Fuel 142 (2015) 2837

Contents lists available at ScienceDirect

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

Direct injection of neat n-butanol for enabling clean low temperature

combustion in a modern diesel engine
Ming Zheng a,, Tie Li b, Xiaoye Han a
a
b

Clean Combustion Engine Laboratory, Department of Mechanical, Automotive and Material Engineering, University of Windsor, Ontario, Canada
State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai, China

h i g h l i g h t s
 A rarely studied fuel, n-butanol, is applied to replace diesel for clean combustion.
 Low temperature combustion is enabled via n-butanol high pressure direct injection.
 Emission benets and control challenges of n-butanol combustion are identied.
 n-Butanol combustion offers ultralow NOx and near-zero smoke emissions without EGR.
 Improved combustion control using multi-pulse injections with moderate EGR.

a r t i c l e

i n f o

Article history:
Received 20 August 2014
Received in revised form 17 October 2014
Accepted 27 October 2014
Available online 7 November 2014
Keywords:
n-Butanol
Direct injection
Low temperature combustion
Ultralow NOx and smoke
Diesel engine efciency

a b s t r a c t
This study investigates the effects of neat n-butanol replacing conventional diesel fuels to enable clean
combustion on a modern common-rail diesel engine. Systematic engine experiments are conducted to
examine the combustion characteristics and exhaust emissions in correlation to n-butanols relatively
high oxygen content and high volatility but low ignitability, and control strategies are thereafter developed for enabling clean and efcient combustion of neat n-butanol. Compared to its diesel counterpart,
the single-shot injection of neat n-butanol offers substantially reduced NOx emissions without the use of
EGR and near-zero soot emissions, but the applicable injection timing window is narrower for n-butanol
limited by high maximum rates of pressure rise and/or unstable combustion. EGR is effective to reduce
the combustion roughness, but it further narrows the applicable injection timing window and deteriorates the HC and CO emissions. A control strategy that deploys multi-shot injections combined with moderate use of EGR is developed and applied to improve the combustion controllability and exhaust
emissions while minimizing the penalties in the engine efciency.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
Butanol can be produced from alcoholic fermentation of biomass feed stocks including edible materials such as corn, sugar
cane, and molasses, as well as agricultural wastes such as wheat
straw, corn stover and other celluloses. It is also reported that
the crude glycerol, a by-product of biodiesel production during
the transesterication processes, can be converted into the valueadded biofuels comprised of mainly butanol [1,2]. As one of the
next generation biofuels, butanol has attracted increasing attention for engine applications in recent years. There are two isomers
(n-butanol and sec-butanol) of butanol with the straight carbon
chain, and two other isomers (isobutanol and tert-butanol) with
the branched carbon chain. Normal-butanol (n-butanol) with a
Corresponding author.
http://dx.doi.org/10.1016/j.fuel.2014.10.075
0016-2361/ 2014 Elsevier Ltd. All rights reserved.

straight carbon chain structure and a hydroxyl at the terminal carbon site is used in this study.
As a fuel, butanol has several advantages over ethanol for combustion engine applications. It is less corrosive and less prone to
water contamination than ethanol, and thus minor or no modications are required for the existing infrastructure for gasoline to be
used for the butanol fuel distribution. The use of butanol replacing
gasoline on spark ignition (SI) engines has been demonstrated in
previous research without any engine modications, such as in
[3]. However, its lower octane number and lower latent heat of
vaporization could be drawbacks of n-butanol as a fuel for SI
engines, compared to ethanol. The lower latent heat of vaporization of n-butanol tends to reduce the charge density and thus
decrease the engine power output, while the lower octane number
leads to higher propensities to engine knock and reduced engine
efciency [4,5].

M. Zheng et al. / Fuel 142 (2015) 2837

29

Nomenclature
ABE
CA
CA5
CA50
CA95
CAI
CI
CO
DI
DOC
dp/dhmax
EGR
FPGA
FTIR
THC
HCCI
HCLD

acetonebutanolethanol
crank angle
crank angle of 5% total heat release
crank angle of 50% total heat release
crank angle of 95% total heat release
California analytical instruments
compression ignition
carbon monoxide
direct injection
diesel oxidation catalyst
maximum rate of pressure rise
exhaust gas recirculation
eld programmable gate array
Fourier transform infrared spectroscopy
total hydrocarbons
homogeneous charge compression ignition
heated chemiluminescence detector

Higher fuel conversion efciency is expected when n-butanol is

used in the lean-burn and high compression ratio diesel engines
instead of conventional SI engines. Compared to the majority of
diesel fuels, n-butanol has certain preferred fuel properties for
enabling low temperature combustion (LTC), including the relatively higher oxygen content and high volatility along with lower
ignitability, which are deemed helpful to improve fuelair mixing
and reduce particulate emissions of diesel engines. Owing to its
less hydrophilicity and higher miscibility with diesel fuel than ethanol, n-butanol is usually blended with diesel for engine tests in
most studies [613]. Rakopoulos et al. conducted a series of experimental studies on the effects of n-butanol diesel blends on the
performance and emissions of a heavy-duty diesel engine in either
steady state operation [6,8] or during transient acceleration cycles
[7]. Dogan [9] and Siwale et al. [13] evaluated the effects of
n-butanol diesel blends on the combustion characteristics and
exhaust emissions of a single-cylinder naturally aspirated diesel
engine and a turbocharged four-cylinder automotive engine,
respectively. These studies demonstrated that the soot emissions
generally decreased remarkably with the increasing use of n-butanol
in the blends, while nitrogen oxides (NOx) decreased or slightly
increased in different studies and under different engine operating
conditions, but the increased NOx emissions could be effectively
suppressed by the use of exhaust gas recirculation (EGR).
Ballesteros et al. conducted carbonyls speciation of the exhaust
gas from an automotive diesel engine fueled with diesel, bioethanol-diesel blends, and butanol-diesel blends [10]. The results indicated that the combustion of bio-alcohol blends produced higher
carbonyl emissions than that of the pure diesel, and these carbonyl
emissions could be effectively reduced through the diesel oxidation catalysts (DOC). Valentino et al. studied the performance
and emissions of a high-speed turbocharged common-rail diesel
engine fueled with blends of n-butanol and diesel under the premixed low temperature combustion mode [11], and simultaneous
reduction of NOx and soot emissions was obtained when the fuel
injection was completed prior to the ignition event. This could be
attributed to the prolonged ignition delay and improved fuelair
mixing before the start of combustion, which was achieved by
appropriate control over the injection pressure, injection timing
and intake oxygen concentration with the joint effect of increased
resistance to auto-ignition and enhanced volatility of n-butanol
blends than those of the pure diesel fuel. Merola et al. conducted
an optical investigation on the spray and combustion processes
of n-butanol diesel blends and, as shown by the test results, a

HFID
HRR
ID
IDmain
IMEP
LTC
NDIR
NOx
PCCI
PFI
ppm
pinj
pint
SI
SOI
SOImain
TDC

heated ame ionization detector

heat release rate
ignition delay
ignition delay for the main injection
indicated mean effective pressure
low temperature combustion
non-dispersive infrared detector
nitrogen oxides
premixed charge compression ignition
port fuel injection
parts per million
pressure of injection
pressure of intake
spark ignition
start of injection
start of main injection
top dead centre

minor increase in NOx emissions but a signicant reduction of soot

emissions were observed in the spray ame of fuel blends compared to those of the pure diesel fuel [12]. Studies using n-butanol
and bio-diesel blends showed similar results to those of n-butanol
and diesel blends in terms of the engine performance and exhaust
emissions [14,15]. Moreover, Lin et al. [16] and Chang et al. [17,18]
investigated the water-containing butanol to simulate the
hydrated n-butanol produced from the acetonebutanolethanol
(ABE) fermentation and a simple distillation treatment. As emphasized in this work, the production of hydrated butanol rather than
dehydrated butanol could substantially reduce the cost and energy
consumption during the fuel rening processes. The results also
indicated that the use of diesel blends with hydrated butanol
(510%) did not require any modications to the engine hardware
while offering improved energy efciency and reduced pollutant
emissions.
Although the n-butanol blends with diesel and/or biodiesel
have been extensively investigated as fuels for diesel or compression ignition (CI) engines, there are relatively limited publications
on the use of neat n-butanol for CI engines. Chen et al. proposed a
dual-fuel concept with port fuel injection (PFI) of n-butanol to prepare a premixed fuelair mixture and direction injection of diesel
fuel to ignite the mixture [19]. Han et al. [20] and He et al. [21],
respectively, investigated the combustion characteristics and
exhaust emissions for the compression ignition of the intake port
injected neat n-butanol. All of these studies exhibited the potential
of n-butanol for enabling efcient and clean combustion in CI
engines. Particularly, preliminary tests of Han et al. [20] suggested
that n-butanol delivered via high-pressure direct injection might
be a more suitable fuel to enable clean combustion than diesel.
Detailed studies on the combustion characteristics of neat
n-butanol using direct injection are rarely reported so far. Compared to the port fuel injection, the direction injection can provide
more degrees of freedom for the control over the injection and the
subsequent combustion process. In particular, the fueling strategies and fuel properties have signicant impacts on engine operations in the low temperature combustion (LTC) mode, while
optimization of the engine design (e.g. compression ratios) and/or
operating parameters (e.g. EGR ratios) can be effective to enable
LTC for simultaneous reduction of NOx and particulate emissions,
as reported in [2232]. In terms of fuel properties, empirical results
suggested that less ignitable fuels could assist to extend the LTC
operation to higher engine loads, while the impact of fuel volatility
was relatively minor [3336]. In comparison to the majority of

M. Zheng et al. / Fuel 142 (2015) 2837

Injection Control System

Single Cylinder

Injection Control System Multiple Cylinders

Diesel pinj

Injection Command
Injection Bench
n-Butanol Tank

HP
Pump

AC
Motor

4th

Original
HP
Pump
Filter

3rd

2nd

2. Experimental setup

Regulator

Feed
Pump

Fuel Cooler

Conditioning &
Measurements

Original Common Rail

diesel fuels, n-butanol has a lower Cetane number (25) and an

oxygen content of 22% by weight. With proper control strategies
over the engine operating parameters such as the air and fuel handling, the application of n-butanol should have great potentials of
enabling LTC on diesel engines.
In this study, the authors intend to explore the potential of
n-butanol direct injection for enabling the low temperature combustion mode on a single-cylinder diesel engine. The investigation
includes (1) the direct replacement of diesel with n-butanol using
single-shot injection, (2) EGR application to reduce combustion
roughness, and (3) combustion control improvement via multishot injections combined with EGR.

n-Butanol pinj

30

Filter
1st

Blocked Ports

2.1. Research engine

n-Butanol Common Rail

The engine tests are conducted on a four-cylinder diesel engine,

and the specications are given in Table 1. This multi-cylinder
engine is recongured to perform single-cylinder research. The
intake and exhaust of the research cylinder are separated from
those of the other three cylinders to prevent cross-contamination.
The single cylinder is comprehensively instrumented for combustion research. The glow plug is replaced by a piezoelectric pressure
transducer (AVL GU13P) to acquire the cylinder pressure. As shown
in Fig. 1, the injector of the single cylinder is decoupled from the
original common rail while keeping others intact; a stand-alone
fuel injection bench is in-house fabricated and connected to the
injector of the single cylinder for implementing the n-butanol
direct injection. This stand-alone fuel bench comprises a highpressure fuel pump driven by an AC motor, a common rail with rail
pressure measurement, a set of fuel conditioning units, and a fuel
ow meter (Ono Sokki FP-213). A set of solenoid injector drivers
(EFS 8232) are used to drive the injectors and injection programs
are developed to independently control the injection events (injection pressure, timing, and duration) for the single cylinder and the
rest three cylinders.
The advanced research platform offers independent control
over the engine intake boost and EGR, as shown in Fig. 2. Dry
and clean air is externally compressed by an auxiliary compressor
to simulate the boosted intake, whose pressure is controlled by a
digital pressure regulator. The air ow rate is measured by an air
ow meter (Roots 2M175) installed upstream of the intake surge
tank. In the exhaust loop, a pneumatically controlled valve is used
to restrict the exhaust outow and thus to control the exhaust
backpressure. The EGR rate can therefore be modulated through
the control of the EGR valve opening and the exhaust backpressure
for a given intake boost. An eddy current dynamometer is connected to the engine for power and speed management. An optical
encoder is mounted on the engine crankshaft to determine the
engine crank angle. External conditioning units are employed to
manage the engine coolant and oil; the temperature is maintained
at 80 C for both the coolant and oil during the engine tests.

Table 1
Engine specications.
Engine type

4-Cylinder, 4-stroke, re-congured

to single cylinder

Displacement (cm3)
Bore  stroke (mm)
Compression ratio ()
Max. Cylinder pressure
Injection system
Injector

1998
86  86
18.2:1
18 MPa
Common-rail (160 MPa)
Solenoid, 6-hole, hole diameter 0.13 mm,
umbrella angle 155

Research
Cylinder

Diesel
Tank

Fig. 1. Fueling system for research engine.

This study focuses on the use of n-butanol on a direct-injection

diesel engine. As a comparison, the engine performance is also
investigated using the regular diesel fuel. The major fuel properties
are tabulated in Table 2 for these two fuels.
2.2. Combustion analyses and emission measurements
Based on the cylinder pressure measurements, the apparent
heat release analysis is applied as the primary approach to analyze
the combustion process using Eq. (1), where HRR is the heat
release rate [J/CA], c is the ratio of specic heats for the cylinder
charge, p is the measured cylinder pressure [N/m2], V is the cylinder volume [m3], and h is the crank angle [CA].

HRR



1
dV
dp
V
 cp
dh
dh
c  1

The ignition timing of the combustion event is dened as the

crank angle when 5% of the total heat is released (CA5), and the
combustion phasing is represented by 50% of the total heat is
released (CA50). The end of combustion is dened as the crank
angle when 95% of the total heat is released (CA95), and thus the
combustion duration is dened as the time period between CA5
and CA95. The ignition delay is dened as the time period between
CA5 and the start of injection (SOI).
In this study, the net IMEP [bar] over one engine cycle is used to
evaluate the engine load level, as calculated using Eq. (2), where Vd
is the engine displacement [m3].

IMEP

720

phdV=V d =105

The indicated engine efciency is calculated from IMEP [bar]

and the measured fuel ow rate using Eq. (3), where n is the engine
_ f is the measured fuel ow rate [g/s], and LHV is the
speed [rpm], m
lower heating value [kJ/g] for the respective fuel.

gind IMEP  V d  n=m_ f  LHV=1:2  100%

Gas analyzers (summarized in Table 3) are congured to

measure the engine intake (CO2 and O2) and exhaust gases (CO2,
O2, NOx, HC, and CO). The soot emissions are measured by an
AVL-415S smoke meter. The use of EGR is evaluated by the ratio
between intake CO2 concentration and exhaust CO2 concentration,
as shown by Eq. (4).

EGR CO2 int =CO2 exh

31

M. Zheng et al. / Fuel 142 (2015) 2837

130

EGR
Cooler

Intake Pressure
Regulator

Compressor
Cylinder Pressure
Transducer

EGR
Valve

Air Flow
Measurement

Compressed Air

Injector
Intake
Surge
Tank
pint

Emission
Analyzers

Exhaust
Surge
Tank
pexh

Back-pressure
Valve

150

Pressure
Regulator
Exhaust

Coolant
conditioning
Tcoolant

Speed Sensor
Eddy Current
Dynamometer

Oil
conditioning
Toil
poil

Driveshaft
Shaft
Encoder

Torque
Sensor

Fig. 2. Schematic of the advanced engine research platform.

Table 2
Major properties of test fuels.

Table 4
Engine operating conditions for comparison tests between n-butanol and diesel.

Fuel

Diesel

n-Butanol

Formula
Density @20 C (kg/m3)
Cetane number ()
Octane number ()
Lower heating value (MJ/kg)
Oxygen content by mass (%)
Boiling temperature (C)
Flash point (C)
Auto-ignition temperature
@1 bar abs (C)
Lubricity, HFRR WSD @60 C (lm)
Kinematic viscosity @40 C (cSt)

CH1.89
846
46.5
25
43.5
0
229337 (T5T95)
73.3
254 285

C4H9OH
810
25
87
33.1
21.62
117.4
35
355

300
2.5

622
2.22, 3.6@20 C

a
b

Injection duration (ls)

SOI sweep (CA)
CA50 (CA)

1500
6.5
600
2
0
20.7
n-Butanol

Diesel

900
331.0347.0
365.8375.1

650
335.0369.0
346.3389.2

Table 5
Selected test points for comparison tests between n-butanol and diesel.

Table 3
Emission analyzers.
Analyzers type

Measured emissions

Model

Flame ionization (HFID)a

Non-dispersive infrared (NDIR)a
Chemiluminescence (HCLD)a
Paramagnetic, NDIRb
FTIRa

THC (ppm)
CO (ppm) CO2 (%)
NO & NO2 (ppm)
O2 (%), CO2 (%)
NO, light HC species,
CO (ppm)
Smoke (FSN)

CAI 300M HFID

CAI 200/300 NDIR
CAI 600 HCLD
CAI 602P
MKS 2030HS

Smoke metera

Engine operating conditions

Engine speed (rpm)
Nominal IMEP (bar)
Injection pressure (bar)
Intake pressure (bar abs)
EGR (%)
Intake O2 (vol.%)

AVL 415S

For emission measurement in the exhaust.

For gas measurement in the intake.

Test Point

Fuel

Remarks

A
B
C

n-Butanol
n-Butanol
n-Butanol

n-Butanol

B0
C0
D0
B00
C00
D00

Diesel
Diesel
Diesel
Diesel
Diesel
Diesel

Early injection, late CA50, inefcient combustion

Representative case of n-butanol stable combustion
Transition from stable combustion to late
and unstable combustion
Retarded injection, late CA50, high tendency
towards misre
Counterpart of test point B with same injection timing
Counterpart of test point C with same injection timing
Counterpart of test point D with same injection timing
Counterpart of test point B with similar CA50
Counterpart of test point C with similar CA50
Counterpart of test point D with similar CA50

3. Results and discussion

Engine tests are performed to investigate the combustion characteristics and exhaust emissions of the n-butanol use on the diesel
research engine. In order to examine the impacts of n-butanol
replacing diesel, tests are also conducted using the regular diesel
under the same engine operating conditions for a comparison. Furthermore, experiments are carried out to improve the controllability of n-butanol combustion via the EGR application and multi-shot
injection strategies.

is controlled at 600 bar, and the intake pressure is boosted to 2 bar

absolute. In this set of tests, the EGR is not applied and the intake
oxygen concentration is at 20.7%. The injection timing is swept for
each fuel.
As shown in Table 5, four test points in the n-butanol test are
selected for detailed analyses, which represent different modes of
the n-butanol combustion; the corresponding diesel test points
are used for the comparison purpose.

3.1. Comparison tests for n-butanol and diesel

3.1.1. Combustion characteristics

Under the tested conditions, n-butanol exhibits signicantly
longer ignition delay durations and only allows a much narrower
range of the injection timing to attain a successful ignition, compared to those in the diesel case (Fig. 3). For both fuels, the ignition

The engine operating conditions are tabulated in Table 4 for the

comparison tests. The engine runs at a medium speed of 1500 rpm
and a middle load of 6.5 bar nominal IMEP. The injection pressure

M. Zheng et al. / Fuel 142 (2015) 2837

Ignition Delay [ms]

32

4.5

Misfire

CD

2.5
1.5

Test conditions are

shown in Table 4

n-Butanol

3.5

High dpmax
B'

Diesel

C' D'

0.5
390

Misfire

CA5 [CA]

380

n-Butanol
370

D
TDC

360
350

High dpmax

Diesel
C' D'

B'

340
390

Misfire

380

CA50 [CA]

n-Butanol

370
360
350

Diesel

B'
340
330

335

TDC

C' D'

High dpmax

340

345

350

355

360

365

370

SOI [CA]
Fig. 3. Comparison tests ignition delay, CA5, and CA50 versus SOI.

200
150

SOI:
A@331CA
B&B'@338CA
C&C'@345CA
D&D'@347CA

B' C' D'

30

Test conditions are

shown in Table 4

B C

mf [mg/cycle]

250

A
D

100
50

A
22

n-Butanol
B

mf_butanol Calculated
mf_butanol Measured
mf_diesel Calculated
mf_diesel Measured

Misfire
C D

18
14

Diesel
High dpmax
335

Test conditions are shown in Table 4

340

345

350

355

360

365

370

SOI [CA]

200

160

50

B'

120

B C

Diesel

40

80

C'

D'

40
0
-40
330

26

10
330

ind [%]

Heat Release Rate [J/CA]

Cylinder Pressure [bar]

delay generally shortens when the injection timing is postponed

towards TDC until the ignition and combustion events are considerably retarded into the expansion stroke (e.g. after 370CA). For
n-butanol, an excessively advanced or postponed injection leads
to retarded ignition timing (CA5) and combustion phasing
(CA50), whereas the diesel case presents a nearly linear correlation
between injection and combustion. Due to the low reactivity,
the ignition of n-butanol generally occurs after TDC, despite the
considerably early injection timings.
The cylinder pressure and heat release rate traces are plotted
in Fig. 4 for n-butanol test points A, B, C, and D, along with the

corresponding diesel test points B0 , C0 , and D0 . When the fuel is

injected at the same timing, the diesel combustion events
generally occur much earlier than those of n-butanol. As shown
by the heat release proles, the diesel combustion generally
presents a two-phase combustion pattern, i.e. the premixed phase
and diffusion phase, whereas the n-butanol combustion exhibits
only the premixed burning.
The fuel ow rates and indicated engine efciency are plotted in
Fig. 5 for this fuel comparison test. In addition to the directly measured fuel ow rates (solid markers), the carbon balance method is
applied to estimate the fuel ow rate (hollow markers) based on
the measurements of exhaust emissions (CO2, HC, and CO) and
the air ow rate. For test point A, the calculated n-butanol consumption is substantially lower than the measured fuel ow rate,
which indicates that a large portion of the injected n-butanol has
been lost rather than efciently burned. Such an n-butanol loss
may be attributed to the oil dilution resulted by wall-wetting
and the condensation of unburned n-butanol and other heavy
hydrocarbon species along the exhaust sampling line.
The injector (umbrella angle 155) and piston bowl (deep
omega) of the test engine are originally designed for diesel combustion, and thus the wall impingement tends to occur when the
injection is substantially advanced from the conventional diesel
injection timings. For test point A where an early injection is commanded at 331CA (29CA before TDC), it is highly possible that a
large portion of the n-butanol spray hits on the piston edge and/
or the cylinder wall rather than into the piston bowl. Due to the
wall wetting and the low reactivity of n-butanol (Cetane number
25), part of the n-butanol (in both liquid and gas phases) in the
crevice volume can survive the combustion process and enter the
engine crank case, while most of the rest is emitted with the engine
exhaust. As a result, the engine efciency for test point A is substantially reduced. Under the tested conditions and with this
engine setup, injection timings earlier than 333CA are deemed
unsuitable to attain efcient n-butanol combustion.
Since the combustion phasing has a major impact on the engine
efciency, the engine efciency results are plotted against CA50 in
Fig. 5. The n-butanol combustion (e.g. test points B and C) can produce higher indicated engine efciencies than the diesel combustion under the same engine operating conditions. For these two
test points (as shown in Fig. 6), the n-butanol combustion presents

30

n-Butanol

20
10

340

350

360

370

380

390

Crank Angle [CA]

Fig. 4. Comparison tests cylinder pressure and HRR for same SOI.

0
345

350

355

360

365

370

375

380

385

390

CA50 [CA]
Fig. 5. Comparison tests fuel consumption and indicated efciency.

33

Test conditions are

shown in Table 4

200

C"

B"
D"

150

100

50

dp/dmax [bar/CA]

200

160

120

80

C"

40

15

B"

12

n-Butanol

9
6

Diesel

25

20

15

n-Butanol

Diesel

10

5
0
360

362

364

366

355

360

365

370

368

370

372

374

376

378

380

CA50 [CA]

D"
350

0
-40
345

Heat Release Rate [J/CA]

Test conditions are

shown in Table 4

18

250

COVIMEP [%]

Cylinder Pressure [bar]

M. Zheng et al. / Fuel 142 (2015) 2837

375

380

385

Fig. 7. Comparison tests COVIMEP and dp/dhmax.

390

2.5

Diesel

1600

B"

C"

D"

2.0

A
1.5

n-Butanol
C
B

1.0

NOx [ppm]

Comb. Duration [ms]

Crank Angle [CA]

0.5

1200
800
400

n-Butanol

0
360 362 364 366 368 370 372 374 376 378 380

3.1.2. Exhaust emissions

The test results of major exhaust emissions are shown in Fig. 8.
Without the use of EGR, the n-butanol combustion produces considerably less NOx emissions (<100 ppm) than the diesel baseline,
which indicates a lowered ame temperature during the premixed
combustion of an overall lean and highly homogeneous n-butanolair mixture. At an engine load of 6.5 bar IMEP and without EGR, the
diesel combustion produces low soot emissions (<0.4 FSN), while
the n-butanol combustion generates near-zero smoke. The HC
and CO emissions are much higher for the n-butanol combustion,
especially in the cases of possible wall wetting (test point A) and
misre (test point D).
As suggested by presented results, the use of n-butanol on the
diesel engine offers great benets in ultralow NOx and soot emissions but penalties of increased HC and CO emissions, and it often

Soot [FSN]

0.3
0.2

Diesel
n-Butanol
BC
A D

0.1
0

HC [ppm]

1200

900

n-Butanol

600
300

BC

Diesel
0
10000

CO [ppm]

sharper heat release proles and shorter combustion durations

than the corresponding diesel combustion (test points B00 and C00 ),
which indicates a higher degree of constant volume heat release
and improved engine efciencies. For test point D, misre tends
to occur and the HC and CO emissions increase substantially (discussed in the emission section later), and thus the indicated engine
efciency is reduced.
In Fig. 7, the coefcient of variation in IMEP (COVIMEP) and the
maximum rate of pressure rise (dp/dhmax) are plotted against
CA50 for both n-butanol and diesel fuels. Despite a narrower range
of the stable operation, the n-butanol combustion (e.g. test point B
and C) shows comparable cycle-to-cycle variations (<3%) to the
diesel baseline. However, the maximum rates of pressure rise are
considerably higher for n-butanol compared with diesel, except
when misre tends to occur.

BC

0.4

CA50 [CA]
Fig. 6. Comparison tests cylinder pressure, HRR, and combustion duration for
similar CA50.

Test conditions are

shown in Table 4

Diesel

7500

n-Butanol

5000

C
B

2500

Diesel
0
345

350

355

360

365

370

375

380

385

390

CA50 [CA]
Fig. 8. Comparison tests NOx, soot, HC, and CO emissions.

encounters combustion control challenges such as narrow ranges

of stable operation and high maximum rates of pressure rise.
3.2. Reduction of combustion roughness via EGR application
In order to reduce the high maximum rates of pressure rise, the
combustion process needs to be slowed down. The use of EGR is
therefore investigated through engine experiments. The engine

Table 6
Engine operating conditions for EGR investigation tests.
Engine operating conditions
Engine speed (rpm)
Nominal IMEP (bar)
Injection pressure (bar)
Intake pressure (bar abs)

1500
6.5
600
2

Experiments

EGR 0%

EGR 27%

EGR 33%

SOI sweep (CA)

333347

333341

333341

EGR 0%
EGR 27%
EGR 33%

150
120

400

Test conditions
in Table 6

300

90

200

60

100

30

SOI@336CA
0
-100
345 350 355 360 365 370 375 380 385 390

Heat Release Rate [J/CA]

M. Zheng et al. / Fuel 142 (2015) 2837

Cylinder Pressure [bar]

34

Crank Angle [CA]

Comb. Duration [ms]

2.0

Test conditions are

shown in Table 6

1.6

30

Prolonged
combustion

Misfire

1.2

24
18

n-Butanol
EGR 0%
EGR 27%
EGR 33%

3000

EGR 0%
EGR 27%
EGR 33%

2400
1800

12

1200

600

0.8

0.03 0.02 0.04

25

Fig. 10. EGR tests cylinder pressure, HRR, and emissions for same SOI.

15
10

dpmax
reduction

Misfire

18

COVIMEP [%]

5
0
50
46

ind [%]

CO [ppm]

20

42
38
34
30
364

n-Butanol
EGR 0%
EGR 27%
EGR 33%

15
12

Test conditions are

shown in Table 6

CA50 window for

stable combustion

Misfire

Misfire

Slightly
reduced
efficiency

0
364

366

368

370

372

374

376

CA50 [CA]
18

366

368

370

372

374

376

CA50 [CA]
Fig. 9. EGR tests combustion duration, dp/dhmax, and gind.

operating conditions are tabulated in Table 6 for these EGR investigation tests. The experiments are conducted at three levels of the
EGR ratio, namely 0%, 27%, and 33%. At each EGR ratio, the injection
timing for n-butanol is postponed from 333CA until misre starts
to occur.
As shown in Fig. 9, for the same combustion phasing (CA50), the
increase of EGR generally prolongs the combustion duration and
reduces the maximum rates of pressure rise, while slightly reducing the engine efciency.
In Fig. 10, the proles of the cylinder pressure and heat release
rate are shown for the same commanded injection timing (336CA)
at different levels of EGR, along with the corresponding emissions.
When EGR is gradually increased, the overall reactivity of the cylinder charge is reduced and, as a result, the n-butanol combustion
event is delayed and the combustion rate is slowed down. In terms
of exhaust emissions, the use of EGR slightly reduces the already
low NOx emissions under the tested conditions, and the soot emissions remain at a near-zero level. However, the HC and CO emissions rise noticeably with the EGR addition.
As shown in Fig. 11, the increase of EGR leads to deteriorated
combustion stability (greater COVIMEP). The applicable timing window for n-butanol injection is further narrowed at higher EGR
rates. Under these tested conditions, the n-butanol combustion

COVIMEP [%]

dp/dmax [bar/CA]

NOx [ppm] Soot [FSN] HC [ppm]

15
12
9

Excessively early
injections
possible wall
impingement

Misfire
Misfire

6
3
0
330

335

340

345

350

SOI [CA]
Fig. 11. EGR tests COVIMEP.

remains stable (COVIMEP < 3%) until the combustion phasing

(CA50) is substantially postponed, e.g. later than 372CA.
The results from the EGR investigation indicate that the use of
EGR is effective to reduce the otherwise high maximum rates of
pressure rise for n-butanol combustion; however, the penalties in
emissions and combustion stability are generally resulted from
such an EGR use, including increased HC and CO emissions and
deteriorated cycle-to-cycle variation (greater COVIMEP). Nonetheless, the NOx emissions are further lowered by EGR while the soot
emissions remain at a near-zero level.
3.3. Improved combustion controllability via multi-shot injections
In order to further improve the controllability of n-butanol
combustion, the multi-shot injection strategy is studied. In this

35

Speed: 1500 rpm

p int: 1.5 bar abs
EGR: 10%
p inj: 600 bar
IMEP: ~10 bar

80
40

NOx, HC, CO [ppm]

120

Delay main injection

100
50

350

360

370

380

390

Crank Angle [CA]

1.2

CA5, CA50 [CA]

390

Ignition delay

385

0.8

380

0.4

375

370

IDmain [ms]

Fig. 12. Double injection tests cylinder pressure and HRR for SOImain sweep.

CA50

365
20

Test conditions are

shown in Figure 12

15

10

0
352

0
354

356

358

360

362

dp/dmax [bar/CA]

CA5
360
12

COVIMEP [%]

1600

0.2

1200

800
400
354

356

358

360

362

364

Fig. 14. Double-shot injection tests NOx, HC, CO, and soot emissions.

Pilot injection @325CA

340

0.6

SOImain [CA]

0
-50
330

Test conditions are

shown in Figure 12

0.4

0
352

354
356
358
360
362

150

HC
Soot

2000

SOI main

364

SOImain [CA]
Fig. 13. Double-shot injection tests CA5, CA50, IDmain, COVIMEP, and dp/dhmax.

set of engine experiments, a pilot injection is applied at an earlier

injection timing (325CA) to form a lean and near-homogeneous
fuelair mixture, and a main injection is commanded near TDC
to initiate the combustion. A small amount of fuel is delivered during the pilot injection event to minimize the potential wall
impingement. The injection durations for the pilot and main injections are 600 ls and 900 ls respectively, and the engine load is
raised to 10 bar IMEP.
The results of the cylinder pressure and heat release rate traces
are shown in Fig. 12 for the double-shot injection tests wherein the
injection timing of the main shot is swept. Stable combustion is
achieved using the double-shot injection strategy. It is noted that,
based on the previous results shown in Section 3.1.1 (e.g. Fig. 3),
misre may occur to either one of the double shots if it is applied
alone. For instance, the reactions of the early pilot fuel may not
release sufcient heat to sustain the combustion events if the main
injection is absent.
In correspondence to the cylinder pressure and heat release
traces (Fig. 12), the ignition and combustion characteristics are
plotted in Fig. 13, including CA5, CA50, COVIMEP, dp/dhmax, and
the ignition delay for the main injection. For the double-shot

Cylinder Pressure [bar]

Heat Release Rate [J/CA]

0
200

NOx
CO

2400

Soot [FSN]

160

Heat Release Rate [J/CA]

Cylinder Pressure [bar]

M. Zheng et al. / Fuel 142 (2015) 2837

200
150

Speed: 1500 rpm

pint: 2.0 bar abs
pinj: 600 bar
IMEP: ~10 bar

EGR 35%

100

EGR 40%
50
0
250
200
150

NOx [ppm]: 50, 34

Soot [FSN]: 0.62, 0.72
dpmax [bar/CA]: 13.3, 13.8
COVIMEP [%]: 1.1, 1.2

450s@285CA
450s@325CA
550s@358CA

100
50
0
-50
330

340

350

360

370

380

390

Crank Angle [CA]

Fig. 15. Triple-shot injection tests cylinder pressure and HRR.

injection strategy, the ignition delay for the main injection is

between 0.3 to 1 ms under the tested conditions, which is
substantially shortened compared to that in the single-shot case
(24 ms). For the single-shot injection strategy, the injection event
is effectively decoupled temporally from the combustion process,
allowing the formation of a highly homogeneous cylinder charge,
but the ignition and combustion processes are essentially subject
to the kinetics of chemical reactions. The double-shot injection
strategy, on the other hand, retains the control over the ignition
timing and the combustion phasing through the timing adjustment
of the main injection that is closely coupled to the combustion
events and, as a result, diffusion burning tends to occur. Such a
double-shot injection strategy offers desired combustion stability
(COVIMEP < 3%) and controllability over the maximum rate of
pressure rise.
While the diffusion burning offers desired combustion controllability, it also leads to increased NOx and soot emissions (Fig. 14),
compared to those in the single-shot case. As the main injection is
postponed, the NOx emissions reduce to a level of 300 ppm; the
soot emissions slightly increase owing to shortened ignition delay.
It is noted that the soot emissions are still below the targeted limit
(i.e. 2 FSN).
In order to improve the engine emissions, EGR can be applied
for NOx reduction. In addition, the use of a less amount of the main
injection is deemed benecial to reduce the extent of the diffusion
burning and in turn the NOx and soot emissions. For a constant

36

M. Zheng et al. / Fuel 142 (2015) 2837

engine load, the fuel quantity reduced from the main injection
needs to be re-distributed to the pilot injection. At the same time,
each pilot injection only allows a small quantity of fuel to avoid
potential wall impingement and/or premature early ignition.
Therefore, the multi-shot injection strategy deploys two early
pilots along with the main injection, and these injections are
450 ls@285CA, 450 ls@325CA, and 550 ls@358CA.
The cylinder pressure and heat release rate traces are shown in
Fig. 15 for the triple-shot injection tests at two EGR ratios (35% and
40%). The engine load is 10 bar IMEP. With the EGR application and
reduced main injection, the NOx emissions are substantially
reduced while the soot emissions are maintained below 1 FSN.
Under the tested conditions, the n-butanol combustion presents
desirable combustion stability (COVIMEP < 3%).

4. Conclusions
Engine experiments and detailed combustion analyses are conducted to study the LTC operation using direct injection of neat
n-butanol on a high compression ratio diesel engine. The singleshot injection and multi-pulse injection strategies are investigated
along with varied intake boost and EGR rates to enable LTC operation up to 10 bar IMEP while the engine runs at a medium speed of
1500 rpm. The conclusions based on the test results are summarized as follows:
1. In comparison to the diesel baseline, the single-shot injection of
n-butanol offers substantial benets of near-zero NOx and soot
emissions under the tested engine operating conditions, but the
combustion of n-butanol (as a low Cetane fuel) exhibits a narrower applicable window for the injection timing at medium
engine loads (67 bar IMEP) without the use of EGR; earlier
injection timings lead to excessively high maximum rates of
pressure rise that can become more problematic at high engine
loads, while retarded injection timings cause misre incidences
due to the low reactivity of n-butanol.
2. Under the tested engine speed and load, the use of EGR is effective to reduce combustion roughness even when the combustion phasing is maintained constant; however the EGR
addition further narrows the applicable injection timing window and deteriorates the HC and CO emissions while minor
soot penalties are observed.
3. By splitting a single-shot injection into a pilot injection and a
main injection, the multi-pulse injection strategy allows more
leverage in the ignition and combustion timing control, but
the main injection incurs diffusion burning, thereby leading to
moderately increased NOx and soot emissions.
4. For engine loads up to 10 bar IMEP, an improved injection strategy that deploys two early pilot injections and a reduced main
injection requires a moderate use of EGR for low exhaust emissions and offers desirable combustion controllability without
signicant penalties of engine efciency.

Acknowledgements
The research in the Clean Combustion Engine Laboratory at the
University of Windsor is sponsored by the BioFuelNet, NSERC-CRD,
CFI, OIT, AUTO21, NSERC-DG, NSERC-RTI, the Canada Research
Chair program, the University of Windsor, and the automotive
OEMs. The support by the China-Canada Joint Initiative on
Clean Automotive Transportation (2014DFG61320) is gratefully
acknowledged.

References
[1] Johnson DT, Taconi KA. The glycerin glut: options for the value-added
conversion of crude glycerol resulting from biodiesel production. Environ
Prog 2007;26:33848.
[2] Rahmat N, Abdullah AZ, Mohamed AR. Recent progress on innovative and
potential technologies for glycerol transformation in to fuel additives: a critical
review. Renew Sust Energy Rev 2010;14:9871000.
[3] Szwaja A, Naber JD. Combustion of n-butanol in a spark-ignition IC engine. Fuel
2010;89:157382.
[4] Yacoub Y, Bara R, Gautam M. The performance and emission characteristics of
C1C5 alcohol-gasoline blends with matched oxygen content in a singlecylinder spark ignition engine. P IMech (A) J Power Energy 1998;212:36379.
[5] Cooney C, Wallner T, McConnell S, et al., Effects of blending gasoline with
ethanol and butanol on engine efciency and emissions. In: The 2009 ASME
ICE spring conference ICES 200976155, Milwaukee, Wisconsin, USA; 2006.
[6] Rakopoulos DC, Rakopoulos CD, Hountalas DT, Kakaras EC, Kakaras EC,
Papagiannakis RG, et al. Investigation of the performance and emissions of a
bus engine operating on butanol/diesel fuel blends. Fuel 2010;89:278190.
[7] Rakopoulos CD, Dimaratos AM, Giakoumis EG, Rakopoulos DC. Investigating
the emissions during acceleration of a turbocharged diesel engine operating
with bio-diesel or n-butanol diesel fuel blends. Energy 2010;35:517384.
[8] Rakopoulos DC, Rakopoulos CD, Papagiannakis RG, Kyritsis DC. Combustion
heat release analysis of ethanol or n-butanol diesel fuel blends in heavy-duty
DI diesel engine. Fuel 2011;90:185567.
[9] Dogan O. The inuence of n-butanol/diesel fuel blends utilization on a small
diesel engine performance and emissions. Fuel 2011;90:246772.
[10] Ballesteros R, Hernandez JJ, Guillen-Flores J. Carbonyls speciation in a typical
European automotive diesel engine using bioethanol/butanol-diesel blends.
Fuel 2012;95:13645.
[11] Valentino G, Corcione FE, Iannuzzi SE, Serra S. Experimental study on
performance and emissions of a high speed diesel engine fueled with nbutanol diesel blends under premixed low temperature combustion. Fuel
2012;92:295307.
[12] Merola SS, Tornatore C, Lannuzzi SE, Marchitto L, Valentino G. Combustion
process investigation in a high speed diesel engine fueled with n-butanol
diesel blend by conventional methods and optical diagnostics. Renewable
Energy 2014;64:22537.
[13] Siwale L, Kristof L, Adam T, Bereczky A, Mbarawa M, Penninger A, et al.
Combustion and emission characteristics of n-butanol/diesel fuel blend in a
turbo-charged compression ignition engine. Fuel 2013;107:40918.
[14] Rakopoulos DC. Combustion and emissions of cottonseed oil and its bio-diesel
in blends with either n-butanol or diethyl ether in HSDI diesel engine. Fuel
2013;105:60313.
[15] Yilmaz N, Vigil M, Benalil K, Davis SM, Calva A. Effect of biodiesel-butanol fuel
blends on emissions and performance characteristics of a diesel engine. Fuel
2014;135:4650.
[16] Lin SL, Lee WJ, Lee CF, Wu YP. Reduction in emissions of nitrogen oxides,
particulate matter, polycyclic aromatic hydrocarbon by adding watercontaining butanol into diesel-fueled engine generator. Fuel 2012;93:36472.
[17] Chang YC, Lee WJ, Lin SL, Wang LC. Green energy: water-containing acetone
butanolethanol diesel blends fueled in diesel engines. Appl Energy
2013;109:18291.
[18] Chang YC, Lee WJ, Wu TS, Wu CY, Chen SJ. Use of water containing acetonebutanol-ethanol for NOx-PM trade-off in the diesel engine fueled with
biodiesel. Energy 2014;64:67887.
[19] Chen Z, Liu J, Wu Z, Lee CF. Effects of port fuel injection (PFI) of n-butanol and
EGR on combustion and emissions of a direct injection diesel engine. Energy
Convers Manag 2013;76:72531.
[20] Han X, Zheng M, Wang J. Fuel suitability for low temperature combustion in
compression ignition engines. Fuel 2013;109:33649.
[21] He BQ, Yuan J, Liu MB, Zhao H. Combustion and emission characteristics of an
n-butanol HCCI engine. Fuel 2014;115:75864.
[22] Kimura S, Aoki O, Ogawa H, Muranaka S, Enomoto Y. New combustion concept
for ultra-clean and high-efciency small DI diesel engines. SAE paper 1999.
1999-01-3681.
[23] Akihama K, Takatori Y, Inagaki K, Sasaki S, Dean AM. Mechanism of the
smokeless rich diesel combustion by reducing temperature. SAE paper 2001.
2001-01-0655.
[24] Shimazaki N, Tsurushima T, Nishimura T. Dual-mode combustion concept
with premixed diesel combustion by direct injection near top dead center. SAE
paper 2003. 2003-01-0742.
[25] Neely GD, Sasaki S, Huang Y, Leet JA, Stewart DW. New diesel emission control
strategy to meet US Tier 2 emissions regulations. SAE paper 2005. 2005-011091.
[26] Alriksson M, Rente T, Denbratt I. Low soot, low NOx in a heavy duty diesel
engine using high levels of EGR. SAE Paper 2005. 2005-01-3836.
[27] Kook S, Bae C, Miles P, Chio D, Picket LM. The inuence of charge dilution and
injection timing on low-temperature diesel combustion and emissions. SAE
paper 2005. 2005-01-3837.
[28] Noehre C, Anderson M, Johansson B, Hultqvist A. Characterization of partially
premixed combustion. SAE paper 2006. 2006-01-3412.
[29] Li T, Izumi H, Shudo T, Ogawa H, Okabe Y. Characterization of low temperature
diesel combustion with various dilution gases. SAE Trans J Engines
2007;116(3):12031.

M. Zheng et al. / Fuel 142 (2015) 2837

[30] Zheng M, Tan Y, Mulenga MC, Wang M. Thermal efciency analyses of diesel
low temperature combustion cycles. SAE Transactions 2007. 2007-01-4019.
[31] Ogawa H, Li T, Miyamoto N. Characteristics of low temperature and low
oxygen diesel combustion with ultra-high EGR. Int J Engine Res
2007;8(4):36578.
[32] Zheng M, Raj K. Implementation of multiple-pulse injection strategies to
enhance homogeneity for simultaneous low-NOx low-soot diesel combustion.
Int J Therm Sci 2009;48(9):182941.
[33] Li T, Okabe Y, Izumi H, Shudo T, Ogawa H. Dependence of ultra-high EGR and
low temperature diesel combustion on fuel properties. SAE paper 2006. 200601-3387.

37

[34] Kook S, Pickett LM. Effect of fuel volatility and ignition quality on combustion
and soot formation at xed premixing conditions. SAE Paper 2009. 2009-012643.
[35] Kalghatgi GT, Hildingsson L, Johansson B, Harrison A. Some effects of fuel
autoignition quality and volatility in premixed compression ignition engines.
SAE paper 2010. 2010-01-0607.
[36] Li T, Moriwaki R, Ogawa H, Kakizaki R, Murase M. Dependence of premixed
low-temperature diesel combustion on fuel ignitability and volatility. Int J
Engine Res 2012;13(1):1427.