.47 (a) Cylinder

pressure and HRR for
DDFS with either diesel or
gasoline as the second
injection fuel. (b) Cylinder
pressure and HRR for
DDFS with a sweep of
SOI-2 from 200 to
30 ATDC (Adapted from
[210])

injection fuel was switched from diesel to gasoline. Higher HRR and shorter
combustion duration is also observed for gasoline injection case. Since fuel injec-
tion pressure and injection duration are the same for both the fuels, therefore
this difference in combustion characteristics is purely a fuel effect. The main fuel
differences are volatility and reactivity, which influence mixing and ignition delay,
respectively. The figure also illustrates the limited control availability with equiv-
alence ratio stratification alone and the flexibility gained by direct injection of
highly reactivity fuel [210]. Figure 2.47a also depicts that diesel injection pressure
had little effect on HRR. Relatively lower thermal efficiency and higher unburned
HC and CO emissions are also observed for the case of gasoline injection.
This observation suggests that φ-stratification alone is insufficient for full potential
of this strategy.
The same study also conducted the effect of diesel fuel ratio and start of injection
(SOI) sweep on combustion characteristics of DDFS. Figure 2.47b shows the effect
of diesel SOI timings on cylinder pressure and HRR in DDFS strategy. Few points
are omitted for clarity and all data points can be seen in original study
[210]. Figure depicts that before 40 ATDC SOI timings, combustion phasing
2.6 Reactivity-Controlled Compression Ignition 117

advances with increase in diesel stratification. At SOI of 30 ATDC, the large
bump at the beginning of the heat release suggests that the diesel may be
transitioning to a diffusion-limited combustion. After this SOI timing, combustion
phasing retards with increase in diesel stratification in the cylinder. Detailed
analysis of heat release curve shows that LTHR is advanced and increased slightly
in magnitude for SOI up to 50 ATDC, and a slight shift is observed in the
beginning of the LTHR at 40 ATDC. Major change in LTHR occurs at
30 ATDC, and LTHR was significantly diminished after this SOI, which is
interpreted by authors as the onset of diffusion-limited combustion of the diesel
injection [210].
This study identified three possible combustion regimes for the second injection
of high reactivity fuel: premixed, RCCI and diffusion-limited. Study found the
RCCI regime is superior in several ways, including improved control over CA50
and noise; substantial reduction of HC, CO and soot emissions; improved indicated
efficiency; and NOx emission comparable to the premixed regime. DDFS has a
unique ability of independent control on the combustion duration (with a near TDC
gasoline injection) and the combustion phasing (with high reactivity diesel injec-
tion). Advantage of this ability can be availed by operating engine at optimum
combustion phasing across the engine map, and combustion duration can be
controlled without retarding the combustion phasing. Engine can be operated at
higher compression ratios, and higher thermal efficiency can be achieved using this
strategy. The combination of the diesel and late gasoline direct injections offers
robust fast-response control mechanisms with a wide range of authority [210].

2.6.5 RCCI vis-

a-vis Other LTC Strategies

Several studies have demonstrated the superior performance of RCCI strategy over
conventional diesel combustion (CDC) strategy. A study comprehensively investi-
gated the RCCI combustion on a HD engine over a wide range of engine operating
loads by varying the gasoline-to-diesel ratio at fixed diesel injection strategy
[215]. The study found that RCCI operation achieved NOx reduction of nearly
three orders of magnitude, six times lower soot and 16.4% higher gross indicated
efficiency in comparison to CDC without EGR. A detailed comparison between
CDC and RCCI combustion at matched conditions of load, speed, boost pressure
and combustion phasing is conducted [216]. The study found 24% lower integrated
piston heat transfer and 25  C lower mean surface temperature for RCCI engine in
comparison to CDC. The SOI timing strongly influences CDC heat flux but has a
negligible effect on RCCI heat flux, even in the limit of near TDC high reactivity
fuel injection timings. These observations indicate that the high reactivity fuel
injection does not have significant effect on the thermal environment.
Figure 2.48 depicts the distributions of the in-cylinder temperature for the CDC,
HCCI and RCCI regimes for constant combustion phasing at 3 CA ATDC. In CDC
engine, the fuel vapour is mainly concentrated around the piston lip and bowl
118 2 Low Temperature Combustion Engines

Fig. 2.48 Distributions of the in-cylinder temperature for the CDC, HCCI and RCCI regimes for
constant combustion phasing at 3 CA ATDC [217]

during the combustion process due to close to TDC direct injection of fuel. This
results into the high temperature region in the cylinder (Fig. 2.48a), which leads to
higher NOx emissions. Sufficiently premixed fuel–air mixture and the avoidance of
local high temperature regions in HCCI and RCCI strategy lead to substantially
lower combustion temperature (Fig. 2.48b, c) and lower NOx emissions. Some
premixed and low temperature fuel vapour that exists near the cylinder liner and in
the crevice region cannot be fully oxidized, resulting in high HC and CO
emissions [217].
A comprehensive comparative investigation of RCCI, PPC and HCCI mode is
conducted using single-cylinder engine for comparison of performance and emis-
sions, combustion sensitivity to intake conditions and the control ability using fuel
injection strategy [218]. This study used PFI for low reactivity fuel and direct
injection of high reactivity fuel in RCCI and only PFI for HCCI combustion. In PPC
strategy, single fuel is used in both PFI and DI systems. All the three strategies have
shown higher indicated thermal efficiency and acceptable coefficient of variation
(COV) of IMEP. The RCCI combustion has longer combustion duration and
subsequently lower peak pressure rise rate compared to HCCI and PPC. The
RCCI operation has highest indicated thermal efficiency but it has lower combus-
tion efficiency. It is suggested that higher efficiency is due to the lower heat transfer
losses stemming from the lower maximum PRR [218]. This study also perturbed the
intake conditions (intake temperature and pressure) from the baseline to analyse the
sensitivity of combustion phasing and its control. It was noted that baseline
combustion phasing can be easily recovered through small changes in the global
fuel reactivity in RCCI strategy. However, in PCC strategy, the baseline combus-
tion phasing was unrecoverable when combustion becomes advanced due to
2.6 Reactivity-Controlled Compression Ignition 119

variations in intake conditions. On the contrary, retarded combustion phasing (due

to variation in intake condition) is recoverable by appropriately increasing the level
of fuel stratification.
Another study compared the RCCI and GCI strategy at high load conditions at
1300 rpm for optimized engine operating conditions [214]. In this study, RCCI
strategy uses dual direct injection of high and low reactivity fuels (Fig. 2.46a). A
non-parametric regression analysis tool, known as component selection and
smoothing operator (COSSO) for multivariate smoothing splines, is used to choose
a single optimum operating point for comparison of RCCI and GCI strategies
[219]. The COSSO fits a response surface model to the genetic algorithm data
from all the runs combined. The combustion and emission characteristics are
compared to optimal engine operating condition. The overall combustion charac-
teristics are very similar with the close to TDC injection initiating combustion for
both strategies. The main differences observed in these two (RCCI and GCI)
strategies are soot emissions. The difference in soot formation in RCCI and GCI
is due to the state of local equivalence ratio and temperature in the cylinder during
combustion. Figure 2.49 depicts the variation of local φ and temperature at various
crank angle positions after start of combustion (ASOC) on φ-T map. The NOx and
soot formation regions in φ-T map are identified from HCCI simulations of
mixtures of gasoline and air from the study [220].
The combustion is delayed in GCI mode due to longer ignition delay as gasoline
has lower cetane index (high octane fuel). In RCCI strategy, combustion occurs
during injection event due to shorter ignition delay of diesel fuel, which results in
higher local φ and more soot formation. In RCCI case, at 10 ASOC (close to
CA50), the final diesel fuel injection is 60% complete and local φ reached above
3 (Fig. 2.49). In GCI case, the final gasoline injection is finished at 7 CA, and post
combustion mixing has reduced the peak local φ to 1.2. The higher local φ during
combustion leads to comparatively higher soot formation in RCCI strategy. How-
ever, at this expense, RCCI strategy has comparatively better control on combus-
tion phasing due to shorter ignition delay. The peak local temperatures achieved in
the RCCI and GCI cases are 2242 K and 2056 K, respectively, due to utilization of
high EGR levels. Thus, lower NOx emissions are obtained in both cases due to low
peak temperatures.
Sensitivity to operating conditions in RCCI and GCI strategy is also evaluated in
the same study [214]. Figure 2.50 depicts the sensitivity of outputs (IMEP, gross
indicated efficiency (GIE), CA50 and peak pressure rise rate (PPRR)) to changes in
inputs (EGR, intake pressure and temperature, direct injection (DI) mass) for RCCI
and GCI strategies.
Figure 2.50 shows that both RCCI and GCI strategies are very sensitive to EGR
fluctuations at this high load condition. In contrast to this observation, a sensitivity
study at a medium load condition found the RCCI strategy to be insensitive to EGR
[221]. Figure 2.50 shows that the GCI strategy is more sensitive to EGR than the
RCCI strategy (Fig. 2.50). The global φ of the GCI strategy is close to stoichio-
metric (0.98), which makes it to be very sensitive to oxygen concentration. Intake
valve closing temperature and pressure also affect the combustion phasing
120 2 Low Temperature Combustion Engines

Fig. 2.49 Ф-T map of the charge at various crank angle positions for the optimum RCCI and
gasoline compression ignition (GCI) combustion strategies [214]
References 121

Fig. 2.50 Sensitivity of

outputs to changes in inputs
for RCCI and GCI strategies
[214]

considerably for the GCI strategy, whereas the RCCI strategy is relatively insensi-
tive to these fluctuations.
The characteristics of dual fuel RCCI and single fuel PPCI show their potential
for higher thermal efficiency along with reduction of NOx and soot emission to an
ultralow level. Detailed combustion, performance and emission characteristics of
HCCI, RCCI and PPC strategies are discussed in Chaps. 6, 7 and 8, respectively.

References

1. Zheng M, Reader GT, Hawley JG (2004) Diesel engine exhaust gas recirculation––a review
on advanced and novel concepts. Energy Convers Manage 45(6):883–900
2. Majewski WA, Khair MK (2006) Diesel emissions and their control, vol. 303. SAE technical
paper
122 2 Low Temperature Combustion Engines

3. Johnson TV (2002) Diesel emission control: 2001 in review (No. 2002-01-0285). SAE
technical paper
4. Dec JE (2009) Advanced compression-ignition engines—understanding the in-cylinder pro-
cesses. Proc Combust Inst 32(2):2727–2742
5. Dec JE (1997) A conceptual model of DI diesel combustion based on laser-sheet imaging
(No. 970873). SAE technical paper
6. Dempsey AB, Curran SJ, Wagner RM (2016) A perspective on the range of gasoline
compression ignition combustion strategies for high engine efficiency and low NOx and
soot emissions: effects of in-cylinder fuel stratification. Int J Engine Res 17(8):897–917
7. Lewander M (2011) Characterization and control of multi-cylinder partially premixed com-
bustion. PhD thesis, Lund University. ISBN 978-91-7473-148-4
8. Kitamura T, Ito T, Senda J, Fujimoto H (2002) Mechanism of smokeless diesel combustion
with oxygenated fuels based on the dependence of the equivalence ration and temperature on
soot particle formation. Int J Engine Res 3(4):223–248
9. Kim D, Ekoto I, Colban WF, Miles PC (2008) In-cylinder CO and UHC imaging in a light-
duty diesel engine during PPCI low-temperature combustion. SAE Int J Fuels Lubr
1:933–956. (2008-01-1602)
10. Dec JE, Canaan RE (1998) PLIF imaging of NO formation in a DI diesel engine
(No. 980147). SAE technical paper
11. Som S, Aggarwal S (2009) Modeling diesel spray flame lift-off using detailed chemistry and a
new primary breakup model. In: 47th AIAA aerospace sciences meeting including the new
horizons forum and aerospace exposition, p 666
12. Pickett LM, Siebers DL (2006) Soot formation in diesel fuel jets near the lift-off length. Int J
Eng Res 7(2):103–130
13. Pickett LM, Siebers DL, Idicheria CA (2005) Relationship between ignition processes and
the lift-off length of diesel fuel jets (No. 2005-01-3843). SAE technical paper
14. Chartier C, Aronsson U, Andersson O, Egnell R, Collin R, Seyfried H et al (2009) Analysis of
smokeless spray combustion in a heavy-duty diesel engine by combined simultaneous optical
diagnostics (No. 2009-01-1353). SAE technical paper
15. Naber JD, Siebers DL (1996).Effects of gas density and vaporization on penetration and
dispersion of diesel sprays (No. 960034). SAE technical paper
16. Akihama K, Takatori Y, Inagaki K, Sasaki S, Dean AM (2001). Mechanism of the smokeless
rich diesel combustion by reducing temperature (No. 2001-01-0655). SAE technical paper
17. Musculus MP, Miles PC, Pickett LM (2013) Conceptual models for partially premixed
low-temperature diesel combustion. Prog Energy Combust Sci 39(2):246–283
18. Dobbins RA (2002) Soot inception temperature and the carbonization rate of precursor
particles. Combust Flame 130(3):204–214
19. Huestis E, Erickson PA, Musculus MP (2007) In-cylinder and exhaust soot in
low-temperature combustion using a wide-range of EGR in a heavy-duty diesel engine
(No. 2007-01-4017). SAE technical paper
20. Colban WF, Miles PC, Oh S (2007) Effect of intake pressure on performance and emissions
in an automotive diesel engine operating in low temperature combustion regimes (No. 2007-
01-4063). SAE technical paper
21. Dec JE (2015) Advanced compression ignition combustion for high efficiency and ultralow
NOx and soot. In: Encyclopedia of automotive engineering, vol 1. Wiley, Chichester,
pp 267–306
22. Richter M, Engstr€om J, Franke A, Aldén M, Hultqvist A, Johansson B (2000) The influence
of charge inhomogeneity on the HCCI combustion process (No. 2000-01-2868). SAE tech-
nical paper
23. Dec JE, Hwang W, Sj€oberg M (2006) An investigation of thermal stratification in HCCI
engines using chemiluminescence imaging (No. 2006-01-1518). SAE technical paper
References 123

24. Hultqvist A, Christensen M, Johansson B, Richter M, Nygren J, Hult J, Aldjn M (2002) The
HCCI combustion process in a single cycle-speed fuel tracer LIF and chemiluminescence
imaging (No. 2002-01-0424). SAE technical paper
25. Johansson B (2007) Homogeneous charge compression ignition: the future of IC engines? Int
J Veh Des 44(1–2):1–19
26. Battin-Leclerc F (2008) Detailed chemical kinetic models for the low-temperature combus-
tion of hydrocarbons with application to gasoline and diesel fuel surrogates. Prog Energy
Combust Sci 34(4):440–498
27. Saxena S, Bedoya ID (2013) Fundamental phenomena affecting low temperature combustion
and HCCI engines, high load limits and strategies for extending these limits. Prog Energy
Combust Sci 39(5):457–488
28. Erlandsson O (2002) Early Swedish hot-bulb engines-efficiency and performance compared
to contemporary gasoline and diesel engines (No. 2002-01-0115). SAE technical paper
29. Gussak LA, Turkish MC, Siegla DC (1975) High chemical activity of incomplete combustion
products and a method of prechamber torch ignition for avalanche activation of combustion
in internal combustion engines (No. 750890). SAE technical paper
30. Onishi S, Jo SH, Shoda K, Do Jo P, Kato S (1979) Active thermo-atmosphere combustion
(ATAC)-a new combustion process for internal combustion engines (No. 790501). SAE
technical paper
31. Noguchi M, Tanaka Y, Tanaka T, Takeuchi Y (1979) A study on gasoline engine combustion
by observation of intermediate reactive products during combustion (No. 790840). SAE
technical paper
32. Iida N (1994) Combustion analysis of methanol-fueled active thermo-atmosphere combus-
tion (ATAC) engine using a spectroscopic observation (No. 940684). SAE technical paper
33. Duret P, Venturi S (1996) Automotive calibration of the IAPAC fluid dynamically controlled
two-stroke combustion process (No. 960363). SAE technical paper
34. Oguma H, Ichikura T, Iida N 1997 A study on adaptability of alternative fuels for lean burn
two-stroke ATAC engine (No. 972097). SAE technical paper
35. Yamaguchi J (1997) Honda readies activated radical combustion two-stroke engine for
production motorcycle. Automot Eng 105(1):90–92
36. Ishibashi Y, Asai M (1998) A low pressure pneumatic direct injection two-stroke engine by
activated radical combustion concept (No. 980757). SAE technical paper
37. Najt PM, Foster DE (1983) Compression-ignited homogeneous charge combustion
(No. 830264). SAE technical paper
38. Thring RH (1989) Homogeneous-charge compression-ignition (HCCI) engines
(No. 892068). SAE technical paper
39. Christensen M, Hultqvist A, Johansson B (1999) Demonstrating the multi fuel capability of a
homogeneous charge compression ignition engine with variable compression ratio
(No. 1999-01-3679). SAE technical paper
40. Yao M, Zheng Z, Liu H (2009) Progress and recent trends in homogeneous charge compres-
sion ignition (HCCI) engines. Prog Energy Combust Sci 35(5):398–437
41. Lu X, Han D, Huang Z (2011) Fuel design and management for the control of advanced
compression-ignition combustion modes. Prog Energy Combust Sci 37(6):741–783
42. Lavy J, Dabadie JC, Angelberger C, Duret P, Willand J, Juretzka A et al (2000) Innovative
ultra-low NOx controlled auto-ignition combustion process for gasoline engines: the
4-SPACE project (No. 2000-01-1837). SAE technical paper
43. Maurya RK, Agarwal AK (2011) Experimental study of combustion and emission charac-
teristics of ethanol fuelled port injected homogeneous charge compression ignition (HCCI)
combustion engine. Appl Energy 88(4):1169–1180
44. Stanglmaier RH, Roberts CE (1999) Homogeneous charge compression ignition (HCCI):
benefits, compromises, and future engine applications (No. 1999-01-3682). SAE technical
paper
124 2 Low Temperature Combustion Engines

45. Affens WA, Sheinson RS (1980) Autoignition: the importance of the cool flame in the
two-stage-process. Loss Prev 13:83–88
46. Machrafi H (2010) HCCI combustion chemistry, reduced kinetic mechanisms and controlling
strategies. In: Lackner M, Winter F, Agarwal AK (eds) Handbook of combustion, New
technologies, vol 5. WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
47. Machrafi H (2008) Experimental validation of a kinetic multi-component mechanism in a
wide HCCI engine operating range for mixtures of n-heptane, iso-octane and toluene:
influence of EGR parameters. Energy Convers Manag 49(11):2956–2965
48. Tanaka S, Ayala F, Keck JC (2003) A reduced chemical kinetic model for HCCI combustion
of primary reference fuels in a rapid compression machine. Combust Flame 133(4):467–481
49. Hwang W, Dec J, Sj€oberg M (2008) Spectroscopic and chemical-kinetic analysis of the
phases of HCCI autoignition and combustion for single-and two-stage ignition fuels. Com-
bust Flame 154(3):387–409
50. Sj€
oberg M, Dec JE (2007) Comparing late-cycle autoignition stability for single-and
two-stage ignition fuels in HCCI engines. Proc Combust Inst 31(2):2895–2902
51. Westbrook CK, Pitz WJ, Curran HJ (2007) Auto-ignition and chemical kinetic mechanisms
of HCCI combustion. In: HCCI and CAI engines for the automotive industry. Woodhead
Publishing Limited, Cambridge, England pp 433–445
52. Zheng J, Yang W, Miller DL, Cernansky NP (2001) Prediction of pre-ignition reactivity and
ignition delay for HCCI using a reduced chemical kinetic model (No. 2001-01-1025). SAE
technical paper
53. Sj€
oberg M, Dec JE (2007) EGR and intake boost for managing HCCI low-temperature heat
release over wide ranges of engine speed (No. 2007-01-0051). SAE technical paper
54. Serinyel Z, Le Moyne L, Guibert P (2007) Homogeneous charge compression ignition as an
alternative combustion mode for the future of internal combustion engines. Int J Veh Des 44
(1–2):20–40
55. Zhao F, Asmus TW, Assanis DN, Dec JE, Eng JA Najt PM (2003) Homogeneous charge
compression ignition (HCCI) engines: key research and development issues. SAE PT-94
56. Silke EJ, Pitz WJ, Westbrook CK, Sj€oberg M, Dec JE (2008) Understanding the chemical
effects of increased boost pressure under HCCI conditions. SAE Int J Fuels Lubr 1:12–25.
(2008-01-0019)
57. Curran HJ, Gaffuri P, Pitz WJ, Westbrook CK (1998) A comprehensive modeling study of
n-heptane oxidation. Combust Flame 114(1):149–177
58. Luong MB, Yu GH, Lu T, Chung SH, Yoo CS (2015) Direct numerical simulations of
ignition of a lean n-heptane/air mixture with temperature and composition inhomogeneities
relevant to HCCI and SCCI combustion. Combust Flame 162(12):4566–4585
59. Coskun G, Jonsson M, Bood J, Tunér M, Algotsson M, Li B et al (2015) Analysis of
in-cylinder H2O2 and HO2 distributions in an HCCI engine–comparison of laser-diagnostic
results with CFD and SRM simulations. Combust Flame 162(9):3131–3139
60. Burns VR, Rapp LA, Koehl WJ, Benson JD, Hochhauser AM, Knepper JC, et al (1995)
Gasoline reformulation and vehicle technology effects on emissions-auto/oil air quality
improvement research program (No. 952509). SAE technical paper
61. Machrafi H, Cavadias S (2008) Three-stage autoignition of gasoline in an HCCI engine: an
experimental and chemical kinetic modeling investigation. Combust Flame 155(4):557–570
62. Vuilleumier D, Kozarac D, Mehl M, Saxena S, Pitz WJ, Dibble RW et al (2014) Intermediate
temperature heat release in an HCCI engine fueled by ethanol/n-heptane mixtures: an
experimental and modeling study. Combust Flame 161(3):680–695
63. Dec JE, Yang Y (2010) Boosted HCCI for high power without engine knock and with ultra-
low NOx emissions-using conventional gasoline. SAE Int J Engines 3:750–767. (2010-01-
1086)
64. Yang J (2007) Four-stroke gasoline HCCI engines with thermal management. In: Zhao H (ed)
HCCI and CAI engines for the automotive industry. Woodhead Publishing, Cambridge,
England ISBN:978-1-84569-128-8
References 125

65. Sun R, Thomas R, Gray CL (2004). An HCCI engine: power plant for a hybrid vehicle
(No. 2004-01-0933). SAE technical paper
66. Shahbakhti M (2009) Modeling and experimental study of an HCCI engine for combustion
timing control. PhD thesis, University of Alberta
67. Borland M, Zhao F (2002) Application of secondary air injection for simultaneously reducing
converter-in emissions and improving catalyst light-off performance (No. 2002-01-2803).
SAE technical paper
68. Bauer H, Haldenwanger HG, Hirth P, Brück R (1999) Thermal management of close coupled
catalysts (No. 1999-01-1231). SAE technical paper
69. Konstantinidis PA, Koltsakis GC, Stamatelos AM (1997) Computer aided assessment and
optimization of catalyst fast light-off techniques. Proc Inst Mech Eng D J Automob Eng 211
(1):21–37
70. Maurya RK, Agarwal AK (2014) Experimental investigations of performance, combustion
and emission characteristics of ethanol and methanol fueled HCCI engine. Fuel Process
Technol 126:30–48
71. Dec JE, Sj€oberg M (2004). Isolating the effects of fuel chemistry on combustion phasing in an
HCCI engine and the potential of fuel stratification for ignition control (No. 2004-01-0557).
SAE technical paper
72. Dec JE, Yang Y, Dronniou N (2011) Boosted HCCI-controlling pressure-rise rates for
performance improvements using partial fuel stratification with conventional gasoline. SAE
Int J Engines 4:1169–1189. (2011-01-0897)
73. Yang Y, Dec JE, Dronniou N, Sj€oberg M (2011) Tailoring HCCI heat-release rates with
partial fuel stratification: comparison of two-stage and single-stage-ignition fuels. Proc
Combust Inst 33(2):3047–3055
74. Yang Y, Dec JE, Dronniou N, Sj€oberg M, Cannella W (2011) Partial fuel stratification to
control HCCI heat release rates: fuel composition and other factors affecting pre-ignition
reactions of two-stage ignition fuels. SAE Int J Engines 4:1903–1920. (2011-01-1359)
75. Buda F, Bounaceur R, Warth V, Glaude PA, Fournet R, Battin-Leclerc F (2005) Progress
toward a unified detailed kinetic model for the autoignition of alkanes from C 4 to C
10 between 600 and 1200 K. Combust Flame 142(1):170–186
76. Ribaucour M, Minetti R, Sochet LR, Curran HJ, Pitz WJ, Westbrook CK (2000) Ignition of
isomers of pentane: an experimental and kinetic modeling study. Proc Combust Inst 28
(2):1671–1678
77. Aceves SM, Flowers DL, Dibble RW, Babajimopoulos A (2007) Overview of modeling
techniques and their application to HCCI/CAI engines. In: HCCI and CAI engines for the
automotive industry. pp 456–474
78. Sj€
oberg M, Dec JE (2006) Smoothing HCCI heat-release rates using partial fuel stratification
with two-stage ignition fuels (No. 2006-01-0629). SAE technical paper
79. Sjoberg M, Dec J (2010) Ethanol autoignition characteristics and HCCI performance for wide
ranges of engine speed, load and boost. SAE Int J Engines 3(1):84–106
80. Sj€
oberg M, Dec JE (2008) Influence of fuel autoignition reactivity on the high-load limits of
HCCI engines. SAE Int J Engines 1:39–58. (2008-01-0054)
81. Martinez-Frias J, Aceves SM, Flowers D, Smith JR, Dibble R (2000) HCCI engine control by
thermal management (No. 2000-01-2869). SAE technical paper
82. Chang J, Filipi Z, Assanis D, Kuo TW, Najt P, Rask R (2005) Characterizing the thermal
sensitivity of a gasoline homogeneous charge compression ignition engine with measure-
ments of instantaneous wall temperature and heat flux. Int J Engine Res 6(4):289–310
83. Hu T, Liu S, Zhou L, Li W (2006) Effects of compression ratio on performance, combustion,
and emission characteristics of an HCCI engine. Proc Inst Mech Eng D J Automob Eng 220
(5):637–645
84. Christensen M, Johansson B (2000) Supercharged homogeneous charge compression ignition
(HCCI) with exhaust gas recirculation and pilot fuel (No. 2000-01-1835). SAE technical
paper
126 2 Low Temperature Combustion Engines

85. Sj€oberg M, Dec JE (2003) Combined effects of fuel-type and engine speed on intake
temperature requirements and completeness of bulk-gas reactions for HCCI combustion
(No. 2003-01-3173). SAE technical paper
86. Maurya RK, Agarwal AK (2015) Combustion and emission characterization of n-butanol
fueled HCCI engine. J Energy Res Technol 137(1):011101
87. Zhao H, Peng Z, Williams J, Ladommatos N (2001) Understanding the effects of recycled
burnt gases on the controlled autoignition (CAI) combustion in four-stroke gasoline engines
(No. 2001-01-3607). SAE technical paper
88. Cairns A, Blaxill H (2005) The effects of combined internal and external exhaust gas
recirculation on gasoline controlled auto-ignition (No. 2005-01-0133). SAE technical paper
89. Sj€oberg M, Dec JE, Hwang W (2007) Thermodynamic and chemical effects of EGR and its
constituents on HCCI autoignition (No. 2007-01-0207). SAE technical paper
90. Cooper B, Jackson N, Penny I, Truscott T, Rawlins D, Seabrook J (2006) Advanced
development techniques for delivering low emissions diesel engines. Proc THIESEL
2006:267–280
91. Dubreuil A, Foucher F, Mounaι__m-Rousselle C (2006) Effect of EGR chemical components
and intake temperature on HCCI combustion development (No. 2006-32-0044). SAE tech-
nical paper
92. Machrafi H, Cavadias S, Guibert P (2008) An experimental and numerical investigation on
the influence of external gas recirculation on the HCCI autoignition process in an engine:
thermal, diluting, and chemical effects. Combust Flame 155(3):476–489
93. Lemel M, Hultqvist A, Vressner A, Nordgren H, Persson H, Johansson B (2005) Quantifica-
tion of the formaldehyde emissions from different HCCI engines running on a range of fuels
(No. 2005-01-3724). SAE technical paper
94. Hasan AO, Abu-jrai A, Ala’a H, Tsolakis A, Xu H (2016) Formaldehyde, acetaldehyde and
other aldehyde emissions from HCCI/SI gasoline engine equipped with prototype catalyst.
Fuel 175:249–256
95. Risberg P, Johansson D, Andrae J, Kalghatgi G, Bj€ornbom P, Ångstr€ om HE (2006) The
influence of NO on the combustion phasing in an HCCI engine (No. 2006-01-0416). SAE
technical paper
96. Machrafi H, Guibert P, Cavadias S (2008) HCCI engine modeling and experimental
investigations–part 2: the composition of a NO-PRF interaction mechanism and the influence
of NO in EGR on auto-ignition. Combust Sci Technol 180(7):1245–1262
97. Sang W (2014) Knock mitigation on boosted controlled auto-ignition engines with fuel
stratifications and exhaust gas recirculation. PhD thesis, Massachusetts Institute of
Technology
98. Sj€oberg M, Dec JE (2009) Influence of EGR quality and Unmixedness on the high-load limits
of HCCI engines. SAE Int J Engines 2:492–510. (2009-01-0666)
99. Sj€oberg M, Dec JE, Cernansky NP (2005). Potential of thermal stratification and combustion
retard for reducing pressure-rise rates in HCCI engines, based on multi-zone modeling and
experiments (No. 2005-01-0113). SAE technical paper
100. Sj€oberg M, Dec JE (2005) Effects of engine speed, fueling rate, and combustion phasing on
the thermal stratification required to limit HCCI knocking intensity (No. 2005-01-2125). SAE
technical paper
101. Dec JE, Hwang W (2009) Characterizing the development of thermal stratification in an
HCCI engine using planar-imaging thermometry. SAE Int J Engines 2:421–438. (2009-01-
0650)
102. Reuss DL, Sick V (2005) Inhomogeneities in HCCI combustion: an imaging study (No. 2005-
01-2122). SAE technical paper
103. Schmitt M, Boulouchos K (2016) Role of the intake generated thermal stratification on the
temperature distribution at top dead center of the compression stroke. Int J Engine Res 17
(8):836–845
References 127

104. Kakuho A, Nagamine M, Amenomori Y, Urushihara T, Itoh T (2006) In-cylinder temperature

distribution measurement and its application to HCCI combustion (No. 2006-01-1202). SAE
technical paper
105. Krasselt J, Foster DE, Ghandhi J, Herold R, Reuss D, Najt P (2009) Investigations into the
effects of thermal and compositional stratification on HCCI combustion–part I: metal engine
results (No. 2009-01-1105). SAE technical paper
106. Herold RE, Krasselt JM, Foster DE, Ghandhi JB, Reuss DL, Najt PM (2009) Investigations
into the effects of thermal and compositional stratification on HCCI combustion–part II:
optical engine results. SAE Int J Engines 2:1034–1053. (2009-01-1106)
107. Sj€oberg M, Dec JE, Babajimopoulos A, Assanis DN (2004) Comparing enhanced natural
thermal stratification against retarded combustion phasing for smoothing of HCCI heat-
release rates (No. 2004-01-2994). SAE technical paper
108. Yang DB, Wang Z, Wang JX, Shuai SJ (2011) Experimental study of fuel stratification for
HCCI high load extension. Appl Energy 88(9):2949–2954
109. Hwang W, Dec JE, Sj€oberg M (2007). Fuel stratification for low-load HCCI combustion:
performance & fuel-PLIF measurements (No. 2007-01-4130). SAE technical paper
110. Dec JE, Yang Y, Dernotte J, Ji C (2015) Effects of gasoline reactivity and ethanol content on
boosted, premixed and partially stratified low-temperature gasoline combustion (LTGC).
SAE Int J Engines 8:935–955. (2015-01-0813)
111. Lavoie GA, Martz J, Wooldridge M, Assanis D (2010) A multi-mode combustion diagram for
spark assisted compression ignition. Combust Flame 157(6):1106–1110
112. Koopmans L, Denbratt I (2001) A four stroke camless engine, operated in homogeneous
charge compression ignition mode with commercial gasoline (No. 2001-01-3610). SAE
technical paper
113. Reuss DL, Kuo TW, Silvas G, Natarajan V, Sick V (2008) Experimental metrics for
identifying origins of combustion variability during spark-assisted compression ignition.
Int J Engine Res 9(5):409–434
114. Urushihara T, Yamaguchi K, Yoshizawa K, Itoh T (2005) A study of a gasoline-fueled
compression ignition engine expansion of HCCI operation range using SI combustion as a
trigger of compression ignition (No. 2005-01-0180). SAE technical paper
115. Persson H, Hultqvist A, Johansson B, Remón A (2007) Investigation of the early flame
development in spark assisted HCCI combustion using high speed chemiluminescence
imaging (No. 2007-01-0212). SAE technical paper
116. Huang Y, Sung CJ, Eng JA (2004) Dilution limits of n-butane/air mixtures under conditions
relevant to HCCI combustion. Combust Flame 136(4):457–466
117. Persson H, Pfeiffer R, Hultqvist A, Johansson B, Str€ om H (2005) Cylinder-to-cylinder and
cycle-to-cycle variations at HCCI operation with trapped residuals (No. 2005-01-0130). SAE
technical paper
118. Middleton RJ, Olesky LKM, Lavoie GA, Wooldridge MS, Assanis DN, Martz JB (2015) The
effect of spark timing and negative valve overlap on spark assisted compression ignition
combustion heat release rate. Proc Combust Inst 35(3):3117–3124
119. Olesky LM, Martz JB, Lavoie GA, Vavra J, Assanis DN, Babajimopoulos A (2013) The
effects of spark timing, unburned gas temperature, and negative valve overlap on the rates of
stoichiometric spark assisted compression ignition combustion. Appl Energy 105:407–417
120. Olesky LM, Lavoie GA, Assanis DN, Wooldridge MS, Martz JB (2014) The effects of diluent
composition on the rates of HCCI and spark assisted compression ignition combustion. Appl
Energy 124:186–198
121. Zigler BT, Keros PE, Helleberg KB, Fatouraie M, Assanis D, Wooldridge MS (2011) An
experimental investigation of the sensitivity of the ignition and combustion properties of a
single-cylinder research engine to spark-assisted HCCI. Int J Engine Res 12(4):353–375
122. Manofsky L, Vavra J, Assanis DN, Babajimopoulos A (2011) Bridging the gap between
HCCI and SI: spark-assisted compression ignition (No. 2011-01-1179). SAE technical paper
128 2 Low Temperature Combustion Engines

123. Martz JB, Middleton RJ, Lavoie GA, Babajimopoulos A, Assanis DN (2011) A computa-
tional study and correlation of premixed isooctane–air laminar reaction front properties under
spark ignited and spark assisted compression ignition engine conditions. Combust Flame 158
(6):1089–1096
124. Quader AA (1976) What limits lean operation in spark ignition engines-flame initiation or
propagation? (No. 760760). SAE technical paper
125. Middleton RJ, Martz JB, Lavoie GA, Babajimopoulos A, Assanis DN (2012) A computa-
tional study and correlation of premixed isooctane air laminar reaction fronts diluted with
EGR. Combust Flame 159(10):3146–3157
126. Yun H, Wermuth N, Najt P (2011) High load HCCI operation using different valving
strategies in a naturally-aspirated gasoline HCCI engine. SAE Int J Engines 4:1190–1201.
(2011-01-0899)
127. Persson H (2008) Spark assisted compression ignition, SACI. PhD thesis, Lund University,
SE-22100 Lund. ISBN 978-91-628-7578-7
128. Wang Z, Wang JX, Shuai SJ, Tian GH, An X, Ma QJ (2006) Study of the effect of spark
ignition on gasoline HCCI combustion. Proc Inst Mech Eng D J Automob Eng 220
(6):817–825
129. Hyv€ onen J, Haraldsson G, Johansson B (2005) Operating conditions using spark assisted
HCCI combustion during combustion mode transfer to SI in a multi-cylinder VCR-HCCI
engine (No. 2005-01-0109). SAE technical paper
130. Lawler B, Mamalis S, Joshi S, Lacey J, Guralp O, Najt P, Filipi Z (2017) Understanding the
effect of operating conditions on thermal stratification and heat release in a homogeneous
charge compression ignition engine. Appl Therm Eng 112:392–402
131. Lawler B, Splitter D, Szybist J, Kaul B (2017) Thermally stratified compression ignition: a
new advanced low temperature combustion mode with load flexibility. Appl Energy
189:122–132
132. Christensen M, Johansson B (1999) Homogeneous charge compression ignition with water
injection (No. 1999-01-0182). SAE technical paper
133. Ogawa H, Miyamoto N, Kaneko N, Ando H (2003) Combustion control and operating range
expansion with direct injection of reaction suppressors in a premixed DME HCCI engine
(No. 2003-01-0746). SAE technical paper
134. Hyv€ onen J, Wilhelmsson C, Johansson B (2006) The effect of displacement on air-diluted
multi-cylinder HCCI engine performance (No. 2006-01-0205). SAE technical paper
135. Manente V (2010) Gasoline partially premixed combustion: an advanced internal combustion
engine concept aimed to high efficiency, low emissions and low acoustic noise in the whole
load range. PhD thesis, Lund University, SE 22100 Lund Sweden. ISBN 978-91-628-8144-3
136. Hardy WL, Reitz RD (2006) A study of the effects of high EGR, high equivalence ratio, and
mixing time on emissions levels in a heavy-duty diesel engine for PCCI combustion
(No. 2006-01-0026). SAE technical paper
137. Kimura S, Aoki O, Kitahara Y, Aiyoshizawa E (2001) Ultra-clean combustion technology
combining a low-temperature and premixed combustion concept for meeting future emission
standards (No. 2001-01-0200). SAE technical paper
138. Wimmer A, Eichlseder H, Klell M, Figer G (2006) Potential of HCCI concepts for DI diesel
engines. Int J Veh Des 41(1–4):32–48
139. Takeda Y, Keiichi N, Keiichi N (1996) Emission characteristics of premixed lean diesel
combustion with extremely early staged fuel injection (No. 961163). SAE technical paper
140. Akagawa H, Miyamoto T, Harada A, Sasaki S, Shimazaki N, Hashizume T, Tsujimura K
(1999) Approaches to solve problems of the premixed lean diesel combustion (No. 1999-01-
0183). SAE technical paper
141. Hashizume T, Miyamoto T, Hisashi A, Tsujimura K (1998) Combustion and emission
characteristics of multiple stage diesel combustion (No. 980505). SAE technical paper
142. Iwabuchi Y, Kawai K, Shoji T, Takeda Y (1999) Trial of new concept diesel combustion
system-premixed compression-ignited combustion (No. 1999-01-0185). SAE technical paper
References 129

143. Yanagihara H, Sato Y, Mizuta JI (1997) A study of DI diesel combustion under uniform
higher-dispersed mixture formation. JSAE Rev 18(4):361–367
144. Walter B, Gatellier B (2002) Development of the high power NADI™ concept using dual
mode diesel combustion to achieve zero NOx and particulate emissions (No. 2002-01-1744).
SAE technical paper
145. Singh S, Reitz RD, Musculus MPB, Lachaux T (2007) Validation of engine combustion
models against detailed in-cylinder optical diagnostics data for a heavy-duty compression-
ignition engine. Int J Engine Res 8(1):97–126
146. Genzale CL, Reitz RD, Musculus MP (2008) Effects of piston bowl geometry on mixture
development and late-injection low-temperature combustion in a heavy-duty diesel engine.
SAE Int J Engines 1:913–937. (2008-01-1330)
147. Noehre C, Andersson M, Johansson B, Hultqvist A (2006) Characterization of partially
premixed combustion (No. 2006-01-3412). SAE technical paper
148. Johansson B (2005) High-load partially premixed combustion in a heavy-duty diesel engine.
In: Diesel engine emissions reduction (DEER) conference. Chicago, 21–25, 2005
149. Sang W, Cheng WK, Maria A (2014) The nature of heat release in gasoline PPCI engines
(No. 2014-01-1295). SAE technical paper
150. Rose KD, Ariztegui J, Cracknell RF, Dubois T, Hamje HDC, Pellegrini L et al (2013)
Exploring a gasoline compression ignition (GCI) engine concept (No. 2013-01-0911). SAE
technical paper
151. Goyal H, Kook S, Hawkes E, Chan QN, Padala S, Ikeda Y. (2017) Influence of engine speed
on gasoline compression ignition (GCI) combustion in a single-cylinder light-duty diesel
engine (No. 2017-01-0742). SAE technical paper
152. Sellnau M, Moore W, Sinnamon J, Hoyer K, Foster M, Husted H (2015) GDCI multi-cylinder
engine for high fuel efficiency and low emissions. SAE Int J Engines 8:775–790. (2015-01-
0834)
153. Kalghatgi GT (2005) Auto-ignition quality of practical fuels and implications for fuel
requirements of future SI and HCCI engines (No. 2005-01-0239). SAE technical paper
oberg M, Edling LO, Eliassen T, Magnusson L, Ångstr€
154. Sj€ om HE (2002) GDI HCCI: effects of
injection timing and air swirl on fuel stratification, combustion and emissions formation
(No. 2002-01-0106). SAE technical paper
155. Aroonsrisopon T, Werner P, Waldman JO, Sohm V, Foster DE, Morikawa T, Iida M (2004)
Expanding the HCCI operation with the charge stratification (No. 2004-01-1756). SAE
technical paper
156. Dempsey AB, Curran S, Wagner R, Cannella W (2015) Effect of premixed fuel preparation
for partially premixed combustion with a low octane gasoline on a light-duty multicylinder
compression ignition engine. J Eng Gas Turbines Power 137(11):111506
157. Loeper P, Ra Y, Adams C, Foster DE, Ghandhi J, Andrie M et al (2013) Experimental
investigation of light-medium load operating sensitivity in a gasoline compression ignition
(GCI) light-duty diesel engine (No. 2013-01-0896). SAE technical paper
158. Ra Y, Loeper P, Andrie M, Krieger R, Foster DE, Reitz RD, Durrett R (2012) Gasoline DICI
engine operation in the LTC regime using triple-pulse injection. SAE Int J Engines
5:1109–1132. (2012-01-1131)
159. Sellnau MC, Sinnamon J, Hoyer K, Kim J, Cavotta M, Husted H (2013) Part-load operation
of gasoline direct-injection compression ignition (GDCI) engine (No. 2013-01-0272). SAE
technical paper
160. Sellnau MC, Sinnamon J, Hoyer K, Husted H (2012) Full-time gasoline direct-injection
compression ignition (GDCI) for high efficiency and low NOx and PM. SAE Int J Engines
5:300–314. (2012-01-0384)
161. Sellnau M, Foster M, Hoyer K, Moore W, Sinnamon J, Husted H (2014) Development of a
gasoline direct injection compression ignition (GDCI) engine. SAE Int J Engines 7:835–851.
(2014-01-1300)
130 2 Low Temperature Combustion Engines

162. Tanov S, Collin R, Johansson B, Tuner M (2014) Combustion stratification with partially
premixed combustion, PPC, using NVO and split injection in a LD-diesel engine. SAE Int J
Engines 7:1911–1919. (2014-01-2677)
163. Dempsey AB, Reitz RD (2011) Computational optimization of a heavy-duty compression
ignition engine fueled with conventional gasoline. SAE Int J Engines 4:338–359. (2011-01-
0356)
164. Kalghatgi GT, Risberg P, Ångstr€om HE (2006) Advantages of fuels with high resistance to
auto-ignition in late-injection, low-temperature, compression ignition combustion (No. 2006-
01-3385). SAE technical paper
165. Manente V, Johansson B, Tunestal P (2009) Partially premixed combustion at high load using
gasoline and ethanol, a comparison with diesel (No. 2009-01-0944). SAE technical paper
166. Manente V, Zander CG, Johansson B, Tunestal P, Cannella W (2010) An advanced internal
combustion engine concept for low emissions and high efficiency from idle to max load using
gasoline partially premixed combustion (No. 2010-01-2198). SAE technical paper
167. Chang J, Kalghatgi G, Amer A, Viollet Y (2012) Enabling high efficiency direct injection
engine with naphtha fuel through partially premixed charge compression ignition combustion
(No. 2012-01-0677). SAE technical paper
168. Adhikary BD, Reitz RD, Ciatti S, Kolodziej C (2014) Computational investigation of low
load operation in a light-duty gasoline direct injection compression ignition [GDICI] engine
using single-injection strategy (No. 2014-01-1297). SAE technical paper
169. Kolodziej CP, Ciatti S, Vuilleumier D, Adhikary BD, Reitz RD (2014) Extension of the lower
load limit of gasoline compression ignition with 87 AKI gasoline by injection timing and
pressure (No. 2014-01-1302). SAE technical paper
170. Kalghatgi GT, Hildingsson L, Harrison AJ, Johansson B (2011) Autoignition quality of
gasoline fuels in partially premixed combustion in diesel engines. Proc Combust Inst 33
(2):3015–3021
171. Manente V, Johansson B, Tunestal P, Cannella W (2009) Effects of different type of gasoline
fuels on heavy duty partially premixed combustion. SAE Int J Engines 2:71–88. (2009-01-
2668)
172. Manente V, Tunestal P, Johansson B, Cannella WJ (2010) Effects of ethanol and different
type of gasoline fuels on partially premixed combustion from low to high load (No. 2010-01-
0871). SAE technical paper
173. Chen Y, Wolk B, Mehl M, Cheng WK, Chen JY, Dibble RW (2017) Development of a
reduced chemical mechanism targeted for a 5-component gasoline surrogate: a case study on
the heat release nature in a GCI engine. Combust Flame 178:268–276
174. Solaka H, Tuner M, Johansson B (2013) Analysis of surrogate fuels effect on ignition delay
and low temperature reaction during partially premixed combustion (No. 2013-01-0903).
SAE technical paper
175. Kokjohn SL (2012) Reactivity controlled compression ignition (RCCI) combustion. PhD
thesis, University of Wisconsin-Madison
176. Bessonette PW, Schleyer CH, Duffy KP, Hardy WL, Liechty MP. (2007) Effects of fuel
property changes on heavy-duty HCCI combustion (No. 2007-01-0191). SAE technical paper
177. Inagaki K, Fuyuto T, Nishikawa K, Nakakita K, Sakata I (2006) Dual-fuel PCI combustion
controlled by in-cylinder stratification of ignitability (No. 2006-01-0028). SAE technical
paper
178. Zeldovich YB (1980) Regime classification of an exothermic reaction with nonuniform initial
conditions. Combust Flame 39(2):211–214
179. Chen JH, Hawkes ER, Sankaran R, Mason SD, Im HG (2006) Direct numerical simulation of
ignition front propagation in a constant volume with temperature inhomogeneities:
I. Fundamental analysis and diagnostics. Combust Flame 145(1):128–144
References 131

180. Reitz RD, Duraisamy G (2015) Review of high efficiency and clean reactivity controlled
compression ignition (RCCI) combustion in internal combustion engines. Prog Energy
Combust Sci 46:12–71
181. Li J, Yang W, Zhou D (2017) Review on the management of RCCI engines. Renew Sust
Energ Rev 69:65–79
182. Eichmeier JU, Reitz RD, Rutland C (2014) A zero-dimensional phenomenological model for
RCCI combustion using reaction kinetics. SAE Int J Engines 7:106–119. (2014-01-1074)
183. Kokjohn S, Reitz RD, Splitter D, Musculus M (2012) Investigation of fuel reactivity
stratification for controlling PCI heat-release rates using high-speed chemiluminescence
imaging and fuel tracer fluorescence. SAE Int J Engines 5:248–269. (2012-01-0375)
184. DelVescovo DA (2016) The effects of fuel stratification and heat release rate shaping in
reactivity controlled compression ignition (RCCI) combustion. PhD thesis, University of
Wisconsin-Madison
185. Kokjohn SL, Musculus MP, Reitz RD (2015) Evaluating temperature and fuel stratification
for heat-release rate control in a reactivity-controlled compression-ignition engine using
optical diagnostics and chemical kinetics modeling. Combust Flame 162(6):2729–2742
186. Kokjohn SL, Reitz RD (2010) Characterization of dual-fuel PCCI combustion in a light-duty
engine. In: 20th international multidimensional engine modeling user’s group meeting,
Detroit, MI, April 12, 2010
187. Paykani A, Kakaee AH, Rahnama P, Reitz RD (2016) Progress and recent trends in
reactivity-controlled compression ignition engines. Int J Engine Res 17(5):481–524
188. Ladommatos N, Parsi M, Knowles A (1996) The effect of fuel cetane improver on diesel
pollutant emissions. Fuel 75(1):8–14
189. Hosseini V, Neill WS, Guo H, Chippior WL, Fairbridge C, Mitchell K (2011) Effects of
different cetane number enhancement strategies on HCCI combustion and emissions. Int J
Engine Res 12(2):89–108
190. Nieman DE, Dempsey AB, Reitz RD (2012) Heavy-duty RCCI operation using natural gas
and diesel. SAE Int J Engines 5:270–285. (2012-01-0379)
191. Wang H, DelVescovo D, Yao M, Reitz RD (2015) Numerical study of RCCI and HCCI
combustion processes using gasoline, diesel, iso-butanol and DTBP cetane improver. SAE Int
J Engines 8:831–845. (2015-01-0850)
192. Dempsey AB, Adhikary BD, Viswanathan S, Reitz RD (2012) Reactivity controlled com-
pression ignition using premixed hydrated ethanol and direct injection diesel. J Eng Gas
Turbines Power 134(8):082806
193. Dempsey AB, Walker NR, Reitz RD (2013) Effect of piston bowl geometry on dual fuel
reactivity controlled compression ignition (RCCI) in a light-duty engine operated with
gasoline/diesel and methanol/diesel. SAE Int J Engines 6:78–100. (2013-01-0264)
194. Li Y, Jia M, Liu Y, Xie M (2013) Numerical study on the combustion and emission
characteristics of a methanol/diesel reactivity controlled compression ignition (RCCI)
engine. Appl Energy 106:184–197
195. Benajes J, Molina S, Garcı́a A, Monsalve-Serrano J (2015) Effects of direct injection timing
and blending ratio on RCCI combustion with different low reactivity fuels. Energy Convers
Manag 99:193–209
196. Hanson R, Curran S, Wagner R, Reitz R (2013) D.. Effects of biofuel blends on RCCI
combustion in a light-duty, multi-cylinder diesel engine. SAE Int J Engines 6:488–503.
(2013-01-1653)
197. Gharehghani A, Hosseini R, Mirsalim M, Jazayeri SA, Yusaf T (2015) An experimental study
on reactivity controlled compression ignition engine fueled with biodiesel/natural gas.
Energy 89:558–567
198. Splitter D, Reitz RD, Hanson R (2010) High efficiency, low emissions RCCI combustion by
use of a fuel additive. SAE Int J Fuels Lubr 3:742–756. (2010-01-2167)
132 2 Low Temperature Combustion Engines

199. Hanson R, Kokjohn S, Splitter D, Reitz R (2011) D.. Fuel effects on reactivity controlled
compression ignition (RCCI) combustion at low load. SAE Int J Engines 4:394–411. (2011-
01-0361)
200. Dempsey AB, Walker NR, Reitz RD (2013) Effect of cetane improvers on gasoline, ethanol,
and methanol reactivity and the implications for RCCI combustion. SAE Int J Fuels Lubr
6:170–187. (2013-01-1678)
201. Ma S, Zheng Z, Liu H, Zhang Q, Yao M (2013) Experimental investigation of the effects of
diesel injection strategy on gasoline/diesel dual-fuel combustion. Appl Energy 109:202–212
202. Benajes J, Garcı́a A, Monsalve-Serrano J, Boronat V (2017) Achieving clean and efficient
engine operation up to full load by combining optimized RCCI and dual-fuel diesel-gasoline
combustion strategies. Energy Convers Manag 136:142–151
203. Walker NR, Dempsey AB, Andrie MJ, Reitz RD (2013) Use of low-pressure direct-injection
for reactivity controlled compression ignition (RCCI) light-duty engine operation. SAE Int J
Engines 6:1222–1237. (2013-01-1605)
204. Splitter D, Wissink M, Kokjohn S, Reitz RD (2012) Effect of compression ratio and piston
geometry on RCCI load limits and efficiency (No. 2012-01-0383). SAE technical paper
205. Li J, Yang WM, Zhou DZ (2016) Modeling study on the effect of piston bowl geometries in a
gasoline/biodiesel fueled RCCI engine at high speed. Energy Convers Manag 112:359–368
206. Hanson R, Curran S, Wagner R, Kokjohn S, Splitter D, Reitz RD (2012) Piston bowl
optimization for RCCI combustion in a light-duty multi-cylinder engine. SAE Int J Engines
5:286–299. (2012-01-0380)
207. Wu Y, Reitz RD (2015) Effects of exhaust gas recirculation and boost pressure on reactivity
controlled compression ignition engine at high load operating conditions. J Energy Resour
Technol 137(3):032210
208. Yu C, Wang J, Yu W, Liu J, Gao D (2013) Research on low temperature combustion of
homogeneous charge induced ignition (HCII) in a light-duty diesel engine. In: Proceedings of
the FISITA 2012 world automotive congress, Springer Berlin Heidelberg, pp 195–204
209. Li Y, Jia M, Chang Y, Fan W, Xie M, Wang T (2015) Evaluation of the necessity of exhaust
gas recirculation employment for a methanol/diesel reactivity controlled compression igni-
tion engine operated at medium loads. Energy Convers Manag 101:40–51
210. Wissink M, Reitz R (2016) Exploring the role of reactivity gradients in direct dual fuel
stratification. SAE Int J Engines 9:1036–1048. (2016-01-0774)
211. Wissink M, Reitz RD (2015) Direct dual fuel stratification, a path to combine the benefits of
RCCI and PPC. SAE Int J Engines 8:878–889. (2015-01-0856)
212. Splitter D, Hanson R, Kokjohn S, Reitz RD (2011) Reactivity controlled compression
ignition (RCCI) heavy-duty engine operation at mid-and high-loads with conventional and
alternative fuels (No. 2011-01-0363). SAE technical paper
213. Dempsey AB, Reitz RD (2011) Computational optimization of reactivity controlled com-
pression ignition in a heavy-duty engine with ultra low compression ratio. SAE Int J Engines
4(2):2222–2239
214. Kavuri C, Paz J, Kokjohn SL (2016) A comparison of reactivity controlled compression
ignition (RCCI) and gasoline compression ignition (GCI) strategies at high load, low speed
conditions. Energy Convers Manag 127:324–341
215. Kokjohn SL, Hanson RM, Splitter DA, Reitz RD (2011) Fuel reactivity controlled compres-
sion ignition (RCCI): a pathway to controlled high-efficiency clean combustion. Int J Engine
Res 12(3):209–226
216. Hendricks TL, Splitter DA, Ghandhi JB (2014) Experimental investigation of piston heat
transfer under conventional diesel and reactivity-controlled compression ignition combustion
regimes. Int J Engine Res 15(6):684–705
217. Li Y, Jia M, Chang Y, Kokjohn SL, Reitz RD (2016) Thermodynamic energy and exergy
analysis of three different engine combustion regimes. Appl Energy 180:849–858
References 133

218. Dempsey AB, Walker NR, Gingrich E, Reitz RD (2014) Comparison of low temperature
combustion strategies for advanced compression ignition engines with a focus on controlla-
bility. Combust Sci Technol 186(2):210–241
219. Lin Y, Zhang HH (2003) Component selection and smoothing in smoothing spline analysis of
variance models. Institute of Statistics Mimeo Series 2556, NUCS
220. Kavuri C, Tiry M, Paz J, Kokjohn SL (2016) Experimental and computational investigation
of soot production from a premixed compression ignition engine using a load extension
injection. Int J Engine Res. https://doi.org/10.1177/1468087416650073
221. Klos D, Kokjohn SL (2015) Investigation of the sources of combustion instability in
low-temperature combustion engines using response surface models. Int J Engine Res 16
(3):419–440