This article was downloaded by: [Universiti Teknologi Malaysia]On: 11 January 2013, At: 20:55Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK
Combustion Science and TechnologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/gcst20
Predicting Fuel Performance for Future HCCI EnginesVi H. Rapp a , William J. Cannella b , J.-Y. Chen a & Robert W. Dibble aa Department of Mechanical Engineering, University of California–Berkeley, Berkeley,California, USAb Chevron Energy Technology Company, Richmond, California, USAAccepted author version posted online: 04 Dec 2012.
To cite this article: Vi H. Rapp , William J. Cannella , J.-Y. Chen & Robert W. Dibble (2012): Predicting Fuel Performance forFuture HCCI Engines, Combustion Science and Technology, DOI:10.1080/00102202.2012.750309
To link to this article: http://dx.doi.org/10.1080/00102202.2012.750309
Disclaimer: This is a version of an unedited manuscript that has been accepted for publication. As a serviceto authors and researchers we are providing this version of the accepted manuscript (AM). Copyediting,typesetting, and review of the resulting proof will be undertaken on this manuscript before final publication ofthe Version of Record (VoR). During production and pre-press, errors may be discovered which could affect thecontent, and all legal disclaimers that apply to the journal relate to this version also.
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1
Predicting Fuel Performance for Future HCCI Engines
Vi H. Rapp1,*, William J. Cannella2, J.-Y. Chen1, Robert W. Dibble1
1Department of Mechanical Engineering, University of California–Berkeley, Berkeley, California, USA, 2Chevron Energy Technology Company, Richmond, California, USA
*Corresponding author: E-mail: [email protected]
Abstract
The purpose of this research is to investigate the impact of fuel composition on auto-
ignition in HCCI engines in order to develop a future metric for predicting fuel
performance in future HCCI engine technology. A single-cylinder, variable compression
ratio engine operating as an HCCI engine was used to test reference fuels and gasoline
blends with Octane numbers (ON) ranging from 60-88. Correlations between fuel
composition, ON, and two existing methods for predicting fuel auto-ignition in HCCI
engines (Kalghatgi’s Octane Index and Shibata and Urushihara’s HCCI Index) are
investigated. Results show that Octane Index and HCCI Index poorly predict the impact
of fuel composition on auto-ignition for fuels with the same ON. The effect of ethanol in
delaying auto-ignition depends on the composition of the original gasoline blend; the
same is true for the addition of naphthenes. Low temperature heat release (LTHR)
correlates well with auto-ignition for gasoline fuels exhibiting LTHR.
KEYWORDS: Auto-ignition, Homogenous charge compression-ignition (HCCI), Fuel
composition
INTRODUCTION
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 2
Increasing concern with climate change has encouraged the development of alternative
fuels and advanced engine technologies that improve efficiency and reduce CO2
emissions. Homogeneous charge compression ignition (HCCI) engines offer the potential
for Diesel-like efficiencies and low nitrogen oxide emissions compared with conventional
gasoline and Diesel engines. HCCI engines also offer fuel flexibility as they can operate
using a wide variety of fuels such as Diesel, gasoline, and alternative fuels (Thring, 1989;
Fuhs, 2008). In the early twentieth century, Weiss and Mietz developed the first HCCI-
like combustion engine, called the hot-bulb engine (Erlandsson, 2002). The hot-bulb
engine offered a simple and durable design that had brake thermal efficiencies
comparable to contemporary Diesel engines. Later, in 1979, Onishi et al. (1979)
published the first research on a gasoline-fueled HCCI engine. The two-stroke gasoline
engine, using a process dubbed Active Thermo-Atmosphere Combustion (ATAC) by the
authors, increased fuel economy and decreased exhaust emissions at part-throttle
operation.
In 1983, Najt and Foster (1983) achieved compression ignition homogenous charge
(CIHC) combustion in a four-stroke gasoline engine. Using the same engine as Najt and
Foster, Thring (1989) studied the effects of exhaust gas recirculation, intake temperature,
and compression ratio; he was also the first to use the acronym HCCI. Seven years after
Thring, the first research burning Diesel fuel in an HCCI engine appeared (Gray and
Ryan, 1997) and led to research testing other fuels, such as alcohols (Oakley et al., 2001),
hydrogen (Shudo and Ono, 2002), natural gas (Christensen, Johansson and Einewall,
1997; Hiltner et al., 2000; Olsson et al., 2002; Stanglmaier, Ryan and Souder, 2001),
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 3
propane(Au et al., 2001; Flowers et al., 2001), and many fuel blends with additives (Eng,
Leppard and Sloane, 2003; Yao, Zheng and Liu, 2009)
Although HCCI engines offer fuel flexibility and a solution for meeting new, strict
pollution requirements, HCCI engines have a limited load range and cannot support high
load demands required by automobiles. Hybrid HCCI engines, such as SI-HCCI (Zhang,
Xie and Zhao, 2009; Koopmans et al., 2003), HCCI-DI (Canova et al., 2007; Helmantel
and Denbratt, 2004), or HCCI-electric(Wu and Zhang, 2012), offer a solution for
reducing emissions and increasing the load operating range. As HCCI engine technology
becomes more widely used in automotive technology, developing fuels to support hybrid
HCCI engines will become increasingly important.
Conventional methods for quantifying fuel auto-ignition, such as Research Octane
Number (RON), Motor Octane Number (MON), Octane Number (ON = ½RON +
½MON) and Cetane Number, poorly predict auto-ignition in HCCI engines (Kalghatgi,
2005; Shibata and Urushihara, 2007). Kalghatgi (2005) developed an Octane Index (OI)
for measuring the auto-ignition or anti-knock quality of a practical fuel at different
operating conditions. The OI, not to be confused with ON, is defined as,
OI = (1 K)RON +(K)MON, (1)
where K is a parameter specified by engine operating conditions. Although the OI may
be applicable for HCCI operation (Kalghatgi, 2005), the OI does not fully describe the
impact of fuel composition on auto-ignition in HCCI engines.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 4
Shibata and Urushihara (2007) investigated the impact of fuel composition on auto-
ignition in HCCI engines and introduced three HCCI Indices. While all three HCCI
Indices predict auto-ignition similarly, the relative HCCI Index (HIrel) predicts auto-
ignition of fuels using the fuel composition and MON (Shibata and Urushihara, 2007).
The HIrel is defined as,
relHI = MON +α(nP) +β(iP)(O) δ(A) + (OX),
(2)
where nP is the percent n-paraffins by volume, iP is the percentiso-paraffins by volume,
O is the percent olefins and cycloalkanes by volume, A is the percent aromatics by
volume, OX is the percent oxygenates by volume, and α, β, γ, δ, and ε are temperature
dependent parameters.
In this paper, the capability of two existing methods (the OI and the HIrel) for predicting
the impact of fuel composition on auto-ignition in HCCI engines is investigated and
correlations between fuel composition and auto-ignition of fuels in a HCCI engine are
explored. Fuels tested consist of following five different blends:
1. Primary reference fuels (PRF):blends of isooctane and n-heptane
2. Toluene reference fuels (TRF): PRF fuels blended with toluene
3. Ethanol reference fuels(E-PRF): PRF fuels blended with Ethanol
4. Gasoline blendstocks
5. Gasoline blendstocks with different pure compounds added (“additives”)
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 5
Following this Introduction, the instrumentation and experimental design are described.
Next, results and discussions are presented. Last, future work is suggested and
conclusions are drawn.
MATERIALS AND METHODS
Engine and Fuel Specifications
Similar to previous metrics for predicting fuel auto-ignition, such as RON and MON
(ASTM RON Standard, 2011; ASTM MON Standard, 2011), experiments were
conducted using a variable compression ratio, single cylinder cooperative fuel research
(CFR) engine operating in HCCI mode. Engine specifications and operating conditions
are listed in Table 1. The engine was preheated by operating in spark-ignition mode
under stoichiometric conditions. Once the coolant temperature reached 80°C, the
equivalence ratio was decreased to �=0.33 (�=3.0). Next, the compression ratio (CR)
was slowly increased until stable auto-ignition (no misfiring) occurred. The lowest CR
limit was determined by decreasing the CR until HCCI operation became unstable. The
highest CR limit for each fuel was determined by increasing the CR until the in-cylinder
pressure exceeded 50 bar (a limit to safe guard the mechanical integrity of CFR engine)
or the ringing intensity became too great (Eng, 2002). For each experiment, equivalence
ratio was held constant at φ=0.33 (λ=3.0). Data were taken at various compression ratios
between the lowest and the highest limits. For a fixed CR, 300 thermodynamic cycles
(each cycle with 720 CAD) of in-cylinder pressure data were collected along with
exhaust emissions before the catalytic converter.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 6
The effects of fuel composition on auto-ignition timing n HCCI engines, measured by
CA50 (the crank angle degree at which 50% of the cumulative heat has been
released),was explored by testing twenty different fuel blends at an engine intake
temperature of 150°C and one fuel, PRF70, at intake temperatures of 70°C, 115°C, and
150°C. Of the twenty fuels, eight fuels were reference fuel blends. The reference fuel
blends consisted of PRF60, PRF70, PRF75, PRF85, PRF88, TRF70, S70, and E-PRF70.
For primary reference fuels (PRF), the number following "PRF" is the RON, MON, and
percent isooctane by volume. TRF70 (46% n-heptane and 54% toluene, by volume) has a
calculated RON of 70.5and MON of 63.5, which were calculated using a linear-by-
volume blending equation created by Morgan et al.(2010). For S70 (64% isooctane, 31%
n-heptane, and 5% toluene by volume) the RON and MON were measured, by Chevron,
as 70.5and 69.6, respectively. A RON of 69.5 and MON of 68.6 were calculated for E-
PRF70 (64% isooctane, 31% n-heptane, and 5% ethanol by volume) using the blending
RON and blending MON values from Anderson et al.(2010). A summary of the reference
fuel blend compositions, RON, and MON are given in Table 2.
Two different base gasolines, typically used in U.S. gasoline blends, were provided by
Chevron, labeled G1 and G2. Hydrocarbon class information and the RON and MON of
the base gasolines are provided in Table 3. The base gasoline fuels were blended with
different “additives”: n-heptane, ethanol, cycloparaffins, and aromatics. RON and MON
for the gasoline fuel blends are listed in Table 4along with the type of additive. Fuels
with a calculated RON and MON were determined using the blending RON and blending
MON of the additive.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 7
Determining Octane Index, Relative HCCI Index, And Heat Release
The K factor for the OI, shown in Eq. (1), is a function of the in-cylinder temperature
when the in-cylinder pressure, during the compression stroke, reaches 15 bar (Tcomp,15bar)
(Kalghatgi, 2005). The K factor is computed using the following equations:
comp,15bar comp,15barK = 0.0426(T ) 35.2 if T 825 ,K (3)
or
comp,15bar comp,15bar0.0056(T ) 4.68 if T 825 .K (4)
For the UC Berkeley CFR engine operating at 600 RPM with an intake temperature of
150°C, Tcomp,15bar is estimated to be 780K, yielding K=-0.312 using Eq. (4) (Kalghatgi,
2005).
Because the OI was used to develop the HIrel, the temperature dependent constants for the
HIrel, shown in Eq.(2), are also functions of Tcomp,15bar. For Tcomp,15bar = 780K, Shibata and
Urushihara list values for the constants as follows (Shibata and Urushihara, 2007):α = -
0.487, β= -0.380, γ= -0.246, δ= -0.222, and ε =0.049. These constants were determined
using a similar method as the K factor in the Octane Index.
In addition to developing the relative HCCI Index, Shibata et al. (2005) suggested that
low temperature heat release (LTHR) might correlate with auto-ignition better than high
temperature heat release (HTHR). LTHR is defined as total heat release (Killingsworth,
2007) d from combustion at in-cylinder temperatures less than 1000K while HTHR is
total heat released from combustion at in-cylinder temperatures greater than 1000K. In
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 8
their study, correlations between auto-ignition and LTHR were investigated by dividing
LTHR by HTHR for each fuel, yielding a heat release ratio. In this paper, the net heat
release per crank angle degree (dQ/dθ) was determined using the first law of
thermodynamics (Heywood, 1988; Stone, 1999),
dQ γ dV 1 dp= p + Vdθ γ 1 dθ γ 1 dθ
(5)
where γ is the specific heat ratio, p is pressure, and V is volume. To avoid numerically
differentiating the discrete pressure measurements and amplifying signal noise, Eq. (6)
was be rewritten as,
dQ 1 d(pV) dV= +pdθ γ 1 dθ dθ'
(6)
and the cumulative net heat release, Qi, was computed as a finite sum instead of a
continuous integral using (Killingsworth, 2007),
i i i 0 0 j0
1Q [p V p V ] p ( ) ,1
i
j
V j
(7)
where i and j imply a discrete measurement of pressure and volume at a given crank
angle degree. The specific heat ratio, γ, was assumed to be constant during compression
and expansion. A linear fit between the compression γ and expansionγ was used to
calculate dQ/dθ during combustion. Figure 1 shows the inflection points in the heat
release rate that were used to distinguish between LTHR and HTHR. We assumed that
LTHR began when the heat release rate was greater than zero and that HTHR ended
when the heat release rate dropped below zero.
Measurement Instrumentation
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 9
In-cylinder pressure was measured using a 6052B Kistler piezoelectric pressure
transducer in conjunction with a 5044A Kistler charge amplifier and was recorded every
0.1 crank angle (CA) degree. The cylinder pressure transducer was mounted in the
cylinder head. Intake pressure was measured using a 4045A5 Kistler piezoresistive
pressure transducer in conjunction with a 4643 Kistler amplifier module. Crank angle
position was determined using an optical encoder, while an electric motor, controlled by
an ABB variable speed frequency drive, controlled the engine speed. A Motec M4 ECU
(Engine Control Unit) controlled injection timing, injection pulse width, and injection
duty cycle. Before collecting data for each experiment, the engine was run until the
coolant temperature reached 80°C and combustion became steady. A Horiba analyzer
was used for measuring exhaust gases (CO, CO2, O2, unburned hydrocarbons (UHC), and
nitrogen oxides (NOx)). For lean complete combustion, emissions measurements were
used to deduce the normalized air-fuel ratio using,
c 2
oH 2c
n [O ]1 ,nn [CO ]n4 2
(8)
where nc is the number of carbon atoms in the fuel, nH is the number of hydrogen atoms
in the fuel, nO is the number of oxygen atoms in the fuel, [O2] is the percent oxygen
measured in the emissions, and [CO2] is the percent carbon dioxide measured in the
emissions. The number of carbon, hydrogen, and oxygen atoms were estimated using the
fuel composition. The uncertainty in the normalized air-fuel ratio is approximately± 0.05.
EFFECTS OF FUEL COMPOSITION ON AUTO-IGNITION
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 10
The following results demonstrate the impact of fuel composition on auto-ignition in
HCCI engines. First, auto-ignition timing, measured by CA50, of each fuel at various
compression ratios is presented. Second, the Octane Index (OI) is compared with
experimental data. Third, the relative HCCI Index (HIrel) is compared with experimental
data. Fourth, correlations between fuel composition, auto-ignition and low temperature
heat release (LTHR) are investigated.
Auto-Ignition Timing (Ca50)
The effects of fuel composition on auto-ignition were first investigated by measuring the
auto-ignition timing (CA50) of each fuel at various compression ratios. Figure 2 plots
CA50 versus CR for the twenty fuels tested showing that fuel blends with similar ON do
not always auto-ignite at the same CR, which is consistent with previous research (Liu et
al., 2009). The uncertainty in CA50 and CR was calculated to be ±0.5 and ±0.1,
respectively (Taylor, 1997). For example, tested fuels with ON~70 are: PRF70, S70,
TRF70, E-PRF70, G2, and G1-H. As seen in Fig. 2, PRF70, S70, and G1-H auto-ignite
at similar CR values. However, TRF70 auto-ignites half a CR higher than PRF70, E-
PRF70 auto-ignites a full CR higher than PRF70, and G2 auto-ignites three CRs higher
than PRF70. The results also suggest that ethanol inhibits auto-ignition more than toluene
as S70 auto-ignites at the same CR as PRF70, while E-PRF70 auto-ignites a full CR
higher than PRF70 and half a CR higher than TRF70.
The results also show adding the same amount of ethanol, 10% by volume, to G1
(RON=87) and G2 (RON=70) does not have the same effect on auto-ignition. As shown
in Fig. 2, G1-E2 (RON=90) auto-ignites about one CR higher than G1, while G2-E2
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 11
(RON=78) auto-ignites three CRs higher than G2. Like ethanol, the naphthene, N1,
affects auto-ignition of G2 more than G1; G2-N1 (RON=N/A) auto-ignites about half a
CR higher than G2 and G1-N1 (RON=86) auto-ignites at about the same CR as G1.
The Octane Index (Oi)
Although the OI was developed for predicting anti-knock qualities of practical fuels in
spark ignited engines, Kalghatgi (Kalghatgi, 2005) suggests that the OI can be used for
predicting auto-ignition of fuels in an HCCI engine. For the CFR engine running at the
same inlet temperature, inlet pressure, and RPM, the compression ratio when CA50=6
deg ATDC is used for quantifying the fuel’s propensity of autoignition. This operating
condition was chosen because data was successfully collected for all fuels when CA50=6
deg ATDC.Figure3 shows the relationship between OI and compression ratio when a
CA50=6 deg ATDC for nineteen of the twenty fuels tested. The OI could not be
calculated for G2-N1 because RON and MON were unavailable. The OI accurately
predicts auto-ignition of the primary reference fuel (PRF) blends, agreeing well with
previous results (Liu et al., 2009; Yao, Zheng and Liu, 2009). These results were
expected because for PRF blends OI = ON. The OI poorly predicts auto-ignition of some
fuels with the same ON. For example, the OI predicts S70, E-PRF70, and G2 will have
similar auto-ignition characteristics; however, E-PRF70 auto-ignites almost one CR
higher than S70 and G2 auto-ignites almost two CR higher than S70.
Additionally, the OI poorly predicts auto-ignition of fuels containing naphthenes. The OI
predicts G1-N1 (RON=86, MON=79) and G1-N2 (RON=87, MON=81) having similar
auto-ignition characteristics, but the experimental results show G1-N2 auto-igniting one
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 12
CR higher than G1-N1. The OI also predicts G1-A2 (RON=91, MON=82) auto-igniting
after G1-N2, but the experimental results show G1-N2 auto-igniting one CR higher than
G1-A2. Although the OI correlates with the RON and MON, it is not sufficient for
predicting auto-ignition of fuels with similar ON.
The Relative HCCI Index (Hirel)
The HIrel, introduced by Shibata and Urushihara (2007), is the first published research for
predicting auto-ignition of fuels in an HCCI engine using the fuel composition and MON.
Like the OI, the HIrel predicts auto-ignition order of fuels; fuels with higher HIrel require
higher CR for auto-ignition (i.e. more difficult to auto-ignite). Figure 4 shows the
relationship between HIrel and CR when CA50=6 deg ATDC for nineteen of the twenty
fuels tested. The HIrel could not be calculated for G2-N1 because MON was unavailable.
The HIrel accurately predicts auto-ignition of some PRF blends, and some gasoline fuels
blended with ethanol. However, the HIrel does not accurately predict ignition order of
some fuels with similar ON. For example, PRF70, S70, G1-H, TRF70, E-PRF70, and G2
have the same HIrel, but experimental results show TRF70, E-PRF70, and G2 auto-
igniting at different CRs than PRF70, S70, and G1-H and G2-E2 auto-igniting at different
CR’s than PRF 75.
The HIrel also poorly predicts auto-ignition of fuel blends containing different aromatics.
The HIrel assumes that fuels containing different aromatic compounds at the same
concentration and approximately same MON will have similar effects on auto-ignition, as
all aromatics are grouped together in Eq. (2). Fig. 4 shows that different aromatics at the
same concentration and approximately same MON can have different effects on auto-
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 13
ignition. For example, G1-A1 has a lower HIrel than G1-A2, but the experimental results
show G1-A1 auto-igniting half a CR higher than G1-A2. Since naphthenes are not
included in Eq. (2), and if their effects are assumed to be the same, the HIrel relation
would predict G1-N1 and G1-N2 having similar effects on auto-ignition (Shibata and
Urushihara, 2007). However, Fig. 4 shows G1-N1 auto-igniting one CR lower than G1-
N2. It should be noted that temperature dependent constants used for computing HIrel
were developed using the K factor in the OI. Therefore, the results from the HIrel were
expected to be similar to the results from the OI.
Low Temperature Heat Release
Previous research (Shibata et al., 2005) suggests low temperature heat release (LTHR)
may correlate with auto-ignition better than high temperature heat release (HTHR).
Figures 5 and 6 show the dependence of the heat release ratio (the ratio of average LTHR
to HTHR) on CR. For gasoline fuels, the heat release ratio decreases almost linearly as
CR (when CA50=6 deg ATDC) increases from 9 to 15 (see Fig. 5). Gasoline fuels with
the same ON show different heat release ratios and correlate well with CR. For example,
G1-H auto-ignites almost 2 CRs lower than G2 and the heat release ratio predicts G1-H
has about 3% less LTHR than G2. The heat release ratio also suggests gasoline fuels with
more LTHR will auto-ignite at lower CR, agreeing with previous research (Shibata et al.,
2005). For gasoline fuels auto-igniting at CRs greater than 15, no LTHR was detectable
using our instrumentation.
Figure 6 shows reference fuels with similar ON (PRF70, S70, TRF70, and E-PRF70)
have similar amounts of LTHR, suggesting the reference fuels should auto-ignite at
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 14
similar CRs. However, these reference fuels do not auto-ignite at the same CR. For
example, ethanol in E-PRF70 was expected to suppress LTHR. Instead, E-PRF70 shows
similar amounts of LTHR as PRF70 while auto-igniting at the same CR as PRF75. The
results suggest that the addition of Ethanol makes auto-ignition more difficult, but
ethanol does not necessarily suppress LTHR. One possible explanation for E-PRF70
exhibiting the same LTHR as PRF70 is that the 31% n-heptane in E-PRF70 may promote
more LTHR than ethanol suppresses. For fuels exhibiting decreasingly low amounts of
LTHR, a water-cooled pressure transducers may provide better resolution (Sjoberg and
Dec, 2003; Stone, 1999).
Overall, the results show that LTHR correlates well with auto-ignition of gasoline blends
exhibiting LTHR, but does not correlate well with reference fuel blends. This suggests
that reference fuels for spark-ignited engines may not be appropriate reference fuels for
HCCI engines.
CONCLUSIONS
In this paper, we investigated the impact of fuel composition on auto-ignition in HCCI
engines in order to develop a future metric for predicting fuel performance in future
HCCI engine technology. The following conclusions were derived:
• For a fixed intake temperature, intake pressure, and equivalence ratio, fuels with
the same Octane Number (ON) do not auto-ignite at similar compression ratios (CR).
Additionally, the effect of ethanol (and naphthene) in delaying auto-ignition is dependent
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 15
on the gasoline blendstocks.
The Octane Index and relative HCCI Index correlate well with auto-ignition of primary
reference fuel but poorly predict auto-ignition of gasoline fuel blends containing
naphthenes, aromatics, and ethanol.
Low temperature heat release (LTHR) correlates well with auto-ignition for gasoline
fuels with measurable LTHR but does not correlate well for reference fuels. The results
suggest that reference fuels may not be appropriate for describing fuel performance in
HCCI engines. For fuels auto-igniting at CRs greater than 15, LTHR could not be
detected.
More than one metric may be required for predicting auto-ignition. For gasoline fuels that
exhibit LTHR, LTHR better predicts auto-ignition order than Octane Index and the
Relative HCCI Index. For fuels that do not exhibit LTHR, a different metric is needed.
To further advance development of a future metric for predicting fuel performance in
future HCCI engine technology, we recommend that more fuel blends containing linear
amounts of toluene, ethanol, and various aromatics, by volume, should be explored to
help identify reference fuels for a standard HCCI number. Additionally, different test
conditions, such higher RPM and lower intake temperatures, should be explored further
for the fuels used in this research. Trends established at different operating conditions
could be used with trends found in this paper to establish a standard HCCI metric.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 16
ACKNOWLEDGEMENTS
Research conducted at the University of California, Berkeley was supported by the
Chevron Corporation. The authors also wish to acknowledge the assistance of T.
Dillstrom, A. Van Blarigan, and M. Wissink in conducting experimental measurements.
ABBREVIATIONS
ATAC Active Thermo-Atmosphere Combustion
ATDC After top dead center
CA50 Crank angle at which 50% of heat has been released
CAD Crank angle degree
CFR Cooperative Fuel Research
CIHC Compression ignition homogenous charge
CR Compression ratio
DI Direct Injection
E-PRF Ethanol Reference Fuel
ECU Engine control unit
HCCI Homogenous Charge Compression Ignition
HIrel, Relative HCCI Index
HTHR High temperature heat release
LTHR Low temperature heat release
MON Motor Octane Number
OI Octane Index
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 17
ON Octane Number
PRF Primary Reference Fuel
RON Research Octane Number
SI Spark Ignition
TRF Toluene Reference Fuel
REFERENCES
Anderson, J.E., Kramer, U., Mueller, S.A., and Wallington, T.J. 2010. Octane numbers of
ethanol and methanol gasoline blends estimated from molar concentrations. Energy and
Fuels, 24, 6576-6585.
ASTM MON Standard 2011. D2700-11, Standard test method for motor octane number
of spark-ignition engine fuel. ASTM International West Conshohocken, PA,
DOI:10.1520/D2700-11, www.astm.org.
ASTM RON Standard 2011. D2699-11, Standard test method for research octane
number of spark-ignition engine fuel. ASTM International West Conshohocken, PA,
DOI:10.1520/D2699-11, www.astm.org.
Au, M.Y., Girard, J.W., Dibble, R.W., Flowers, D., Aceves, S.M., Martinez-Frias, J., Ray
Smith, C.S., and Mass, U. 2001. 1.9-liter four cylinder HCCI engine operation with
exhaust gas recirculation. SAE Technical Paper 2001-01-1894.
Canova, M., Chiara, F., Cowgill, J., Midlam-Mohler, S., Guezennec, Y., and Rizzoni, G.
2007. Experimental Characterization of Mixed-Mode HCCI/DI Combustion on a
Common Rail Diesel Engine. SAE Technical Paper 2007-24-0085.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 18
Christensen, M., Johansson, B., and Einewall, P. 1997. Homogeneous charge
compression ignition (HCCI) using isooctane, ethanol and natural gas - a comparison
with spark-ignition operation. SAE Technical Paper 972874.
Eng, J.A. 2002. Characterization of pressure waves in HCCI combustion. SAE Technical
Paper 2002-01-2859.
Eng, J., Leppard, W., and Sloane, T. 2003. The effect of di-tertiary butyl peroxide
(DTBP) addition to gasoline on HCCI combustion. SAE Technical Paper 2003-01-3170.
Erlandsson, O. 2002. Early Swedish hot-bulb engines - Efficiency and performance
compared to contemporary gasoline and diese engines. SAE Technical Paper 2002-01-
0115.
Flowers, D., Aceves, S.M., Martinez-Frias, J., Smith, J.R., Au, M.Y., Girard, J.W., and
Dibble, R.W. 2001. Operation of a four-cylinder 1.9L propane-fueled homogeneous
charge compression ignition engine: Basic operating characteristics and cylinder-to-
cylinder effects. SAE Technical Paper 2001-01-1895.
Fuhs, A. 2008. Hybrid Vehicles and the Future of Personal Transport. Boca Raton: CRC
Press.
Gray, A.W., and Ryan, T.W. 1997. Homogeneous charge com- pression ignition (HCCI)
of diesel fuel. SAE Technical Paper 971676.
Helmantel, A., and Denbratt, I. 2004. HCCI operation of a passenger car common rail DI
Diesel engine with early injection of Conventional Diesel fuel. SAE Technical Paper
2004-01-0935.
Heywood, J.B. 1988. Internal Combustion Engine Fundamentals. McGraw-Hill, Inc.
Hiltner, J., Agama, R., Mauss, F., Johansson, B., and Christensen, M. 2000. HCCI
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 19
operation with natural gas: Fuel composition implications. Proceedings of the 2000
ASME International Combustion Engine Fall Technical Conference., 35(2), 11-19.
Kalghatgi, G.T. 2005. Auto-Ignition Quality of Practical Fuels and Implications for Fuel
Requirements of Future SI and HCCI Engines. SAE Technical Paper 2005-01-0239.
Killingsworth, N.J. 2007. HCCI Engine Control and Optimization. University of
California, San Diego.
Koopmans, L., Strom, H., Ludgren, S., Baklund, O., and Denbratt, I. 2003.
Demonstrating a SI-HCCI-SI Mode Change on a Volvo 5-Cylinder Electronic Valve
Control Engine. SAE Technical Paper 2003-01-0753.
Liu, H., Yao, M., Zhang, B. and Zheng, Z. 2009. Influence of fuel and operating
conditions on combustion characteristics of a homogeneous charge compression ignition
engine. Energy and Fuels, 23, 1422-1430.
Morgan, N., Smallbone, A., Bhave, A., Kraft, M., Cracknell, R., and Kalghatgi, G. 2010.
Mapping surrogate gasoline compositions into RON/MON space. Combustion and Flame,
157, 1122-1131.
Najt, P.M., and Foster, D.E. 1983. Compression-ignited homogeneous charge combustion.
SAE Technical Paper 830264.
Noguchi, M., Tanaka, Y., Tanaka, T., and Takeuchi, Y. 1979. A study on gasoline engine
combustion by observation of intermediate reactive products during combustion. SAE
Technical Paper 790840.
Oakley, A., Zhao, H., Ma, T., and Ladommatos, N. 2001. Dilution effects on the
controlled auto-ignition (CAI) combustion of hydrocarbon and alcohol fuels. SAE
Technical Paper 2001-01-3606.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 20
Olsson, J.-O., Tunestal, P., Johansson, B., Fiveland, S., Rey Agama, M.W., and Assanis,
D. 2002. Compression ratio influence on maximum load of a natural gas fueled HCCI
engine. SAE Technical Paper 2002-01-0111.
Onishi, S., Jo, S.H., Shoda, K., Jo, P.D., and Kato, S. 1979. Active thermo-atmosphere
combustion (ATAC) - a new combustion process for internal combus- tion engines. SAE
Technical Paper 790501.
Shibata, G., Oyama, K., Urushihara, T., and Nakano, T. 2005. Correlation of Low
Temperature Heat Release With Fuel Composition and HCCI Engine Combustion. SAE
Technical Paper 2005-01-0138.
Shibata, G., and Urushihara, T. 2007. Auto-Ignition Characteristics of Hydrocarbons and
Development of HCCI Fuel Index. SAE Technical Paper 2007-01-0220.
Shudo, T., and Ono, Y. 2002. HCCI combustion of hydrogen, carbon monoxide and
dimethyl ether. SAE Technical Paper 2002-01-0112.
Sjoberg, M., and Dec, J.E. 2003. Combined Effects of Fuel-Type and Engine Speed on
Intake Temperature Requirements and Completeness of Bulk-Gas Reactions for HCCI
Combustion. SAE Technical Paper 2003-01-3173.
Stanglmaier, R.H., Ryan, T.W., and Souder, J.S. 2001. HCCI operation of a dual-fuel
natural gas engine for improved fuel efficiency and ultra-low NOx emissions at low to
moderate engine loads. SAE Technical Paper 2001-01-1897.
Stone, R. 1999. Introduction to internal combustion engines, 3rd ed., Warrendale:
Society of Automotive Engineers, Inc.
Taylor, J.R. 1997. An Introduction to Error Analysis: The Study of Uncertainties in
Physical Measurements, 2nd ed., University Science Books Sausalito, California.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 21
Thring, R.H. 1989. Homogeneous-charge compression ignition (HCCI) engines. SAE
Technical Paper 892068.
Wu, J., and Zhang, H. 2012. Analysis on Application of HCCI Technology for Hybrid
Electric Vehicles. Applied Mechanics and Materials, 128-129, 803-806.
Yao, M., Zheng, S., and Liu, H. 2009. Progress and recent trends in homogenous charge
compression ignition (HCCI) engines. Progress in Energy and Combustion Science, 35,
398-437.
Zhang, Y., Xie, H., and Zhao, H. 2009. Investigation of SI-HCCI Hybrid Combustion and
Control Strategies for Combustion Mode Switching in a Four- Stroke Gasoline Engine.
Combustion Science and Technology, 181(5), 782-799.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 22
Table 1. CFR engine specifications
Displacement 0.616 L
Stroke 114.3 mm
Bore 82.8 mm
Connecting Rod 254 mm
Engine Speed 600 RPM
Coolant Temperature 80°C ±1°C
Intake Pressure 1.035 bar
Intake Temperature 150°C ±1°C
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 23
Table 2. Reference fuel blend compositions by volume percent with RON and MON
Fuel Name % iso-ocatne % n-heptane % Toluene % Ethanol RON MON
PRF 60 60 40 0 0 60 60
PRF 70 70 30 0 0 70 70
PRF 75 75 25 0 0 75 75
PRF 85 85 15 0 0 85 85
PRF 88 88 12 0 0 88 88
TRF 70* 0 46 54 0 70.5 63.5
S 70 64 31 5 0 70.5 69.6
E-PRF 70+ 64 31 0 5 69.5 68.6
n-heptane 0 100 0 0 0 0
*RON and MON were calculated using method described by Morgan et al. (2010).
+RON and MON were calculated using bRON and bMON values from Anderson et al.
(2010).
RON and MON of remaining fuels were assumed.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 24
Table 3. Base gasoline hydrocarbon class information (percent by volume)
Fuel
Name
%N-
Paraffins
%Iso-
Paraffins
%Ole
fins
%Cycloparaffins %Arom
atics
RO
N
MON
G1 14.2 44.9 5.2 9.8 25.9 87 80
G2 4.7 48.2 0.3 34.4 12.4 70 65
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 25
Table 4. Gasoline fuel blend information
Fuel Name Additive RON MON
G1-H N-heptane 70b 65b
G2-H N-heptane 60a 55a
G1-E1 Ethanol 93a 84a
G1-E2 Ethanol 90a 82a
G2-E2 Ethanol 78a 68a
G1-A1 Toluene 91a 82a
G1-A2 O-Xylene 91b 82b
G1-N1 Methylcyclohexane 86b 79b
G2-N1 Methylcyclohexane N/A N/A
G1-N2 Cyclohexane 87b 81b
“N/A” implies not available
aRON or MON was estimated
bRON or MON was provided by Chevron
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 26
Figure 1. Distinction between low temperature heat release (LTHR) from low temperature combustion and high temperature heat release (HTHR) from high temperature combustion.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 27
Figure 2. Auto-ignition in HCCI engines varies almost linearly with compression ratio. Fuels with the same octane number (ON), such as G2 and G1, do not auto-ignite at the same compression ratios. Error bars are suppressed for visibility. Error in CA50 is typically ± 0.5 of the shown value.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 28
Figure 3. The Octane Index poorly predicts auto-ignition fuels with similar Octane Number (ON) but shows an almost linear relationship (R2=0.90) with compression ratio for all fuel blends.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 29
Figure 4. HCCI index correlates well with primary reference fuel blends but poorly predicts auto-ignition of fuel blends containing ethanol, aromatics, or naphthenes.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 30
Figure 5. For gasoline fuels, the ratio of LTHR to HTHR shows an almost linear decrease with the compression ratio at a CA50=6 deg ATDC.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 31
Figure 6. Reference fuels with the same ON have similar amounts of LTHR even though they auto-ignite at different compression ratios. PRF blends decrease with compression ratio at a CA50=6 deg ATDC.
Dow
nloa
ded
by [
Uni
vers
iti T
ekno
logi
Mal
aysi
a] a
t 20:
55 1
1 Ja
nuar
y 20
13