Fuel 230 (2018) 440–451

Contents lists available at ScienceDirect

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

Full Length Article

High efficiency ethanol-diesel dual-fuel combustion: A comparison against T

conventional diesel combustion from low to full engine load
⁎
Vinícius B. Pedrozo , Ian May, Wei Guan, Hua Zhao
Centre for Advanced Powertrain and Fuels Research (CAPF), Brunel University London, Kingston Lane, Uxbridge, Middlesex UB8 3PH, United Kingdom

A R T I C LE I N FO A B S T R A C T

Keywords: Comparisons between dual-fuel combustion and conventional diesel combustion (CDC) are often performed
Ethanol using different engine hardware setups, exhaust gas recirculation rates, as well as intake and exhaust manifold
Low carbon fuel pressures. These modifications are usually made in order to curb in-cylinder pressure rise rates and meet exhaust
Dual-fuel combustion emissions targets during the dual-fuel operation. To ensure a fair comparison, an experimental investigation into
Diesel engine
dual-fuel combustion has been carried out from low to full engine load with the same engine hardware and
Greenhouse gas emissions
identical operating conditions to those of the CDC baseline. The experiments were executed on a single cylinder
heavy-duty diesel engine at a constant speed of 1200 rpm and various steady-state loads between 0.3 and
2.4 MPa net indicated mean effective pressure (IMEP). Ethanol was port fuel injected while diesel was direct
injected using a high pressure common rail injection system. The start of diesel injection was optimised for the
maximum net indicated efficiency in both combustion modes. Varied ethanol energy fractions and different
diesel injection strategies were required to control the in-cylinder pressure rise rate and achieve highly efficient
and clean dual-fuel operation. In terms of performance, dual-fuel combustion attained higher net indicated
efficiency than the CDC mode from 0.6 to 2.4 MPa IMEP, with a maximum value of 47.2% at 1.2 MPa IMEP. The
comparison also shows the use of ethanol resulted in 26% to 90% lower nitrogen oxides (NOx) emissions than
the CDC operation. At the lowest engine load of 0.3 MPa IMEP, the dual-fuel operation led to simultaneous low
NOx and soot emissions at the expense of a relatively low net indicated efficiency of 38.9%. In particular, the
reduction in NOx emissions introduced by the utilisation of ethanol has the potential to decrease the engine
running costs via lower consumption of aqueous urea solution in the selective catalyst reduction system.
Moreover, the dual-fuel combustion with a low carbon fuel such as ethanol is an effective means of decreasing
the use of fossil fuel and associated greenhouse gas emissions.

1. Introduction However, there is a lot of concern over the greenhouse gas (GHG)
emissions produced from the combustion of diesel and other fossil fuels
Heavy-duty (HD) vehicles are typically powered by diesel engines [1]. This is due to a recent increase in the atmospheric concentration of
due to their cost-effectiveness and high fuel conversion efficiency. GHGs such as carbon dioxide (CO2) [2], which can cause irreversible

Abbreviations: ATDC, after firing top dead centre; CA10, crank angle of 10% cumulative heat release; CA10–CA90, combustion duration or 10–90% cumulative heat release; CA50, crank
angle of 50% cumulative heat release; CA90, crank angle of 90% cumulative heat release; CAD, crank angle degree; CDC, conventional diesel combustion; CH4, methane; CO, carbon
monoxide; CO2, carbon dioxide; CO2eq, CO2 equivalent; COV_IMEP, coefficient of variation of IMEP; COV_Pmax, coefficient of variation of Pmax; DAQ, data acquisition; ECR, effective
compression ratio; ECU, engine control unit; EF, ethanol energy fraction; EGR, exhaust gas recirculation; EGT, exhaust gas temperature; FID, flame ionisation detector; FSN, filter smoke
number; GHG, greenhouse gas; GWP, global warming potential; HC, hydrocarbons; HD, heavy-duty; HRR, apparent net heat release rate; iEGR, internal EGR; iLUC, indirect land use
change; IMEP, net indicated mean effective pressure; ISCO, net indicated specific emissions of CO; ISCO2, net indicated specific emissions of CO2; ISHC, net indicated specific emissions of
actual unburnt HC; ISNOx, net indicated specific emissions of NOx;LHVdiesel , lower heating value of diesel; LHVethanol,, lower heating value of ethanol; LHVfuel,, lower heating value; ṁ air ,
fresh air mass flow rate; ṁ diesel , diesel mass flow rate; ṁ ethanol , ethanol mass flow rate; ṁ urea , estimated consumption of aqueous urea solution in the SCR system; MCO2 , normalised molar
mass of CO2; Mdiesel , normalised molar mass of diesel;Methanol , normalised molar mass of ethanol; Mfuel , normalised molar mass; MFB, mass fraction burnt; N2O, nitrous oxide;
Net Indicated Eff .SCRcorr. , SCR corrected net indicated efficiency; NOx, nitrogen oxides; O2, Oxygen; Pind , net indicated power; PFI, port fuel injector; Pmax, peak in-cylinder gas pressure;
PRR, pressure rise rate; RON, research octane number; SCR, selective catalyst reduction; SOC, start of combustion; SOI_main, actual start of main diesel injection; SOI_mai–SOC, ignition
delay; SOI_pre, actual start of diesel pre-injection; TDC, firing top dead centre; TTW, tank-to-wheels; VVA, variable valve actuation; WTT, well-to-tank; WTTdiesel , WTT CO2eq emissions for
fossil diesel; WTTethanol , WTT CO2eq emissions for ethanol; WTW, well-to-wheels; %Cfuel , carbon mass content; %Hfuel , hydrogen mass content; %Ofuel , carbon mass content; γ, ratio of
specific heats; ρfuel , density; ρdiesel , diesel density; ρethanol , ethanol density; Φ, global fuel/air equivalence ratio
⁎
Corresponding author.
E-mail address: [email protected] (V.B. Pedrozo).

https://doi.org/10.1016/j.fuel.2018.05.034
Received 19 February 2018; Received in revised form 5 May 2018; Accepted 8 May 2018
0016-2361/ © 2018 Elsevier Ltd. All rights reserved.
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

climate change and negatively impact the health of living organisms triggered by direct injections of diesel [6,35]. It should be noted that
across the globe [1]. the use of a low carbon fuel like ethanol [36–39] can help decrease the
In 2010, HD vehicles were responsible for approximately 34% of the dependence on fossil fuels and minimise GHG emissions from the global
GHGs emitted by the global transport sector and 46.5% of the road transport sector [40].
transport CO2 emissions [3]. This disproportionate contribution is Despite the advantages of dual-fuel operation, it is challenging to
highlighted by the fact the HD fleet represents only 11% of the world obtain direct comparisons against the CDC mode from low to high en-
motor vehicles [4]. Substantial and continuous reduction in fossil fuel gine loads (e.g. above 2.0 MPa IMEP). This is often due to modifications
energy use and GHG emissions must be achieved in order to address the in engine hardware and/or test conditions that are used to control the
transport sector’s impact on the environment. emissions of NOx and the in-cylinder pressure rise rates from dual-fuel
Additional cause for concern is that CDC is prone to a wide range of combustion. These alterations typically include the use of a different
local in-cylinder gas temperatures and fuel/air equivalence ratios that piston design and/or compression ratio [41,42] as well as changes in
can lead to the formation of noxious emissions, such as NOx and soot the levels of exhaust gas recirculation [43].
[5,6]. NOx emissions are mainly formed in near-stoichiometric high This study aims to explore the potential of ethanol–diesel dual-fuel
temperatures regions close to the diesel diffusion flame [7]. Soot for- combustion to achieve high fuel conversion efficiency and low exhaust
mation occurs in high fuel/air equivalence ratio and intermediate emissions using the same combustion system and identical engine
temperature zones within the diesel spray [8,9]. These pollutants are testing conditions to those of the CDC baseline. To the best of our
linked to premature deaths caused by cardiovascular and respiratory knowledge, this is the first attempt to experimentally compare the
diseases [10,11]. controllability, emissions, and fuel conversion efficiency of the above-
Stringent fuel conversion efficiency and exhaust emissions regula- mentioned combustion modes from low to full engine load (e.g.
tions have been implemented to limit the levels of GHG and noxious 0.3–2.4 MPa IMEP). In addition to this, practical considerations have
emissions from HD vehicles [12–15]. Manufacturers are incorporating been raised and the potential CO2 reduction has been discussed on both
costly engine design elements [16–20] and aftertreatment technologies a tank-to-wheels and well-to-wheels basis [37,44].
[21,22] to comply with the aforementioned emissions standards and The investigation was performed on a single cylinder HD diesel
GHG targets [12,13]. The use of improved selective catalyst reduction engine at a steady-state speed of 1200 rpm. The diesel injection timings
(SCR) systems for NOx mitigation, flexible and high pressure diesel and the number of injections per cycle were optimised in both the
injection equipment, as well as high efficiency turbocharging and air combustion modes in order to maximise the fuel conversion efficiency,
handling systems are some of the ways in which this is being achieved. which was given by the net indicated efficiency. Moreover, dual-fuel
Reaching a balance between engine running costs and exhaust operation was carried out using ethanol energy fractions that achieved
emissions can represent a challenge for HD engine manufacturers the highest net indicated efficiency with minimal NOx and soot emis-
[16,17]. Both in-cylinder and aftertreatment measures are considered sions, as determined in our previous studies [29–31,45,46].
and are often linked. An improvement of 1% in fuel conversion effi-
ciency can increase the levels of engine-out NOx from 10 g/kWh to 2. Experimental setup
14 g/kWh [18]. This adversely affects the total cost of ownership due to
a higher consumption of aqueous urea solution in the SCR system 2.1. Experimental facilities
[23–26]. On the other hand, CDC operation with very low engine-out
NOx emissions can result in low fuel conversion efficiency and ex- A schematic diagram of the single cylinder HD engine experimental
cessive levels of soot due the different formation mechanisms [27,28]. setup is shown in Fig. 2. A Froude Hofmann AG150 eddy current dy-
Previous studies into dual-fuel compression ignition combustion namometer was used to absorb the power produced by the engine.
have demonstrated that the strategy has the potential to resolve these Fresh intake air was supplied to the engine via an AVL 515 sliding vanes
issues, increasing the fuel conversion efficiency while decreasing both compressor with a closed loop control for the boost pressure. A throttle
the NOx and soot emissions [6,29–32]. This has been attributed to si- valve located upstream of a surge tank provided fine control over the
multaneous reductions in local fuel/air equivalence ratio and tem- intake manifold pressure. The fresh air mass flow rate (ṁ air ) was
perature, shorter combustion duration, and lower heat transfer losses measured with an Endress + Hauser Proline t-mass 65F thermal mass
[6,32]. flow meter.
Fig. 1 shows an example of a dual-fuel system, which can be Another surge tank was installed in the exhaust manifold to damp
achieved by the installation of a port fuel injection system of a low out pressure fluctuations prior to the exhaust gas recirculation (EGR)
reactivity fuel such as gasoline [32], ethanol [33], or natural gas [34] circuit. An electronically controlled butterfly valve located downstream
on a diesel engine. The ignition of the premixed charge is generally of the exhaust surge tank was used to set the required back pressure
(e.g. exhaust manifold pressure). High-pressure loop cooled external
EGR was supplied to the engine intake system by opening a pulse width
modulation-controlled EGR valve. Boosted intake air and external EGR
temperatures were controlled using water cooled heat exchangers.

2.2. Engine specifications

Base hardware specifications are outlined in Table 1. The combus-

tion system consisted of a 4-valve swirl-oriented cylinder head and a
stepped-lip piston bowl design with a geometric compression ratio of
16.8.
The diesel introduction was controlled via a dedicated engine con-
trol unit (ECU) with the ability to support up to three injections per
cycle. The intake valve lift profile was adjusted via a lost-motion vari-
able valve actuation (VVA) system based on a normally open high-
speed solenoid valve assembly and a special intake cam design [47].
Fig. 1. Schematic diagram of a dual-fuel engine with direct injections of diesel Coolant and oil pumps were not coupled to the engine and were
and port fuel injection of ethanol. driven by separate electric motors. Engine coolant and oil temperatures

441
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

Fig. 2. Schematic diagram of the engine experimental setup.

Table 1
Single cylinder HD engine specifications.
Parameter Value

Displaced volume 2.026 dm3

Stroke 155 mm
Bore 129 mm
Connecting rod length 256 mm
Number of valves 4
Piston type Stepped-lip bowl
Geometric compression ratio 16.8
Peak in-cylinder pressure (Pmax) limitation 18 MPa
Diesel injection system Bosch common rail, injection pressure of 50–220 MPa, 8 holes with nominal diameter of
0.176 mm, included spray angle of 150°
Ethanol injection system PFI Marelli IWP069, included spray angle of 15°

were set to 353 ± 3 K. The oil pressure was held at 450 ± 10 kPa Table 2
throughout the experiments. Fuel properties.
Property Diesel Ethanol
2.3. Fuel properties and delivery
Product name Red diesel (gas oil) Absolute ethanol 100
Standard/specification BS 2869 Class A2 Anhydrous ethanol
The relevant properties of the fuel used in this work are listed in
Density at 293 K (ρfuel) 0.827 kg/dm3 0.790 kg/dm3 [50]
Table 2. The diesel fuel was supplied to the engine using a high pressure Cetane number > 45 n/a
common rail injection system. Two Endress + Hauser Promass 83A Research octane number (RON) n/a [51] ∼107 [51]
Coriolis flow meters were used to determine the diesel mass flow rate Alcohol content in volume n/a 99.9%
(ṁ diesel ) by measuring the total fuel supplied to and from the diesel high Water content < 0.20 g/kg [52] 1.7 g/kg [50]
Sulphur content < 0.01 g/kg n/a
pressure pump and injector.
Heat of vaporisation 270 kJ/kg [51] 840 kJ/kg [51]
While the use of two fuel flow meters is not ideal due to the com- Carbon mass content (%Cfuel) 86.6% 52.1% [51]
pounding of measurement error, this setup was the only option for Hydrogen mass content (%Hfuel) 13.2% 13.1% [51]
regulating the backpressure on the diesel injector return line. Tests Oxygen mass content (%Ofuel) 0.2% 34.8% [51]
Normalised molecular composition CH1.825O0.0014 CH3O0.5
were carried out to validate the consistency of the diesel fuel mea-
Lower heating value (LHVfuel) 42.9 MJ/kg 26.9 MJ/kg [51]
surements with two fuel flow meters, achieving a combined uncertainty
[48,49] smaller than ± 1.7%.
In order to enable dual-fuel operation, an ethanol fuel injection
system was designed and fitted to the engine. Ethanol was injected Endress + Hauser Proline Promass 80A Coriolis flow meter, allowing
through a port fuel injector (PFI) installed in the intake manifold. An in- for measurements with an accuracy of ± 0.15% of reading. The injec-
house injector driver controlled the injector pulse width, which was tion pressure was continuously monitored by a pressure transducer, so
adjusted according to the desired ethanol energy fraction. that a constant relative pressure of 300 kPa could be maintained across
The ethanol mass flow rate (ṁ ethanol ) was measured using an the injector.

442
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

2.4. Exhaust emissions measurements and analysis Crank angle based in-cylinder pressure traces were averaged over
200 consecutive cycles for each operating point and used to calculate
Gaseous emissions such as NOx, CO2, carbon monoxide (CO), the IMEP and the apparent net heat release rate (HRR). The pressure
oxygen (O2), and unburnt hydrocarbon (HC) were taken with a Horiba rise rate (PRR) was represented by the average of the maximum pres-
MEXA-7170 DEGR emissions analyser. The EGR rate was determined by sure variations of 200 cycles of cylinder pressure versus crank angle.
calculating the ratio of the intake to the exhaust manifold CO2 con- Combustion and in-cylinder flow stability were monitored by the
centration measured by the same emissions analyser. A high pressure coefficient of variation of IMEP (COV_IMEP) and Pmax (COV_Pmax) over
module allowed for high-pressure sampling upstream of the back the sampled cycles.
pressure valve while a heated line was used to prevent water con- Since the absolute value of the heat released is not as important to
densation. this study as the bulk shape of the curve with respect to crank angle, a
The measurement of unburnt HC was performed on a wet basis by constant ratio of specific heats (γ) of 1.33 was assumed throughout the
the Horiba’s heated flame ionisation detector (FID). However, the HC engine cycle. The mass fraction burnt (MFB) was given by the ratio of
emissions measured with the FID can lead to misinterpretation of the the integral of the HRR and the maximum cumulative heat release.
results due to the relative insensitivity of the device towards alcohols Combustion phasing was determined by the crank angle of 50% (CA50)
and aldehydes [53,54]. Therefore, the FID response was corrected by cumulative heat release. Combustion duration was represented by the
the method developed by Kar and Cheng [53] with an updated response period of time between the crank angles of 10% (CA10) and 90%
factor of 0.68 for the oxygenated organic species resultant from ethanol (CA90) cumulative heat release.
combustion [54]. This procedure has been reported in our previous A current probe was used to acquire the electric current signal sent
work [31] and allows for the determination of the actual unburnt HC from the ECU to the diesel injector solenoid. The signal was corrected
emissions. by adding the respective energising time delay, which was previously
The concentration of NOx, CO, and unburnt HC in the exhaust measured in a constant volume chamber [60]. The resulting diesel in-
stream were converted from parts per million (ppm) to net indicated jector current signal allowed for the determination of the actual start of
specific emissions using the methodology described in the Regulation diesel injection.
number 49 of the Economic Commission for Europe of the United Ignition delay was defined as the period of time between the actual
Nations [52]. NOx and CO were converted to a wet basis by applying a start of main diesel injection (SOI_main) and the start of combustion
correction factor for the raw exhaust gas according to the in-cylinder (SOC), set to 0.3% MFB point of the averaged cycle. After the calcula-
fuel mixture composition. tion of the combustion characteristics (e.g. CA50) and ignition delay,
An AVL 415SE smoke meter was used to measure the concentration the average in-cylinder pressure and the resulting HRR were smoothed
of black carbon containing soot. The measurements were reported on a using a Savitzky-Golay filter.
filter smoke number (FSN) basis, as previous studies [55–58] have re- Net indicated efficiency was determined by calculating the ratio of
ported that low levels of soot from dual-fuel combustion do not corre- the net indicated power to the rate of fuel energy supplied to the en-
late well with the amount of particle mass produced. The actual particle gine. Combustion efficiency calculations were based on the emissions
mass is expected to be higher, particularly at low engine loads, because products not fully oxidised during the combustion process except soot
semi-volatile organic species can condense on small solid cores (e.g. using the method reported in [46].
carbonaceous) [59].

2.5. Data acquisition 3. Methodology

The in-cylinder pressure was measured by a Kistler 6125C piezo- Fig. 3 shows the location of the test points over an estimated speed
electric pressure sensor coupled with an AVL FI Piezo charge amplifier. and load map of a HD diesel engine. Testing was carried out under a
Intake and exhaust manifold pressures were measured by two Kistler steady-state engine speed of 1200 rpm over a range of loads from 0.3 to
4049A water cooled piezoresistive absolute pressure sensors coupled to 2.4 MPa IMEP. A PRR of 2.0 MPa/CAD and a Pmax of 18 MPa were
Kistler 4622A amplifiers. Temperatures and pressures at relevant lo- considered as the upper bounds for calibration. Stable engine operation
cations were measured by K-type thermocouples and pressure gauges, was quantified by COV_IMEP values < 5%.
respectively.
Two National Instruments data acquisition (DAQ) cards and a per-
sonal computer were used to acquire the signals from the measurement
device. A USB-6251 high speed DAQ card received the crank angle
resolved data synchronized with an optical encoder of 0.25 crank angle
degrees (CAD) resolution. A USB-6210 low speed DAQ card acquired
the low frequency engine operation conditions. These data were dis-
played live by an in-house developed DAQ program and combustion
analyser.

2.6. Data analysis

A relevant parameter for the dual-fuel operation was the ethanol

energy fraction (EF), which was defined as the ratio of the energy
content of the ethanol to the total fuel energy supplied by
ṁ ethanol LHVethanol
EF =
(ṁ ethanol LHVethanol ) + (ṁ diesel LHVdiesel ) (1)
The global fuel/air equivalence ratio (Φ) was given by
(14.5ṁ diesel + 9.0ṁ ethanol ) Fig. 3. Experimental test points over an estimated HD diesel engine speed-load
ϕ=
ṁ air (2) map.

443
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

Table 3
Operating conditions for the CDC and ethanol–diesel dual-fuel operation from low to full engine load at 1200 rpm.
Engine load Intake manifold pressure Exhaust manifold pressure Intake manifold air temp. EGR rate Diesel injection pressure
(CDC) (Dual-fuel)

MPa IMEP kPa kPa K % MPa MPa

0.3 115 125 307 25 105 50

0.6 125 135 310 25 125 90
0.9 155 165 315 25 140 110
1.2 190 200 319 25 155 125
1.5 230 240 324 25 170 140
1.8 260 270 324 20 190 160
2.4 300 310 323 11 220 190

The experiments were performed with a pressure-based effective operation via improved diesel atomisation and enhanced the fuel–air
compression ratio (ECR) of 16.8 [46]. The ECR was calculated using the mixing process. The use of the same diesel injection pressure on the
effective in-cylinder volume at intake valve closing, which was ob- dual-fuel mode could have caused overmixing at low engine loads and
tained from the intersection of the intake manifold pressure and an excessive PRR at medium and high engine loads.
extrapolated polytropic compression curve fitted to the experimental All comparisons were carried out for the cases that attained the
in-cylinder pressure [61,62]. The expansion ratio remained constant as highest net indicated efficiencies after sweeps of diesel injection tim-
a result of the fixed exhaust camshaft timing. ings. Additionally, the diesel injection strategy (i.e. number of diesel
Table 3 summarises the test conditions for the CDC and etha- injections per cycle) was optimised and varied as the engine load was
nol–diesel dual-fuel operating modes. The intake manifold pressure set increased. In the dual-fuel mode, the ethanol energy fraction was also
point was taken from a Euro V compliant multi-cylinder HD diesel optimised for minimum NOx and soot emissions, as supported by our
engine in order to provide a sensible starting point, since an external previous dual-fuel studies [29–31,45,46]. A maximum EF of 0.79 was
boosting device was used in place of a turbocharger. The exhaust achieved at 1.2 MPa IMEP. Advanced dual-fuel combustion control
manifold pressure was varied to maintain a constant pressure differ- strategies such as internal exhaust gas recirculation (iEGR) [31] and
ential across the cylinder of 10 kPa. This difference of 10 kPa between Miller cycle [46] were not explored in this study as they would require
exhaust and intake manifold pressures was used to simulate the different operating conditions.
pumping losses typically associated with a turbocharger and to drive
the requested amounts of EGR. 4. Results and discussion
The EGR rate was held constant at a given engine load in order to
ensure a fair comparison between CDC and dual-fuel operation. The 4.1. Overview of the load sweep
EGR rate was limited at 25% between 0.3 and 1.5 MPa IMEP to avoid
excessive smoke and a decrease in net indicated efficiency. At 1.8 and Fig. 4 depicts the effect of engine load on both the operating modes.
2.4 MPa IMEP, the EGR rate was reduced to 20% and 11%, respectively. CDC operation was characterised by longer mixing-controlled com-
This was required to achieve lean and efficient high load operation bustion phase as the load was increased. This was attributed to longer
using the same levels of boost pressure as the multi-cylinder engine. diesel injection periods and increased amount of fuel, which limited the
Diesel injection pressures were set to be 30 to 55 MPa higher in the fuel vapour-air mixing process [51,63]. The optimum CA50 in CDC
CDC mode than those in the dual-fuel combustion due to the relatively mode varied as the engine load was increased, allowing for more ad-
higher diesel flow rates and longer injection durations at a given engine vanced burn rates at mid-loads and delayed combustion events at high
load. This was necessary to minimise soot emissions from the CDC loads. The reasons behind this are described in the next subsection.

Fig. 4. The effect of engine load on CDC and ethanol–diesel dual-fuel operation at 1200 rpm.

444
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

Fig. 5. Optimum ethanol energy fraction for varied engine loads at 1200 rpm.

The dual-fuel operation led to higher peak heat release than the CDC
mode at all engine loads except 0.3 MPa IMEP. This required different
diesel injection strategies and eventually later combustion process in
order to control the PRR as the engine load was increased. The com-
bustion was triggered by and initiated after the start of diesel injection
at low and medium load operations between 0.3 and 1.5 MPa IMEP.
Higher compression pressures and temperatures accelerated the auto-
ignition of the premixed ethanol fuel prior to the diesel injection at high
engine loads of 1.8 and 2.4 MPa IMEP.
Fig. 5 shows the optimum EF was rapidly reduced from 0.76 to 0.25
when increasing the engine load from 1.5 to 1.8 MPa IMEP. This was
necessary in order to minimise the PRR associated with the early au-
toignition of ethanol. It is important to highlight that modifications in
the engine hardware (e.g. lower ECR via Miller cycle) and/or test
procedure (e.g. lower intake manifold air temperature) can increase the
maximum EF at higher loads [46].

4.2. Combustion control

Fig. 6 shows the actual start of diesel pre-injection (SOI_pre),

SOI_main, and in-cylinder pressure characteristics for optimised CDC
and dual-fuel operation. In the CDC mode, a 3 mm3 diesel pre-injection
with a constant dwell time of 1 ms was used to reduce the levels of PRR
[45] between the engine loads of 0.3 MPa IMEP and 1.5 MPa IMEP. The
lower PRR were associated with the shorter ignition delay produced by
the combustion of the diesel pre-injection, which likely formed of a hot
and reactive mixture prior to the main diesel injection [64].
At high engine loads of 1.8 and 2.4 MPa IMEP, relatively shorter
ignition delays introduced by lower EGR rates and higher in-cylinder
pressures and temperatures allowed for the use of a single diesel in-
jection near firing top dead centre (TDC). The maximum SOI_main
advance was limited by the Pmax while the PRR was maintained within
the limit of 2.0 MPa/CAD.
In the dual-fuel operation, the combination of an early single diesel
injection at about -36 CAD after top dead centre (ATDC) and ethanol Fig. 6. Diesel injection timings and combustion characteristics for optimised
energy fractions of 0.56 and 0.65 allowed for long ignition delays CDC and ethanol–diesel dual-fuel operation at 1200 rpm.
(SOI_main–SOC) and better mixture preparation at 0.3 and 0.6 MPa
IMEP. This enhanced the combustion process via a more progressive peak typically observed with a late single diesel injection strategy [45].
and probably sequential combustion from high to low reactivity regions Despite the controlled levels of PRR, the diesel injection timings were
[8]. This has also been identified in computational simulations per- delayed by up to 10.5 CAD when compared to those of the CDC op-
formed by Desantes et al. [65] and is supported by the low levels of eration, helping to decrease the Pmax levels.
PRR. However, the Pmax was increased when compared to that of the At 1.8 and 2.4 MPa IMEP, the premixed ethanol fuel autoignited
CDC operation due to earlier CA50 and shorter combustion for the dual- prior to the diesel injection. Low ethanol energy fractions and a single
fuel mode at these particular loads (see Fig. 7). diesel injection near TDC were used to control the burn rate as well as
At mid-loads between 0.9 and 1.5 MPa IMEP, less partially premixed the resulting PRR and Pmax. The introduction of a diesel pre-injection
diesel fuel could be used in order to prevent an early ignition of the in- would increase the PRR levels at these loads due to simultaneous and
cylinder charge. Therefore, the mass of the diesel was divided into two early combustion of the ethanol and pre-injected diesel fuel.
direct injections using the same strategy employed in the CDC cases.
The injection of a small amount of diesel prior to the SOI_main was
essential to mitigate excessive PRR. This was a result of a shorter
SOI_main–SOC period and elimination of the premixed combustion

445
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

the engine load was increased in order to control the PRR within the
limit of 2.0 MPa/CAD. At high loads of 1.8 and 2.4 MPa IMEP, the CA50
and CA90 were similar for both the combustion modes due to the Pmax
limitation of 18 MPa and lower ethanol energy fractions used in the
dual-fuel mode.
In general, an increase in engine load led to later CA90 and longer
CA10–CA90 period as a result of the higher fuel flow rates. The higher
degree of premixed combustion in the dual-fuel mode was likely the
cause for the relatively earlier CA90 and faster CA10–CA90 period
between 0.3 and 1.5 MPa IMEP. Nonetheless, the early ignition of the
ethanol fuel produced longer burn rates than the CDC operation at 1.8
and 2.4 MPa IMEP.
In terms of combustion stability, the mixing-controlled combustion
of the CDC operation effectively decreased the COV_IMEP and
COV_Pmax to 0.5% as the engine load was increased to 2.4 MPa IMEP. In
the dual-fuel mode, later CA50 and a more premixed combustion
yielded higher levels of COV_IMEP than the CDC case between 0.9 and
2.4 MPa IMEP. The increase in COV_IMEP was proportional to the EF. In
addition, the dual-fuel operation resulted in higher COV_Pmax than the
CDC mode at all engine loads except 0.3 MPa IMEP. Nevertheless, the
COV_IMEP and COV_Pmax could be controlled between 1.0% and 3.0%.

4.4. Engine-out emissions

Fig. 8 shows the engine-out emissions for the optimum cases over a
sweep of load. An EGR rate of 25% was used in order to minimise NOx
emissions at engine loads up to 1.5 MPa IMEP. This allowed for CDC
operation with net indicated specific emissions of NOx (ISNOx) of
3.9 g/kWh, on average, between 0.3 and 1.5 MPa IMEP. The use of
lower EGR rates of 20% and 11% increased the combustion tempera-
tures at 1.8 and 2.4 MPa IMEP, yielding higher ISNOx of 4.4 and 5.7 g/
kWh, respectively.
Alternatively, the optimised dual-fuel operation achieved lower
ISNOx than the CDC mode at all engine loads. This was linked to the
premixed ethanol fuel, which likely decreased the amount of in-cy-
linder regions of high combustion temperature. Reductions in NOx
emissions varied from 26% at 2.4 MPa IMEP up to 90% at 0.3 MPa
IMEP for ethanol energy fractions of 0.19 and 0.56, respectively.
The lowest levels of ISNOx were attained at 0.3 and 0.6 MPa IMEP
due to longer ignition delays and a more homogenous combustion
process when compared to the other dual-fuel cases with diesel injec-
tions closer to TDC. NOx emissions were decreased when increasing the
engine load from 0.9 to 1.5 MPa IMEP because of a later optimum CA50
and thus lower combustion temperatures. At high engine loads of 1.8
and 2.4 MPa IMEP, the ethanol autoignition process and shorter diesel
mixing-controlled combustion helped reduce the peak in-cylinder gas
temperatures [46], decreasing the ISNOx when compared with those of
the CDC operation.
Fig. 7. Heat release characteristics for optimised CDC and ethanol–diesel dual- Smoke levels from dual-fuel combustion were maintained com-
fuel operation at 1200 rpm. paratively low despite the use of lower diesel injection pressures than
the CDC operation (see Section 3). Early diesel injections and high EF
4.3. Heat release analysis curbed fuel rich combustion and limited soot emissions to 0.05 FSN at
0.3 and 0.6 MPa IMEP. However, the levels of soot were increased to
Fig. 7 depicts the heat release characteristics for the CDC and dual- 0.235 FSN at 1.5 MPa IMEP. This was likely a result of higher local fuel/
fuel operation. The optimum CA50 for the maximum net indicated ef- air equivalence ratio and reduced combustion temperatures introduced
ficiency was initially advanced and then retarded in the CDC mode. The by the shorter ignition delay and later CA50. In the CDC mode, higher
advance in the CA50 when increasing the engine load from 0.3 to diesel injection pressures and in-cylinder gas temperatures helped re-
0.6 MPa IMEP was likely linked to the longer CA10–CA90 period and duce soot emissions as the engine load was increased.
relatively lower heat transfer losses obtained at 0.6 MPa IMEP. The Net indicated specific emissions of CO (ISCO) and unburnt HC
CA50 delay at high load operations of 1.8 and 2.4 MPa IMEP was as- (ISHC) increased significantly in the dual-fuel combustion when com-
sociated with the peak in-cylinder pressure limitation. Additionally, pared to the CDC operation. This was possibly a result of premixed fuel
lower levels of EGR and higher combustion temperatures shortened the trapped in the crevice volumes of the stock diesel piston as well as
CA10–CA90 period and probably increased heat transfer losses of the lower local in-cylinder gas temperatures [6].
CDC operation at these high load conditions. High levels of ISCO and ISHC were measured for the dual-fuel op-
In comparison, the dual-fuel operation often required later CA50 as eration at 0.3 MPa IMEP. This can be attributed to excessively low
combustion temperatures and overly lean regions that did not release

446
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

Fig. 8. Engine-out emissions for optimised CDC and ethanol–diesel dual-fuel Fig. 9. Engine performance for optimised CDC and ethanol–diesel dual-fuel
operation at 1200 rpm. operation at 1200 rpm.

enough heat in order to effectively oxidise the fuel [6]. At 1.8 and the CDC cases as supported by the higher levels of ISCO and ISHC in
2.4 MPa IMEP, the use of lower EF as well as lower EGR rates likely Fig. 8. At the lowest load of 0.3 MPa IMEP, a relatively low combustion
increased combustion temperatures, decreasing CO and unburnt HC efficiency of 88.7% adversely affected the performance of the dual-fuel
emissions. operation, limiting the net indicated efficiency to 38.9% and the EGT to
463 K. This can represent a challenge for HD engine manufacturers due
to a reduction in the effectiveness of the oxidation catalyst in reducing
4.5. Engine performance CO and unburnt HC emissions [66,67].
In-cylinder control strategies such as intake throttling and iEGR can
Fig. 9 depicts the engine performance metrics for optimised CDC help increase the EGT while simultaneously minimising the levels of
and ethanol–diesel dual-fuel operation. The global fuel/air equivalence ISCO and ISHC at light engine loads [31]. As the engine load was in-
ratio (Φ) of the dual-fuel combustion was either comparable or lower creased, higher Φ and local in-cylinder gas temperatures helped
than that of the CDC mode at a given engine load. This was attributed to maintain the combustion efficiency above 96.3% despite the use of
minor variations in the intake air flow rate (within 3% and not shown ethanol energy fractions of up to 0.79.
for the sake of brevity) and improvements in net indicated efficiency. Finally, ethanol–diesel dual-fuel combustion resulted in higher net
Differences in LHVfuel and total fuel flow rate possibly balanced out indicated efficiencies than the CDC operation between 0.6 and 2.4 MPa
changes in stoichiometric air/fuel ratio. IMEP. A peak net indicated efficiency of 47.2% was attained at 1.2 MPa
Higher exhaust gas temperatures (EGT) were measured as engine IMEP and represented an increase of 4.4% over the 45.2% of the CDC
load was increased due to later CA90 and higher levels of fuel energy mode. The ethanol autoignition process likely helped decrease the
supplied. However, the dual-fuel operation produced EGTs up to 20 K combustion temperatures and thus the heat transfer losses [51], as
lower than those of the corresponding CDC case. This was perhaps a supported by the NOx reduction in Fig. 8.
result of a more homogenous and lower temperature combustion pro- The use of a late CA50 at 1.5 MPa IMEP and relatively low EF at 1.8
cess for an engine operation with premixed ethanol fuel [31,46]. and 2.4 MPa IMEP limited improvements in the net indicated efficiency
The dual-fuel mode also yielded lower combustion efficiencies than

447
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

of the dual-fuel operation, but were necessary in order to control the due to a combination of a higher total fuel flow rate (e.g. increased
PRR below 2.0 MPa/CAD. The maximum net indicated efficiency production of NOx emissions in g/h) with reductions in the EGR rate at
achieved by the CDC operation was 45.7% at 1.5 MPa IMEP. 1.8 and 2.4 MPa IMEP.
The lower urea consumption in the dual-fuel mode allowed for
4.6. Additional practical considerations higher Net Indicated Eff .SCRcorr.between 0.6 and 2.4 MPa IMEP. The
maximum Net Indicated Eff .SCRcorr.of 46.5% was achieved at 0.6 MPa
Additional practical aspects for ethanol–diesel dual-fuel operation IMEP and represented a relative increase of 8.4% over the CDC mode.
were assessed in order to evaluate whether the combustion strategy can Lowered combustion efficiency limited the Net Indicated Eff .SCRcorr..of
be successfully used in a Euro VI HD engine. The analysis focused on the the dual-fuel mode at 0.3 MPa IMEP, despite the low engine-out NOx of
total fuel flow rate, estimated consumption of aqueous urea solution in 0.4 g/kWh and ṁ urea = 0 .
the SCR system (ṁ urea ) to meet the Euro VI NOx limit of 0.4 g/kWh, and These improvements can reduce the engine running costs depending
SCR corrected net indicated efficiency (Net Indicated Eff .SCRcorr.). The on the volumetric price ratio between ethanol and diesel fuel [45] as
methodology for the calculation of these performance metrics has been well as the cost of aqueous urea solution. Nevertheless, the im-
described in our previous study [45], where ṁ urea was estimated at 1% plementation of this dual-fuel combustion strategy on a HD engine
of the diesel equivalent fuel flow rate per g/kWh reduction in NOx would have to weigh the higher efficiency and lower NOx emissions
emissions [24–26]: against the additional complexity and upfront cost of a port fuel in-
jection system and extra fuel tank.
LHVethanol ⎞
ṁ urea = 0.01(ISNOx −0.4) ⎛ṁ diesel + ṁ ethanol
⎜ ⎟

⎝ LHVdiesel ⎠ (3) 4.7. Potential CO2 reduction

Fig. 10 shows the optimised ethanol-diesel dual-fuel combustion
increased the total fuel flow rate by up to 45.8% in comparison with the The data in Table 4 reveal that the complete combustion of ethanol
CDC mode (8.12 kg/h vs. 5.57 kg/h at 1.5 MPa IMEP). This is attributed can reduce the emissions of CO2 by ∼4% when compared against the
to the relatively lower density (ρethanol ) and energy content (LHVethanol ) combustion of diesel at a given energy input. However, practical
of the ethanol fuel. The total fuel flow rate dropped when increasing the ethanol energy fractions in dual-fuel mode vary between 0.00 and
engine load from 1.5 to 1.8 MPa IMEP due to a reduction in EF. Ap- ∼0.80 while the actual fuel energy consumption changes with the net
propriate volumes of diesel and ethanol fuel tanks will have to be de- indicated efficiency.
signed according to the application of the engine and duty cycle. The use of the engine-out CO2 emissions in the calculation of net
In terms of NOx aftertreatment, the ethanol–diesel dual-fuel com- indicated specific emissions of CO2 (lSCO2) would result in incorrect
bustion attained lower levels of ISNOx than the CDC operation, effec- trends for dual-fuel operation, with significant reductions at all engine
tively decreasing the ṁ urea required. Higher ṁ urea were estimated for loads. This is due to the partial oxidation of hydrocarbons, which in-
both the combustion modes as the engine load was increased. This was crease the formation of CO and reduce the concentration of CO2 in the
exhaust stream.
To remove the effect of incomplete combustion, the lSCO2 (in g/
kWh) was estimated using Eq. (4), which assumed a complete oxidation
of the fuel injected to CO2, either inside the cylinder or in the after-
treatment system. The CO2 emissions resulting from aqueous urea so-
lution reactions in the SCR system [26] have been neglected because
ṁ urea was estimated and its impact on ISCO2 is expected to be very small
[68].

ṁ ṁ MCO2 ⎞
ISCO2 = ⎛ diesel + ethanol ⎞ ⎛
⎜ ⎟⎜ × 103
⎟

⎝ Mdiesel Methanol ⎠ ⎝ Pind ⎠ (4)

where MCO2 is the molar mass of CO2 of 44.01 g/mol [52] and Pind is the
net indicated power in kW.
Fig. 11 shows the optimised dual-fuel operation can achieve lower
lSCO2 than the CDC mode from 0.6 to 2.4 MPa IMEP. The potential CO2
reduction introduced by the ethanol–diesel dual-fuel combustion varied
between 1.8% and 7.5%. This improvement was a result of the increase
in the net indicated efficiency combined with the higher hydrogen to
carbon ratio of the ethanol fuel [69,70]. The low net indicated effi-
ciency at 0.3 MPa IMEP prevented any CO2 reduction and increased the
lSCO2 by 3.7% when compared to the CDC mode.
In order to provide additional insight into the CO2 reduction, a tank-

Table 4
Hypothetical CO2 emissions for diesel and ethanol combustion.
Property Diesel Ethanol

Normalised molecular composition CH1.825O0.0014 CH3O0.5

Lower heating value (LHVfuel) 42.9 MJ/kg 26.9 MJ/kg [51]
Normalised molar mass (Mfuel) 13.87 g/mol 23.03 g/mol
Mass of CO2 emissions per mole of fuel 44.01 gCO2/mol 44.01 gCO2/mol
Mass of CO2 emissions per mass of fuel 3.17 gCO2/g 1.91 gCO2/g
Mass of CO2 emissions per MJ of fuel 73.9 gCO2/MJ 71 gCO2/MJ
Fig. 10. Practical considerations for optimised CDC and ethanol–diesel dual- Specific CO2 emissions reduction n/a ∼4%
fuel operation on a Euro VI HD engine.

448
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

If one considers that the CO2 emissions produced from bioethanol

combustion can be absorbed by plants during photosynthesis [37,44],
the TTW CO2eq emissions for a bioethanol-diesel dual-fuel engine will
be determined by those emitted from diesel combustion only:
TTW CO2eq = 73.9(1−EF ) (6)
From Eq. (6), the WTW CO2eq emissions were given by
WTW CO2eq = [WTTdiesel (1−EF ) + WTTethanol (EF )] + TTW CO2eq (7)
where WTTdiesel is the WTT CO2eq emissions for fossil diesel fuel of
15.4 g/MJ [38,39] and WTTethanol is the WTT CO2eq emissions for
bioethanol, which varies with the biomass and fuel production process.
Dual-fuel operation WTW CO2eq emissions were calculated for corn
Fig. 11. Estimated ISCO2 for optimised CDC and ethanol–diesel dual-fuel op- ethanol produced in the United States (WTTethanol of 68.9 g/MJ [38,39]);
eration at 1200 rpm. wheat ethanol produced in the European Union with production energy
provided by a natural gas-fired combined heat and power plant
(WTTethanol of 64.8 g/MJ [38,39]); and sugarcane ethanol produced in
Brazil (WTTethanol of 24.8 g/MJ [38,39]).
It must be noted that WTTethanol excluded CO2eq emissions produced
by indirect land use change (iLUC). This is due to the uncertainty over
the predictions [72–75] and the potential bonus if biomass is obtained
from restored degraded land [36].
Fig. 13 shows the theoretical TTW CO2eq and WTW CO2eq emissions
for both engine operating modes. Bioethanol-diesel dual-fuel combus-
tion yielded lower TTW CO2eq emissions than the CDC mode. The
lowest TTW CO2eq emissions were attained at mid-loads, where both
the net indicated efficiency and EF were maximised.
The relatively lower TTW CO2eq emissions helped decrease the
WTW CO2eq emissions resulting from bioethanol-diesel dual-fuel op-
eration. However, the biomass feedstock significantly affected the
Fig. 12. Estimated TTW CO2 emissions for CDC and ethanol–diesel dual-fuel CO2eq reduction benefit, which highlights the importance of the source
operation. of the biofuel.
Dual-fuel combustion with sugarcane ethanol demonstrated the best
to-wheels (TTW) analysis was performed by calculating the ratio of the results, decreasing the WTW CO2eq emissions by up to 57% in com-
estimated mass of CO2 emissions to the total fuel energy supplied to the parison with the 89.3 g/MJ for a CDC operation. The reductions in TTW
engine (in MJ) as CO2eq and WTW CO2eq emissions can help combat climate change and

ISCO2 Pind
TTW CO2 =
(ṁ diesel LHVdiesel + ṁ ethanol LHVethanol ) (5)

Fig. 12 indicates the optimised ethanol–diesel dual-fuel combustion

reduced the levels of TTW CO2 emissions by up to 3.2% when compared
against a constant 73.9 g/MJ produced by the CDC operation. This was
attributed to the usage of the ethanol fuel, as the TTW CO2 emissions
are heavily dependent on the in-cylinder fuel characteristics (e.g. Mfuel
and LHVfuel ).
It is important to note that the data plotted in Figs. 11 and 12 were
obtained by assuming complete conversion of the fuel into CO2. For a
more comprehensive analysis, the actual TTW CO2 emissions should be
measured downstream of the aftertreatment system during the appro-
priate engine/vehicle test cycle.

4.8. Theoretical well-to-wheels analysis

A well-to-wheels (WTW) analysis can be used to assess the GHG

emissions and energy expended over the production and use of a given
fuel [37,44]. This holistic methodology now combines the TTW results
with the well-to-tank (WTT). The WTT takes into consideration the
GHGs emitted during the extraction or cultivation of raw materials,
processing, transportation, and other processes necessary to physically
get the fuel into the fuel tank.
The levels of GHGs were expressed as grams of CO2 equivalent
(CO2eq) emissions per MJ of fuel injected. This was required because of
the higher global warming potentials (GWPs) for methane (CH4) and
nitrous oxide (N2O) compounds, which have GWPs equivalent to 25 Fig. 13. Theoretical TTW and WTW CO2eq emissions for CDC and bioethanol-
and 298 times that of the CO2 over a time span of 100 years [71]. diesel dual-fuel operation.

449
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

achieve a more sustainable energy source for the transport sector. bioethanol (e.g. produced from sugarcane) can reduce the use of fossil
fuels and effectively minimise the GHG emissions of future HD engines,
5. Conclusions as supported by the lower tank-to-wheels and well-to-wheels CO2
equivalent emissions.
In this study, experiments were performed to compare the con-
trollability, exhaust emissions, and fuel conversion efficiency of etha- Acknowledgments
nol–diesel dual-fuel combustion to those of conventional diesel com-
bustion (CDC). The investigation was conducted using identical V.B. Pedrozo would like to acknowledge CAPES Foundation
operating conditions at a constant engine speed of 1200 rpm and dif- (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), re-
ferent loads ranging between 0.3 and 2.4 MPa IMEP. search grant No. 13455/13–03, for supporting his PhD study at Brunel
Testing was carried out on a HD diesel engine with a stock piston University London under the supervision of Prof. Hua Zhao. V.B.
and compression ratio of 16.8. Peak in-cylinder pressure and pressure Pedrozo also would like to thank J.C. Bridi, M. Dalla Nora, M. Dhanji,
rise rate (PRR) were limited at 18 MPa and 2.0 MPa/CAD, respectively. and W.H. Au for the comments on the manuscript.
The diesel injection timings and fuel delivery were optimised for the
maximum net indicated efficiency in both the combustion modes. The References
comparison against the CDC operation allowed for a better under-
standing of the potentials, requirements, and limitations of the etha- [1] Intergovernmental Panel on Climate Change (IPCC). Climate Change 2014 –
nol–diesel dual-fuel combustion, which can be summarised as follows: Synthesis Report. IPCC Fifth Assessment Report 2015: 1–112.
[2] United States Environmental Protection Agency. Climate Change Indicators:
Atmospheric Concentrations of Greenhouse Gases. EPA’s Climate Change Indicators
1. The dual-fuel combustion attained significantly lower NOx and soot 2016. https://www.epa.gov/climate-indicators/climate-change-indicators-atmo-
emissions than the CDC cases at low engine loads of 0.3 and 0.6 MPa spheric-concentrations-greenhouse-gases (accessed 4.07.17).
[3] Miller JD, Façanha C. The state of clean transport policy - A 2014 synthesis of
IMEP. This was attributed to the combustion of a more homo- vehicle and fuel policy developments. The ICCT Report 2014: 73.
geneous charge obtained by the use of an early single diesel injec- [4] Kodjak, D. Policies to reduce fuel consumption, air pollution, and carbon emissions
tion at approximately -36 CAD ATDC and large percentages of from vehicles in G20 nations. The ICCT Briefing Paper 2015: 28.
[5] Dec JE. A conceptual model of DI diesel combustion based on laser sheet imaging.
premixed ethanol (with energy fractions of 0.56 and 0.65). SAE Technical Paper; 1997. doi: 10.4271/970873.
2. The dual-fuel mode experienced a relatively low net indicated effi- [6] Kokjohn SL, Hanson RM, Splitter Da, Reitz RD. Fuel reactivity controlled com-
ciency of 38.9% at the lowest load of 0.3 MPa IMEP. This was as- pression ignition (RCCI): a pathway to controlled high-efficiency clean combustion.
Int J Eng Res 2011;12:209–26. http://dx.doi.org/10.1177/1468087411401548.
sociated with the reduced combustion efficiency of 88.7% caused by
[7] Musculus MPB, Miles PC, Pickett LM. Conceptual models for partially premixed low-
excessively lean and low temperature combustion. This region of temperature diesel combustion. Prog Energy Combus Sci 2013;39. http://dx.doi.
engine speed-load map also suffered from a low exhaust gas tem- org/10.1016/j.pecs.2012.09.001.
perature of 463 K, which can adversely affect the effectiveness of the [8] Reitz RD. Directions in internal combustion engine research. Combust Flame
2013;160:1–8. http://dx.doi.org/10.1016/j.combustflame.2012.11.002.
oxidation catalyst. [9] Iwabuchi Y, Kawai K, Shoji T, Takeda Y. Trial of new concept diesel combustion
3. Higher ethanol energy fractions of up to 0.79 and different diesel system - premixed compression-ignited combustion. SAE Tech Paper 1999. http://
injections were required as the engine load was increased from 0.6 dx.doi.org/10.4271/1999-01-0185.
[10] World Health Organization. WHO Air quality guidelines for particulate matter,
to 1.5 MPa IMEP. A transition zone was observed between 0.6 and ozone, nitrogen dioxide and sulfur dioxide; 2006.
0.9 MPa IMEP where less diesel fuel could be partially premixed in [11] World Health Organization. Global health risks – Mortality and burden of disease
order to avoid early ignition and control the levels of PRR. attributable to selected major risks. Genova; 2009.
[12] Environmental Protection Agency (EPA) – National Highway Traffic Safety
4. At mid-loads of 0.9, 1.2, and 1.5 MPa IMEP, optimised dual-fuel Administration (NHTSA) - Department of Transportation (DOT). Greenhouse Gas
combustion was achieved with a 3 mm3 diesel pre-injection prior to Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and
the main diesel injection. The relatively higher degree of fuel stra- Vehicles. Federal Register – Rules and Regulations 2011; 76.
[13] Environmental Protection Agency (EPA) – National Highway Traffic Safety
tification increased the levels of NOx and soot when compared to Administration (NHTSA) – Department of Transportation (DOT). Greenhouse Gas
those obtained at low engine loads. Nevertheless, mid-load dual-fuel Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and
operation attained lower NOx emissions and up to 4.4% higher net Vehicles – Phase 2. Federal Register – Rules and Regulations 2016; 81.
[14] The European Parliament and the Council of the European Union. Regulation (EC)
indicated efficiencies than the CDC cases.
No 595/2009. Official Journal of the European Union 2009; 188.
5. At high engine loads of 1.8 and 2.4 MPa IMEP, early autoignition of [15] The European Parliament and the Council of the European Union. Commission
the ethanol fuel increased the PRR and limited the maximum Regulation (EU) No 582/2011. Official Journal of the European Union 2011; 167.
ethanol energy fractions to 0.25 and 0.19, respectively. This was [16] Posada F, Chambliss S, Blumberg K. Costs of emission reduction technologies for
heavy-duty diesel vehicles. The ICCT White Paper 2016.
linked to the high in-cylinder gas temperatures and pressures prior [17] Delgado O, Rodríguez F, Muncrief R. Fuel Efficiency Technology in European
to the start of combustion. Nonetheless, the ethanol compression Heavy-Duty Vehicles: Baseline and Potential for the 2020–2030 Time Frame. The
ignition combustion helped increase the net indicated efficiency and ICCT White Paper; 2017.
[18] Görsmann C. Improving air quality while reducing the emission of greenhouse
reduce NOx emissions in comparison with the CDC mode. This was gases. Johnson Matthey Technol Rev 2015;59:139–51. http://dx.doi.org/10.1595/
primarily due to a shorter diesel mixing-controlled combustion. 205651315X687524.
6. The ethanol–diesel dual-fuel combustion increased the total fuel [19] O’Connor J, Borz M, Ruth D, Han J, Paul C, Imren A, et al. Optimization of an
Advanced Combustion Strategy Towards 55% BTE for the Volvo SuperTruck
flow rate by up to 45.8% when compared to the CDC operation. This Program. SAE International Journal of Engines 2017; 10: 2017–01 – 0723. doi: 10.
was a result of the differences in fuel characteristics (e.g. LHVfuel ) 4271/2017-01-0723.
and will require the design of appropriate fuel tank volumes. [20] Stanton D, Charlton S, Vajapeyazula P. Diesel Engine Technologies Enabling
Powertrain Optimization to Meet U.S. Greenhouse Gas Emissions. SAE International
Journal of Engines 2013; 6. doi:10.4271/2013-24-0094.
Overall, the optimisation of the diesel injection strategy and ethanol [21] López – De Jesús YM, Chigada PI, Watling TC, Arulraj K, Thorén A, Greenham N,
energy fraction was a key enabler for controlling the PRR. This allowed et al. NOx and PM Reduction from Diesel Exhaust Using Vanadia SCRF®. SAE
International Journal of Engines 2016;9:2016–01 – 0914. doi: 10.4271/2016-01-
for dual-fuel combustion with higher net indicated efficiencies than the
0914.
CDC operation between 0.6 and 2.4 MPa IMEP, with a peak of 47.2% at [22] Walker A, Johnson Matthey. Catalyst-Based Emission Control Solutions for the
1.2 MPa IMEP. Furthermore, the ethanol–diesel dual-fuel combustion Global HDD Market – What Does the Future Hold? Presentation at SAE 2016 Heavy-
attained lower NOx emissions (up to 90%) than the CDC mode from low Duty Diesel Emissions Control Symposium, Gothenburg: 2016.
[23] Delgado O, Lutsey N. The U.S. SuperTruck Program: Expediting the development of
to full engine load. This can decrease the consumption of aqueous urea advanced heavy-duty vehicle efficiency technologies. The ICCT White Paper 2014.
solution in the exhaust aftertreatment system and help to lower the [24] Charlton S, Dollmeyer T, Grana T. Meeting the US Heavy-Duty EPA 2010 Standards
engine running cost. Finally, the substitution of diesel in favour of and Providing Increased Value for the Customer. SAE International Journal of

450
V.B. Pedrozo et al. Fuel 230 (2018) 440–451

Commercial Vehicles 2010; 3. doi: 10.4271/2010-01-1934. [51] Heywood JB. Internal Combustion Engine Fundamentals. First Ed. McGraw-Hill,
[25] Johnson TV. Diesel Emissions in Review. SAE International Journal of Engines Inc.; 1988.
2011; 4. doi:10.4271/2011-01-0304. [52] Economic Commission for Europe of the United Nations (UN/ECE). Regulation No
[26] Stanton DW. Systematic development of highly efficient and clean engines to meet 49 – Uniform provisions concerning the measures to be taken against the emission
future commercial vehicle greenhouse gas regulations. SAE Int J Eng 2013;6. of gaseous and particulate pollutants from compression-ignition engines and posi-
http://dx.doi.org/10.4271/2013-01-2421. tive ignition engines for use in vehicles. Official Journal of the European Union
[27] Liu J, Wang H, Zheng Z, Zou Z, Yao M. Effects of different turbocharging systems on 2013; 171.
performance in a HD diesel engine with different emission control technical routes. [53] Kar K, Cheng WK. Speciated Engine-Out Organic Gas Emissions from a PFI-SI
SAE Tech Paper 2016. http://dx.doi.org/10.4271/2016-01-2185. Engine Operating on Ethanol/Gasoline Mixtures. SAE International Journal of Fuels
[28] Dallmann T, Menon A. Technology Pathways for Diesel Engines Used in Non-Road and Lubricants 2009; 2. doi: 10.4271/2009-01-2673.
Vehicles and Equipment. The ICCT White Paper; 2016. [54] Wallner T. Correlation Between Speciated Hydrocarbon Emissions and Flame
[29] Pedrozo VB, May I, Dalla Nora M, Cairns A, Zhao H. Experimental analysis of Ionization Detector Response for Gasoline/Alcohol Blends. J Eng Gas Turb Power
ethanol dual-fuel combustion in a heavy-duty diesel engine: an optimisation at low 2011;133. http://dx.doi.org/10.1115/1.4002893.
load. Appl Energy 2016;165:166–82. http://dx.doi.org/10.1016/j.apenergy.2015. [55] Prikhodko VY, Curran SJ, Barone TL, Lewis Sa, Storey JM, Cho K, et al. Emission
12.052. characteristics of a diesel engine operating with in-cylinder gasoline and diesel fuel
[30] Pedrozo VB, May I, Zhao H. Characterization of Low Load Ethanol Dual-Fuel blending. SAE Tech Paper 2010;3:946–55.
Combustion using Single and Split Diesel Injections on a Heavy-Duty Engine. SAE [56] Dempsey A, Curran S, Storey J, Eibl M, Pihl J, Prikhodko V, et al. Particulate matter
Tech Paper 2016. http://dx.doi.org/10.4271/2016-01-0778. characterization of reactivity controlled compression ignition (RCCI) on a light duty
[31] Pedrozo VB, May I, Lanzanova TDM, Zhao H. Potential of internal EGR and throt- engine. SAE Tech Paper 2014. http://dx.doi.org/10.4271/2014-01-1596.
tled operation for low load extension of ethanol–diesel dual-fuel reactivity con- [57] Benajes J, Garcia A, Monsalve-Serrano J, Boronat V. Particulates size distribution of
trolled compression ignition combustion on a heavy-duty engine. Fuel reactivity controlled compression ignition (RCCI) on a medium-duty engine fueled
2016;179:391–405. http://dx.doi.org/10.1016/j.fuel.2016.03.090. with diesel and gasoline at different engine speeds. SAE Int J Eng 2017;10. http://
[32] Tong D, Ren S, Li Y, Wang Z, Zhang H, Wang Z, et al. Performance and emissions of dx.doi.org/10.4271/2017-24-0085. 2017–24–0085.
gasoline Homogeneous Charge Induced Ignition (HCII) by diesel through whole [58] Benajes J, García A, Monsalve-Serrano J, Boronat V. An investigation on the par-
operating range on a heavy-duty multi-cylinder engine. Fuel 2017;197:259–71. ticulate number and size distributions over the whole engine map from an opti-
http://dx.doi.org/10.1016/j.fuel.2017.02.003. mized combustion strategy combining RCCI and dual-fuel diesel-gasoline. Energy
[33] Divekar P, Han X, Tan Q, Asad U, Yanai T, Chen X, et al. Mode switching to improve Convers Manage 2017;140:98–108. http://dx.doi.org/10.1016/j.enconman.2017.
low load efficiency of an ethanol-diesel dual-fuel engine. SAE Tech Paper 2017. 02.073.
http://dx.doi.org/10.4271/2017-01-0771. [59] Storey JM, Curran SJ, Lewis Sa, Barone TL, Dempsey aB, Moses-DeBusk M, et al.
[34] May I, Pedrozo V, Zhao H, Cairns A, Whelan S, Wong H, et al. Characterization and Evolution and current understanding of physicochemical characterization of par-
potential of premixed dual-fuel combustion in a heavy duty natural gas/diesel en- ticulate matter from reactivity controlled compression ignition combustion on a
gine. SAE Tech Paper 2016. http://dx.doi.org/10.4271/2016-01-0790. multicylinder light-duty engine. Int J Eng Res 2016. http://dx.doi.org/10.1177/
[35] Li J, Yang W, Zhou D. Review on the management of RCCI engines. Renew Sustain 146808741666163.
Energy Rev 2017;69:65–79. http://dx.doi.org/10.1016/j.rser.2016.11.159. [60] May IA. An experimental investigation of lean-burn dual-fuel combustion in a heavy
[36] The European Parliament and the Council of the European Union. Directive 2009/ duty diesel engine. Brunel University London; 2017.
28/EC. Official Journal of the European Union 2009; 140. [61] He X, Durrett RP, Sun Z. Late intake valve closing as an emissions control strategy at
[37] Edwards R, Larivé J-F, Rickeard D, Weindorf W. Well-to-Wheels analysis of future tier 2 Bin 5 engine-out NOx level. SAE Int J Eng 2008;1. http://dx.doi.org/10.4271/
automotive fuels and powertrains in the European context: Well-to-Tank Report – 2008-01-0637. 2008–01 – 0637.
Version 4.a. Joint Research Centre of the European Commission, EUCAR, and [62] Ickes A, Hanson R, Wallner T. Impact of effective compression ratio on gasoline-
CONCAWE 2014; 4.a. doi: 10.2790/95629. diesel dual-fuel combustion in a heavy-duty engine using variable valve actuation.
[38] Edwards R, Larivé J-F, Rickeard D, Weindorf W. Well-to-Wheels analysis of future SAE Tech Paper 2015. http://dx.doi.org/10.4271/2015-01-1796.
automotive fuels and powertrains in the European context: Well-to-Tank Appendix [63] Zhao H. Advanced direct injection combustion engine technologies and develop-
2 - Version 4a. Joint Research Centre of the European Commission, EUCAR, and ment – Volume 2: Diesel engines. Cambridge: Woodhead Publishing Limited; 2010.
CONCAWE; 2014: 1–133. doi:10.2790/95629. [64] Park C, Busch S. The influence of pilot injection on high-temperature ignition
[39] Edwards R, Hass H, Larivé J-F, Rickeard D, Weindorf W. Well-to-Wheels analysis of processes and early flame structure in a high-speed direct injection diesel engine.
future automotive fuels and powertrains in the European context: Well-to-Tank Int J Engine Res 2017. http://dx.doi.org/10.1177/1468087417728630.
Appendix 4 - Version 4a. Joint Research Centre of the European Commission, [65] Desantes JM, Benajes J, García A, Monsalve-Serrano J. The role of the in-cylinder
EUCAR, and CONCAWE 2014. doi:10.2790/95629. gas temperature and oxygen concentration over low load reactivity controlled
[40] Jaiswal D, De Souza AP, Larsen S, LeBauer DS, Miguez FE, Sparovek G, et al. compression ignition combustion efficiency. Energy 2014;78:854–68. http://dx.
Brazilian sugarcane ethanol as an expandable green alternative to crude oil use. Nat doi.org/10.1016/j.energy.2014.10.080.
Clim Change 2017;7:788–92. http://dx.doi.org/10.1038/nclimate3410. [66] Tsang KS, Zhang ZH, Cheung CS, Chan TL. Reducing emissions of a diesel engine
[41] García a, Pastor JM, Belarte E, Balloul I. Operating range extension of RCCI com- using fumigation ethanol and a diesel oxidation catalyst. Energy Fuels 2010;24.
bustion concept from low to full load in a heavy-duty engine. Appl Energy http://dx.doi.org/10.1021/ef100899z.
2015;143:211–27. http://dx.doi.org/10.1016/j.apenergy.2015.01.035. [67] Prikhodko VY, Curran SJ, Parks JE, Wagner RM. Effectiveness of diesel oxidation
[42] Benajes J, García A, Monsalve-Serrano J, Boronat V. Achieving clean and efficient catalyst in reducing HC and CO emissions from reactivity controlled compression
engine operation up to full load by combining optimized RCCI and dual-fuel diesel- ignition. SAE Int J Fuel Lubr 2013;6. http://dx.doi.org/10.4271/2013-01-0515.
gasoline combustion strategies. Energy Convers Manage 2017;136:142–51. http:// [68] Stelwagen U, Ligterink N. CO2 emission from urea consumption in SCR after-
dx.doi.org/10.1016/j.enconman.2017.01.010. treatment systems in heavy-duty vehicles. TNO Report (TNO 2014 R11513) 2014.
[43] Benajes J, García A, Monsalve-Serrano J, Boronat V. Gaseous emissions and particle http://www.emissieregistratie.nl/erpubliek/documenten/Lucht (Air)/Verkeer en
size distribution of dual-mode dual-fuel diesel-gasoline concept from low to full Vervoer (Transport)/Wegverkeer/TNO (2014) CO2 emission from urea consump-
load. Appl Therm Eng 2017;120:138–49. http://dx.doi.org/10.1016/j. tion in SCR after-treatment systems in HD vehicles.pdf (accessed 2.05.18.).
applthermaleng.2017.04.005. [69] Di Blasio G, BeatriceS. C, Molina M. Effect of port injected ethanol on combustion
[44] Ramachandran S, Stimming U. Well to wheel analysis of low carbon alternatives for characteristics in a dual-fuel light duty diesel engine. SAE Tech Paper 2013;01.
road traffic. Energy Environ Sci 2015;8:3313–24. http://dx.doi.org/10.1039/ http://dx.doi.org/10.4271/2013-01-1692.
C5EE01512J. [70] Han X, Yang Z, Wang M, Tjong J, Zheng M. Clean combustion of n-butanol as a next
[45] Pedrozo VB, May I, Zhao H. Exploring the mid-load potential of ethanol-diesel dual- generation biofuel for diesel engines. Appl Energy 2016. http://dx.doi.org/10.
fuel combustion with and without EGR. Appl Energy 2017;193:263–75. http://dx. 1016/j.apenergy.2016.12.059.
doi.org/10.1016/j.apenergy.2017.02.043. [71] Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: The
[46] Pedrozo VB, Zhao H. Improvement in high load ethanol-diesel dual-fuel combustion Physical Science Basis. Contribution of Working Group I to the Fourth Assessment
by Miller cycle and charge air cooling. Appl Energy 2018;210:138–51. http://dx. Report of the Intergovernmental Panel on Climate Change; 2007.
doi.org/10.1016/j.apenergy.2017.10.092. [72] Laborde D. Assessing the Land Use Change Consequences of European Biofuel
[47] Schwoerer J, Kumar K, Ruggiero B, Swanbon B. Lost-Motion VVA Systems for Policies. International Food Policy Institute (IFPRI); 2011.
Enabling Next Generation Diesel Engine Efficiency and After-Treatment [73] Wang M, Han J, Dunn JB, Cai H, Elgowainy A. Well-to-wheels energy use and
Optimization. SAE Tech Paper 2010. http://dx.doi.org/10.4271/2010-01-1189. greenhouse gas emissions of ethanol from corn, sugarcane and cellulosic biomass
[48] Melo TCC de, Brito MFM de, Moreira MF, Machado GB, Fleischman R. Calculation for US use. Environ Res Lett 2012;7:045905. http://dx.doi.org/10.1088/1748-
of Uncertainty of Measurement for Diesel Engine ESC Test Emissions. SAE Technical 9326/7/4/045905.
Paper; 2013. [74] Yan X, Boies AM. Quantifying the uncertainties in life cycle greenhouse gas emis-
[49] Joint Committee For Guides In Metrology (JCGM). Evaluation of measurement data sions for UK wheat ethanol. Environ Res Lett 2013;8:015024. http://dx.doi.org/10.
– Guide to the expression of uncertainty in measurement. JCGM 100:2008 (GUM 1088/1748-9326/8/1/015024.
1995 with Minor Corrections); 2008. https://www.bipm.org/en/publications/ [75] Morganti K, Al-Abdullah M, Alzubail A, Kalghatgi G, Viollet Y, Head R, et al.
guides/gum.html. Synergistic engine-fuel technologies for light-duty vehicles: fuel economy and
[50] International Organisation of Legal Metrology (IOLM). International Greenhouse Gas Emissions. Appl Energy 2017;208:1538–61. http://dx.doi.org/10.
Recommendation No22 - Alcoholometry. First Ed. Paris: 1973. 1016/j.apenergy.2017.08.213.

451