Applied Thermal Engineering 130 (2018) 1081–1089

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Ethanol in gasoline fuel blends: Effect on fuel consumption and engine

out emissions of SI engines in cold operating conditions
Paolo Iodice ⇑, Giuseppe Langella, Amedeo Amoresano
Dipartimento di Ingegneria Industriale – Università degli Studi di Napoli Federico II, Italy

h i g h l i g h t s

Ethanol/gasoline blended fuels were used to assess exhaust emissions and energy consumption of an SI engine.

Fuel consumption and emissions were measured during the cold transient phase while performing chassis–dynamometer tests.

Fuel consumption and emissions were calculated as a function of oxygen content of the ethanol/gasoline test fuels.

Reductions in cold emission factors were associated with well–defined percentages of ethanol content in gasoline fuel blends.

a r t i c l e i n f o a b s t r a c t

Article history: Ethanol is considered a clean and renewable alternative fuel for SI engines when used in blends with
Received 3 June 2017 gasoline in different fractions to increase oxygen content, thus decreasing exhaust emissions of incom-
Revised 23 October 2017 plete combustion products and dependency on fossil fuels. However, the scientific literature on fuel con-
Accepted 20 November 2017
sumption and exhaust emissions of last generation SI engines during cold operating conditions and using
Available online 21 November 2017
ethanol/gasoline blends is far from comprehensive. In this experimental investigation, ethanol obtained
from grape pomace was used in fuels prepared by blending it in increasing ratios (10, 20 and 30 vol.%)
Keywords:
with oxygen-free gasoline. These ethanol/gasoline blended fuels were used to assess emissions and
Ethanol/gasoline blends
Motorcycle exhaust emissions
energy consumption of a motorcycle equipped with a large displacement four–stroke SI engine. Fuel con-
SI engine sumption and exhaust emissions were measured during the cold transient over the execution of chassis–
Fuel consumption dynamometer tests, and calculated as a function of oxygen content of all ethanol/gasoline test fuels.
Cold-start transient Significant reductions in cold emission factors compared to the commercial gasoline were associated
with well–defined percentages of ethanol content in gasoline fuel blends.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction as carbon monoxide (CO), nitrogen oxides (NOX), unburned hydro-

carbons (HC) and particulate matter (PM) are increasing signifi-
The aim of this study was to highlight some important aspects cantly worldwide [2,3]. Such pollutants greatly affect the state of
related to ethanol as an alternative transportation fuel in SI (Spark– air quality in urban contexts [4,5]. In addition, CO2 emissions from
Ignition) engines. In recent years, the development of advanced the road transport sector are estimated to increase by 92% in 2020,
technologies for the best use of alternative fuels has become abso- compared to 1990, with about nine billion metric tons of CO2 emis-
lutely essential, both to meet the growing demand for energy and sions that will be released from 2020 to 2035, as recently stated by
to reduce the emission levels of incomplete combustion products. the International Energy Agency (IEA) [6].
In this regard, strict legal standards applied to emissions from the For all these reasons, nations worldwide are striving to develop
road transport sector to improve the state of air quality in the cleaner alternative fuels from renewable sources to abate haz-
urban areas, hence reducing negative impacts on human health, ardous air pollutant emissions and reduce the demand for fossil
together with the need to decrease dependency on fossil fuels, fuels. Spark–ignition internal combustion engines can be fuelled
have bolstered research in its quest for clean and renewable fuels with several gasoline fuel blends (e.g. LPG and gasoline, CNG and
[1]. Indeed, air pollutants produced by road transportation such gasoline) [7,8]. Amongst the renewable fuels available for the auto-
motive sector ethanol is recognized as an essential alternative fuel
⇑ Corresponding author at: Università degli Studi di Napoli Federico II, Diparti- for SI engines when blended with gasoline, thanks to its recently
mento di Ingegneria Industriale, Via Claudio 21, 80125 Napoli, Italy. reduced production costs [2,4] and also to its physical and
E-mail address: [email protected] (P. Iodice). chemical properties: lower greenhouse effect, lower dangerous

https://doi.org/10.1016/j.applthermaleng.2017.11.090
1359-4311/Ó 2017 Elsevier Ltd. All rights reserved.
1082 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089

air pollutant emissions, higher octane number, high laminar flame generation spark–ignition engines during cold operating condi-
propagation and the ability to blend with gasoline homogeneously. tions and using ethanol/gasoline blends is far from comprehensive,
Ethanol is generally used in SI engines by mixing it with gaso- further insights into the matter in question are required. This was
line in well-definite fractions, injecting it into the inlet manifold precisely the purpose of the present experimental investigation: to
and recently by direct injection within engine cylinders [9–12]. evaluate the influence of ethanol addition in gasoline fuel blends
Under current EU legislation, 10% v is the maximum percentage on energy consumption and exhaust emissions during the cold-
of ethanol in gasoline. At such a low fraction of ethanol, ethanol/- start transient of last generation SI engines.
gasoline blended fuels do not require any change to the engine In this study, ethanol, obtained from grape pomace, was used in
design [2,12]. The advantages associated to ethanol use in SI engi- fuels prepared by being blended it in increasing ratios (10, 20 and
nes in comparison with pure gasoline can be examined in more 30 vol.%) with oxygen-free gasoline. These ethanol/gasoline
detail below. blended fuels were used to assess emissions and energy consump-
First, in SI engines increasing the compression ratio to increase tion during the warm–up phase of a large-size motorcycle,
the thermal efficiency and engine output power is known to cause equipped with a four-cylinder four-stroke SI engine and belonging
higher pressure and temperature of the air/fuel mixture inside the to the Euro_3 legislative category. This vehicle can be considered
cylinder, thus also causing engine knock problems. Since ethanol representative of the new high–performance motorcycles, which
has a higher octane number than that of gasoline, using it as alter- adopt efficient electronic fuel injection systems and sophisticated
native fuel in SI engines causes fewer risks of engine knock [13]. three-way catalytic converters to improve combustion efficiency
Secondly, the high oxygen content of ethanol is the parameter and to reduce fuel consumption and exhaust pollutants. Fuel con-
which most influences the further oxidation within engine cylin- sumption and exhaust emissions of CO, HC, NOX and CO2 were
ders of incomplete combustion products from ethanol/gasoline measured during cold operating conditions over the execution of
blends. In these conditions, exhaust emissions of carbon monoxide chassis-dynamometer tests, and calculated as a function of oxygen
and unburned hydrocarbons in SI engines can be reduced by using content of all ethanol/gasoline test fuels. These experimental
ethanol/gasoline blended fuels, and this reduction is connected to investigations were performed at the Istituto Motori of the
the increase in oxygen content in the blends. Indeed, ethanol/gaso- National Research Council (IM-CNR, Italy). In comparison with
line blends provide more oxygen than pure gasoline and, under the use of pure gasoline, significant reductions in CO, HC and CO2
these fuel-lean conditions, the high oxygen content in the ethanol cold emission factors were detected and associated with well–de-
molecule improves the oxidation of CO and HC in the fuel-rich fined percentages of ethanol content in gasoline fuel blends, and
regions of the combustion chamber, ensuring more complete com- important findings were obtained about the energy consumption
bustion [14,15]. Since ethanol contains fewer carbon atoms than of SI engines fueled with ethanol/gasoline blends.
gasoline, also ethanol/gasoline blended fuels have a low carbon
number. Therefore ethanol/gasoline mixtures may produce lower
emission levels of CO2, which is the main greenhouse gas (GHG) 2. Material and methods
in road transport emissions. Finally, since heat of vaporization
and flame propagation speed of ethanol are higher than those of 2.1. The emission laboratory
gasoline, using ethanol in gasoline fuel blends can ensure higher
volumetric efficiencies and quicker combustions [16,17]. For all Roller test bench measurements on the vehicle under investiga-
of the above reasons, ethanol can be used in SI internal combustion tion were performed in the emission laboratory belonging to the
engines both on its own (as neat ethanol without any blending) Istituto Motori of National Research Council (Italy). The two-
and by blending it with pure gasoline in several fractions [18,19]. wheeler chassis dynamometer on which the vehicle was tested is
However, ethanol is not without its drawbacks: it has a lower an AVL Zollner 2000, and can reproduce road load conditions
Reid vapor pressure compared to gasoline (which can make a (including vehicle inertia) and measure the exhaust emissions of
cold-start of SI engines difficult), and some experimental investiga- the motorcycle during the selected driving cycles.
tions have found a rise in carbonyl compounds and nitrogen oxide During the tests, quality control procedures were followed to
emissions from ethanol in gasoline fuel blends [20–22]. Another meet the requirements of the European emission regulations for
disadvantage of ethanol consists in it originating from a number compliance testing. In particular, the chassis dynamometer was
of crops such as potatoes, sugarcane, and corn, in other words all set by fixing the following vehicle parameters: the equivalent iner-
products which occupy huge acreages of fertile agricultural land tia mass, the rolling resistance of the front wheel and the aerody-
[23]. namic drag coefficient, as specified in the procedures of Directive
For last generation SI engines, owing to improvements in the 97/24/EC [26]. A driver’s aid shows in detail the speed trace of
field of catalytic converters, electronic fuel injection strategies the driving cycle to follow with a tolerance of around ±1 km/h,
and closed-loop control systems, both fuel consumption and and a variable speed cooling blower is positioned opposite the
exhaust emissions of CO and HC during the warm-up phase are vehicle so as to direct the cooling air in a way which can simulate
far higher than those obtained in hot conditions. This is due to sev- real operating conditions.
eral factors which occur during the cold transient: thermal ineffi- During the test bench experiments, the exhaust gases are first
ciency of the engine, partial and inefficient combustion, catalyst diluted with ambient filtered air by Constant Volume Sampling
inefficiency, and increased frictions. For the above reasons, signifi- with a Critical Flow Venturi (AVL CFV-CVS unit) and then passed
cant improvements have been introduced through international through a dilution tunnel to obtain stable flow conditions. Finally,
legislation with reference to normalized cycles; under the regula- a portion of the diluted gases is sampled downstream of the dilu-
tions now in force and for each vehicle category (passenger cars, tion tunnel, with sample bags being filled for measuring the aver-
motorcycles, heavy-duty vehicles), a more complete assessment age concentration of carbon monoxide (CO), unburned
of the emissive behavior is required, including a cold-start phase hydrocarbons (HC), nitrogen oxides (NOX) and carbon dioxide
in the type-approval driving cycles [21]. (CO2) by means of an exhaust gas analysis system (AVL AMA 4000).
As a general rule, since the heating value of ethanol is much In order to ensure cold-start conditions during the cold-start
lower than that of gasoline, specific fuel consumption of ethanol/ transient, before each test the motorcycle was kept at a quite con-
gasoline blends could increase [24,25]. However, since the scien- stant temperature around 20 °C for 8 h (according to the proce-
tific literature on fuel consumption and exhaust emissions of last dures laid down in Directive 97/24/EC [26]). In this experimental
 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089 1083

campaign, the background pollutant concentrations of indoor air phases: cold phase (ECE_cold, including two base modules) and hot
were also evaluated, subtracting the relevant measured values phase (ECE_hot, including four base modules). WMTC is the world-
from the exhaust gas emission results. wide type-approval driving cycle for motorcycles with maximum
speeds exceeding 140 km/h. The latter requires the execution of
2.2. The vehicle and driving cycles three phases: WMTC_1 (in cold operating conditions), WMTC_2
and WMTC_3. The main kinematic characteristics of the selected
The vehicle used in the experimental campaign was a high per- driving cycles are shown in Table 2.
formance motorcycle equipped with a large displacement four-
stroke engine (1000 cm3 engine displacement and maximum 2.3. Testing fuels
power of 132.4 kW@12,500 rpm) and an ECU (electronic control
unit). It complies with the Euro_3 emission category and, as pollu- Five fuels were tested in this experimental campaign. The first
tant abatement, a three-way catalytic converter is adopted was commercial gasoline (CE0) with oxygenated additive (1.8%
together with a closed loop control system to achieve accurate tun- w/w) and without ethanol addition. The second fuel (E0) was an
ing of the air/fuel ratio. This vehicle adopts an efficient electronic oxygen-free gasoline (without ethanol addition), also used as base
fuel injection strategy allowing the control of fuel feeding for each fuel for the composition of the other ethanol/gasoline testing fuels,
test fuel also during the cold transient phase. Given these premises, which contain 10%, 20% and 30% ethanol v/v (called E10, E20 and
the effects of ethanol addition on the exhaust emissions and fuel E30, respectively). The main fuel properties of all the test blends
consumption were investigated for all the ethanol/gasoline blends and pure ethanol are shown in Table 3. This table evidently
under the original fuel injection system. Regarding the road load explains that, with the addition of ethanol to reference fuel E0,
conditions fixed for this motorcycle, the equivalent vehicle inertia the research octane number, the oxygen content and the density
was set to 280 kg, which corresponds to a rolling resistance of 26.4 of the ethanol/gasoline blended fuels increase whereas the carbon
N, while the aerodynamic drag coefficient was set to 0.024 N/(km/ content, the heating value and the stoichiometric air/fuel ratio con-
h)2. The chief technical characteristics of this motorcycle are sum- currently decrease. The ethanol used for the composition of these
marized in Table 1. blended fuels is anhydrous, and is obtained from grape pomace
Roller test bench measurements were performed on this vehicle produced in Italy during the wine production process.
over the following driving cycles:
3. Results and discussion

ECE_cold (cold-start phase) and ECE_hot phases of the European
type-approval driving cycle (ECE + EUDC) 3.1. CO, HC and NOX exhaust emissions

WMTC_1 phase (cold-start phase) of the World-wide Motorcy-
cle Emissions Test Cycle (WMTC) In Figs. 1–3, CO, HC and NOX emission factors are shown
throughout the cold and hot phases of the European type-
The ECE + EUDC driving cycle is characterized by an urban (ECE) approval driving cycle (ECE_cold and ECE_hot, respectively). These
and an extra urban part (EUDC). The urban part is divided into two emission factors, expressed in terms of mass released per kilome-
ter traveled and reported for CE0, E0, E10, E20 and E30 blended
fuels, are presented for the two phases as mean levels of five test
Table 1
Technical characteristics of the tested vehicle. repetitions (except for test fuel E30 which was repeated once),
with the relative standard deviations in evidence in order to
Engine 4-stroke
appraise the variability of tests in the laboratory.
Displacement [cm3] 1000 While for NOX emission levels (Fig. 3) slight differences exist
Fuel system Electronic fuel injection systems
between cold and hot emissions detected for the engine under
Cooling system Liquid cooling
Maximum power [kW]@12,500 132 investigation, it emerges clearly from Figs. 1 and 2 that, for each
rpm ethanol/gasoline test fuel, a significant difference occurs between
Weight [kg] 189 cold and hot exhaust emissions of incomplete combustion prod-
After-treatment control system Three-way catalyst with lambda sensor ucts (CO and HC), as an effect of the new generation SI engine
European emissions standard Euro_3
and catalytic converter adopted on this motorcycle. Although the

Table 2
Main kinematic properties of the driving cycles.

Driving cycles Duration [s] Length [km] Maximum speed [km/h] Mean speed [km/h]
UDC_cold 390 2 50 18.4
UDC_hot 780 4 50 18.4
WMTC_1 600 4.1 60 24.3

Table 3
Properties of test fuels and ethanol.

Tested fuels CE0 E0 E10 E20 E30 Ethanol

Density [kg/m3] 730 745 750 755 760 792
Lower heating value [MJ/kg] 44.3 44.3 42.3 40.5 38.5 27.0
Stoichiometric air/fuel ratio 14.4 14.6 14.1 12.6 11.7 9.0
Research Octane Number (RON) 95.5 94.7 96.7 98.8 100.8 115
Carbon content [wt%] 86.56 86.56 81.20 76.47 71.97 52.20
Oxygen content [wt%] 1.8 0.0 5.4 10.2 14.4 34.7
Aromatic content [vol.%] 33.3 33.7 30.3 27.0 23.6 0
1084 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089

10 1.00

9 CE0 0.98
8 E0
0.96
7 E10

Lambda ratio - λ
0.94
6 E20
CO [g/km]

5 E30 0.92
CE0
4 0.90 E30
3
0.88
2
0.86
1
0.84
0
0 100 200 300 400 500 600 700
EC E _ c ol d E C E_ h o t
Time [s]

Fig. 1. CO emission factors during cold and hot phases of the ECE driving cycle.
Fig. 4. Air/fuel equivalence ratio k measured during the ECE driving cycle:
comparison between commercial gasoline (CEO) and E30 blend fuel (30 vol.%
ethanol).

1.80

1.60 Under such operating conditions this engine uses a rich mixture
CE0 to overcome the low mixing of the inlet charge, raising unburned
1.40
E0 hydrocarbon emissions due to there being too much fuel flow to
1.20
E10
achieve complete combustion. In addition, the gasoline that
returns from creating a surface film inside the combustion cham-
HC [g/km]

1.00 E20
ber is poorly mixed and thus is poorly burned [27]. Therefore, dur-
0.80 E30 ing the cold-start transient phase, until the enrichment of the
0.60 air/fuel mixture is reduced and the cylinder walls warm up, CO
and HC exhaust emissions are increased.
0.40
Furthermore, the efficiency of sophisticated three-way catalytic
0.20 converters is also reduced owing to rich air/fuel mixtures [28]. The
0.00 range of air/fuel ratio, for the best conversion efficiency of high
EC E_ c o l d E CE _ ho t mileage catalysts, is very limited near the stoichiometric value
(around 0.1). However, this range also depends on the particular
Fig. 2. HC emission factors during cold and hot phases of the ECE driving cycle.
engine operating conditions and catalyst characteristics [29].
Therefore, during cold operating conditions of last generation SI
engines, both the required rich air/fuel mixture and the limits of
0.18
catalytic technology produce higher CO and HC exhaust emissions.
Obviously, such operating conditions cannot lead to a simultane-
CE0
0.16 ous and comparable increase in NOX cold emissive levels.
0.14 E20 As shown clearly in Fig. 4, during hot operating conditions (after
180 s) fixing the air/fuel ratio exactly on the selected near-
E30
0.12 stoichiometric values is no straightforward assumption also for
electronic fuel injection and closed-loop control systems. Indeed,
NOX [g/km]

0.10
for each test fuel, the equivalence air/fuel ratio detected on this
0.08 vehicle during the UDC driving cycle oscillates around the fixed
0.06 stoichiometric value with a periodic trend as the fuel flow is chan-
ged. Thus, also for ethanol/gasoline blends the lambda ratio
0.04 remains only stoichiometric on average over time, varying
0.02 instantly at particular engine loads.
With regard to CO exhaust emissions and for all test fuels
0.00
EC E _ c o ld E C E_ h o t
(Fig. 1), cold-start emission factors were higher than the values
measured during hot operating conditions owing to mixture
Fig. 3. NOX emission factors during the cold and hot phases of the ECE driving cycle. enrichment, partial combustion and catalyst inefficiency. The
(NOX emission factors of E0 and E10 blends are missing due to sampling errors). reduction in the CO emission factor, during cold operating condi-
tions, is considerable for E20 blended fuel (6.5 g/km) in compar-
ison with base fuel E0 and commercial gasoline CE0 (8.4 g/km
light-off temperature of the three-way catalyst is dominant in con- and 7.7 g/km respectively). This reduction may be attributed to
trolling its efficiency, other parameters must be considered to the increase in oxygen content in the ethanol/gasoline blended
achieve best performance and reduced exhaust emissions. A fur- fuels which improves the oxidation of CO in the fuel-rich regions
ther critical variable is the air/fuel equivalence ratio. During the of the combustion chamber, ensuring a more complete combustion
warming-up process of the engine, fuel evaporation is slow due during the cold transient phase.
to the cylinder walls being cold. Hence the fuel flow is increased On the other hand, in Fig. 1 it is also evident that E30 test fuel
to deliver simply combustible fuel-rich blends, as shown in (30 vol.% ethanol) throughout the cold transient phase has a higher
Fig. 4, in which the lambda ratio of CE0 and E30 test fuels is mea- CO emission factor (7.0 g/km) than E20 blended fuel. This effect
sured during the ECE driving cycle. may be elucidated by evaluating other properties of ethanol: with
 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089 1085

an increasing ethanol percentage in the blended fuels [27], the emission factors, reported for CE0, E0, E10, E20 and E30 blended
latent heat of vaporization increases while the lower heating value fuels in terms of mass released per distance travelled, are pre-
decreases, thereby resulting in lower flame temperature and burn- sented for the two phases as mean levels of five test repetitions
ing velocity than E20 test fuel. This leads to a less complete and (except for E30 test fuel which underwent a single repetition), with
efficient combustion process and hence to higher CO emission fac- the relative standard deviations in evidence. As is well known,
tors during cold operating conditions. combustion of fuels with carbon compounds always produces
As clearly shown by Fig. 2, when ethanol content rises in the CO2 exhaust emissions. There should be an inverse relation
blends, HC emission factors measured during the cold transient between the variation in CO2 emissions of the tested fuels and that
phase decrease compared with E0 reference fuel, but only up to a of CO emission levels since the high oxygen content in the
certain percentage of ethanol in the blended fuels. The higher oxy- ethanol/gasoline blends raises the air/fuel ratio in the fuel-rich
gen fraction in the blends, during cold operating conditions, is regions of the combustion chamber, thereby improving the com-
more effective at improving oxidation of HC emissions in compar- bustion process. Under these conditions, therefore, CO2 emissions
ison with E0 base fuel: with the rising ethanol fraction in the should increase with an increasing ethanol percentage in the test-
blends, the air/fuel ratios move from rich toward lower values (sto- ing fuels while, at the same time, the amount of non-complete
ichiometric ratios). combustion products should decrease.
However, in order to explain this reduction, other properties of It is worth stressing that the fuel injection strategy and the elec-
ethanol have to be considered. When the ethanol fraction increases tronic open-loop control adopted on this vehicle during the cold
in the ethanol/gasoline blends, at first the Reid vapor pressure transient are the same for all mixtures tested, and the stoichiomet-
increases (thus favoring the fuel vaporization during the warm- ric air/fuel ratios of gasoline and ethanol are very different (about
up phase), peaking at about 15% v/v of ethanol, whereas at ethanol 14.7 and 9, respectively), while ethanol/gasoline blends have stoi-
fractions higher than 15% v/v the volatility of ethanol/gasoline chiometric ratios which vary linearly with the amount of ethanol
blends decays since the Reid vapor pressure of ethanol is very in the mixtures (Table 3). For this reason, the amount of air
lower (17 kPa) than that of gasoline (53  60 kPa) [12,30]. For required by pure gasoline to obtain fuel-rich conditions during cold
the same reason, the HC emission factor (1.34 g/km) of E30 test operating conditions could be too high for the other mixtures
fuel detected during the cold transient is higher than the E20 tested, thus reducing CO cold emission factors and increasing
blended fuel (1.0 g/km). Indeed, for ethanol fractions in the mix- CO2 cold emissions. Such results are also reported in several scien-
tures higher than 20% v/v, the volatility of E30 ethanol/gasoline tific publications [31], in which the high ethanol fraction (60 vol.%)
test fuel is lower than that of E20 test fuel [31]. in ethanol/gasoline blended fuels was found accountable for the
In addition, a higher percentage of ethanol in the blends reduces high CO2 exhaust emissions in SI engines owing to the increased
the lower heating value and concomitantly increases the latent oxygen content from the alcohols and the improved combustion
heat of vaporization of the blends, thus reducing the combustion process. Likewise, in other investigations, CO2 engine out emis-
temperature which leads to the rise of HC cold emission factors sions rose with the addition of butanol, ethanol and methanol in
and, at the same time, to the decrease in NOX cold emission levels. gasoline fuel blends. This is due to the fact that alcohol blends with
Indeed, owing to the higher latent heat and lower heating value of a higher oxygen content of 7.5 wt% ensures more efficient and
ethanol/gasoline blends, the temperature of the air/fuel mixture complete combustion of such blends inside the cylinder, closer to
decreases at the end of the intake stroke in comparison with pure the full combustion process, thereby increasing the total quantity
gasoline; under these conditions, the combustion temperature will of CO2 exhaust emissions [32].
also decrease which sequentially will affect thermal NOx produc- On the other hand, since ethanol contains fewer carbon atoms
tion. For this reason, NOx emission decreases with the increase than gasoline, also ethanol/gasoline blended fuels are character-
in content of ethanol in the blended fuels [20,31] (Fig. 3). ized by a lower carbon number than pure gasoline (CE0). Therefore
the test mixtures could result in lower CO2 emission levels with
increasing ethanol percentages in the test fuels. The chemical com-
3.2. Energy consumption and carbon dioxide emissions bustion reactions of ethanol and gasoline are shown in Eqs. (1) and
(2), respectively. The stoichiometric results of these full combus-
As illustrated in Fig. 5, CO2 emission factors were measured tion equations show the production of 1.90 kg of CO2 per kg of pure
throughout the cold and hot phases of the European type- ethanol, contrasting with 3.17 kg of CO2 per kg of gasoline (assum-
approval driving cycle (ECE_cold and ECE_hot, respectively). The ing CnH1.87n as average molecular composition) [29,33]. Therefore,
ethanol added to gasoline fuel blends could really lower the rela-
tive CO2 cold emissions.
320
CE0 E0 E10 E20 E30

280
C 2 H6 O þ 3ðO2 þ 3:76ÞN2 ! 2CO2 þ 3H2 O þ 11:28N2 ð1Þ

240
C n H1:87n þ 1:47nðO2 þ 3:77ÞN2 ! nCO2 þ 0:94nH2 O þ 5:55nN2
200 ð2Þ
CO2 [g/km]

160
According to these observations and the experimental results on the
test engine provided in Fig. 5, these two reverse effects clearly com-
120
pensate each other, resulting in CO2 cold emissions of ethanol/gaso-
80 line blends not specifically related to the reduced CO cold emission
levels with increasing ethanol percentage in the test fuels. Indeed,
40 CO2 emission levels detected during the transient phase decrease
somewhat with increasing ethanol percentage in the mixtures in
0
E C E_ c o l d EC E _ h ot
comparison with pure gasoline CE0. Therefore, low carbon number
of ethanol is accountable for higher CO2 cold emissions of ethanol/-
Fig. 5. CO2 emission factors during cold and hot phases of the ECE driving cycle. gasoline blended fuels in comparison with commercial gasoline.
1086 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089

E30 test fuel revealed lower CO2 cold emission factors (265 g/ heating value of ethanol compared to gasoline, higher flow rates
km) than the E10 and E20 blends, (around 282 g/km for both of ethanol/gasoline mixtures should be injected with increasing
blends), a difference which could be ascribed to a particular aspect. of ethanol fraction in the blends. Secondly, although during cold
As clarified by some experimental investigations [34], during cold operating conditions the electronic fuel injection system operates
operating conditions of an SI engine, if the air/fuel ratio exceeds a the engine with rich air/fuel ratios, high ethanol fractions added
certain lean value, incomplete combustion can occur, resulting in to gasoline fuel blends can produce a leaning effect, which
frequent misfires and increased unburned hydrocarbon emissions. improves the fuel combustion process. These two aspects compete
In addition, the flow rate of ethanol/gasoline mixtures may not be with each other during the cold transient phase, resulting in a not
adjusted straightaway according to real combustion conditions, significant difference in fuel consumption between using commer-
even if the engine under investigation adopts an electronic fuel cial gasoline and using E20 and E30 ethanol/gasoline blends.
injection strategy. This further effect also results in irregular com- To measure the fuel efficiency of this engine using ethanol/ga-
bustion conditions, involving higher CO and HC cold emissions and soline blended fuels, energy consumptions were also calculated
hence lower CO2 cold emission factors of blend E30 than the other as kJ/km for the same driving cycles and test fuels. Knowing the
ethanol/gasoline test fuels [35]. In the same Fig. 5, it is also evident density and heating value of all test fuels, and starting from the
that, for all blended fuels, the variance of the results is very accept- experimental fuel consumptions shown in Fig. 7, energy consump-
able in comparison with base fuel. It may thus be inferred that the tions of all test fuels were calculated and plotted (Fig. 8) against
combustion stability of the test engine is highly acceptable even the oxygen content of the same blends. According to these last
when using higher fractions of ethanol added to gasoline fuel experimental results on the test engine, it is evident that energy
blends. consumptions decrease with increasing ethanol fraction in the
In Figs. 6 and 7, CO2 cold emission factors and fuel consump- testing blends. In comparison with fuel consumptions calculated
tions of all test fuels (detrmined during ECE_cold and WMTC_1 driv- as l/100 km in Fig. 7, with an increasing ethanol percentage in
ing cycles) are plotted against the oxygen content of the same the test blends, a decrease in heating value prevails over density
blends. In relation to the difference of CO2 cold emission factors increase.
and fuel consumptions obtained between the two driving cycles However, the measure for the fuel efficiency of any internal
during the cold transient phase, the kinematic features of the per- combustion engine is also the brake-specific fuel consumption
tinent driving patterns can be analysed as follows. First, for similar (BSFC) expressed in g/kWh, which is the ratio between the rate
average speeds, high levels of acceleration (characteristic of the of fuel consumption (expressed in g/s) and the brake power
WMTC_1 driving cycle) during hot operating conditions usually (expressed in kW). Obviously, lower values of this parameter are
involve higher engine loads and a rise in energy demand to per- always required. It has been extensively shown [31] that, with
form the related driving cycle, with the resulting enrichment of increasing ethanol percentages in gasoline fuel blends, the brake-
the air/fuel ratio (outside the optimum range of catalyst efficiency). specific fuel consumptions of ethanol/gasoline blends could
Secondly, for lower engine loads related to lower levels of average increase, owing to lower heating value of ethanol in comparison
speed (characteristic of the ECE_cold driving cycle), the corre- with pure gasoline. If the heating value of ethanol/gasoline mix-
sponding leaning of the air/fuel ratio during the cold transient tures decreases with increasing ethanol content in fuel blends,
phase can result in irregular operating conditions of the engine. the energy content of these blends is also reduced. For this reason,
Starting from these statements and according to the experimental in comparison to the use of commercial gasoline and under similar
results shown in Figs. 6 and 7, higher CO2 cold emission factors and operating conditions, more fuel is required to achieve the same
fuel consumptions were obtained over the ECE_cold driving cycle engine power output when using ethanol/gasoline blends.
due to extreme leaning of the air/fuel mixture and incomplete As revealed by considerable scientific research [31,33], in SI
combustion during the transient phase of this cycle. engines the addition of ethanol in ethanol/gasoline blends can
In order to clarify the experimental results shown in Fig. 7, two increase the torque and brake power in comparison to the use of
reverse effects have to be pointed out. First, owing to the lower pure gasoline. Oxygenated fuels such as ethanol/gasoline mixtures

310
CE0
300
E0
290
E20
280
E10 E30
270
CO2 [g/km]

260
ECE_cold
250
WMTC_1
240 CE0
230
E0 E20
220 E30
E10
210
200
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Oxygen content [wt %]

Fig. 6. CO2 emission factors during the cold phase of ECE and WMTC driving cycles plotted against the oxygen content of test fuels.
 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089 1087

14.5

14.0
E20 E30
13.5 CE0
EO
13.0 E10
Fuel consumption [l/100km]
12.5

12.0
ECE_cold
11.5
WMTC_1
11.0 CE0
E20 E30
10.5

10.0 E0
E10
9.5

9.0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Oxygen content [wt %]

Fig. 7. Fuel consumption during the cold phase of ECE and WMTC driving cycles plotted against the oxygen content of test fuels.

4600
4500 CE0
4400
E0 E20
4300
Energy consumption [kJ/km]

4200
E10 E30
4100
4000
3900
3800 ECE_cold
3700 WMTC_1
3600 CE0
3500
E20
3400
E10
3300 E0 E30
3200
3100
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Oxygen content [wt %]

Fig. 8. Energy consumption during the cold phase of ECE and WMTC driving cycles plotted against the oxygen content of testing fuels.

can ensure better combustion performance, thus giving a higher soline blends improves the combustion process and hence the
engine torque without varying the injection strategy and throttle brake thermal efficiency of SI engines. Analogous results were
opening. Such results are due to the leaning air/fuel mixture pro- reported elsewhere [37], describing an improvement in brake ther-
duced by ethanol in gasoline fuel blends which raises the lambda mal efficiency by using ethanol/gasoline blended fuels owing to
ratio (k). With the addition of ethanol in the blends, the torque better combustion efficiency. In addition, due to ethanol’s high
and brake power are also increased owing to the heat of vaporiza- latent heat of vaporization, the air/fuel mixture can be compressed
tion of ethanol (about 910 kJ/kg), more than twice that of gasoline. more effectively with increasing ethanol percentage in the blends,
Indeed, during vaporization of ethanol/gasoline blended fuels and thereby improving volumetric efficiency and the brake thermal
under hot operating conditions, fuel flow tends to absorb more efficiency of SI engines [38]. Lastly, with increasing ethanol fraction
heat from the cylinder than that of gasoline. In these conditions, in the blended fuels, the temperature and pressure tend to
the charge and intake manifold temperatures decrease, thereby decrease at the beginning of the combustion phase, thereby
increasing volumetric efficiency. Moreover, with the addition of increasing the indicated work and the indicated efficiency of SI
ethanol in ethanol/gasoline blends, also the faster flame speed engines.
ensures high brake power.
Further, the brake thermal efficiency of SI engines can be
4. Conclusions
enhanced with increasing ethanol percentage in the blends, due
to high oxygen content (low carbon number) and heat of vaporiza-
In recent years, due to rising fuel prices, increasing demand for
tion of ethanol [31,36]. Indeed, high oxygen content of ethanol/ga-
energy, and stricter air pollution regulations in the road transport
1088 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089

sector, researchers world-wide are striving to develop cleaner [2] M. Mofijur, M.G. Rasul, J. Hyde, A.K. Azad, R. Mamat, M.M.K. Bhuiya, Role of
biofuel and their binary (diesel–biodiesel) and ternary (ethanol–biodiesel–
renewable fuels, with a view to abating harmful air pollutants
diesel) blends on internal combustion engines emission reduction, Renew.
and lessening the dependency on fossil fuels. Amongst the alterna- Sustain. Energy Rev. 53 (2016) 265–278.
tive fuels available, ethanol in gasoline fuel blends is one of the [3] A. Di Palma et al., Atmospheric particulate matter intercepted by moss-bags:
most widely used renewable fuels for internal combustion engines. relations to moss trace element uptake and land use, Chemosphere 176 (2017)
361–368.
In the present research, the effects of ethanol/gasoline blends [4] A.M. Liaquat, M.A. Kalam, H.H. Masjuki, M.H. Jayed, Potential emissions
on fuel consumption and exhaust emissions of SI engines during reduction in road transport sector using biofuel in developing countries,
the warm–up phase were investigated experimentally in the labo- Atmos. Environ. 44 (2010) 3869–3877.
[5] L. Chapman, Transport and climate change: a review, J. Transp. Geogr. 15
ratories of the National Research Council (IM-CNR, Italy). For this (2007) 354–367.
purpose, different percentages of ethanol mixed in gasoline fuel [6] International Energy Administration (IEA). CO2 emissions from fuel
blends (10, 20 and 30 vol.%) were characterized through the execu- combustion. Cited on [5thDecember2014]. Available from: <http://www.
ieaorg/publications/freepublications/
tion of chassis-dynamometer tests performed on a high–perfor- publication/co2emissionsfromfuelcombustionhighlights2013pdf>, 2013.
mance motorcycle, without any change to the engine design and [7] E. Pipitone, G. Genchi, NOX reduction and efficiency improvements by means of
under the original fuel injection system. the double fuel HCCI combustion of natural gas-gasoline mixtures, Appl.
Therm. Eng. 102 (2016) 1001–1010.
The results obtained in this study revealed that the ethanol [8] E. Pipitone, S. Beccari, Performances improvement of a S.I. CNG BI-fuel engine
added into oxygen-free gasoline reduced cold emissions, with the by means of double-fuel injection. SAE Technical Papers, 2009, No. 2009-24-
E20 blend (20% v/v ethanol content) revealing the greatest emis- 0058.
[9] M.B. Çelik, Experimental determination of suitable ethanol–gasoline blend rate
sion improvement amongst the test fuels. Such test fuel, during
at high compression ratio for gasoline engine, Appl. Therm. Eng. 28 (2008)
the transient phase, shown a considerable reduction in CO and 396–404.
NOX emission factors in comparison with commercial gasoline [10] M. Koç, Y. Sekmen, T. Topgül, H.S. Yücesu, The effects of ethanol–unleaded
CE0 and a significant decrease in HC cold emission factor in com- gasoline blends on engine performance an exhaust emissions in a
sparkignition engine, Renew. Energy 34 (2009) 2101–2106.
parison with oxygen-free gasoline E0. The reduction of CO and [11] J.M. Storey, T. Barone, K. Norman, S. Lewis, Ethanol blend effects on direct
HC cold emission factors is associated to the high oxygen content injection spark-ignition gasoline vehicle particulate matter emissions, SAE Int.
in the ethanol molecule (which improves the oxidation of incom- J. Fuels Lubr. 3 (2010) 650–659.
[12] P. Gomes et al., Study on boosted direct injection SI combustion with ethanol
plete combustion products in the fuel-rich regions of the combus- blends and the influence on the ignition system, SAE Technical Paper, 2011,
tion chamber), and the high volatility of E20 fuel which enhances No. 2011-36-0196.
fuel evaporation during cold operating conditions, thus facilitating [13] S. Yousufuddin, S.N. Mehdi, Effect of ignition timing, equivalence ratio, and
compression ratio on the performance and emission characteristics of variable
the cold-start and decreasing exhaust emissions in the first engine compression ratio SI engine using ethanol-unleaded gasoline blends, Int. J.
cycles. Furthermore, the addition of ethanol in gasoline fuel blends Eng., Iran, IJE Trans. B: Appl. 21 (2008) 97–106.
(20% and 30% v/v ethanol content) produced lower flame temper- [14] Wei-Dong Hsieh et al., Engine performance and pollutant emission of an SI
engine using ethanol-gasoline blended fuels, Atmos. Environ. 36 (2002) 403–
ature which led to lower exhaust temperature, thereby producing 410.
a lower amount of NOx emission. [15] S.H. Yoon, C.S. Lee, Effect of undiluted bioethanol on combustion and
CO2 emission levels detected during the transient phase were emissions reduction in a SI engine at various charge air conditions, Fuel 97
(2012) 887–890.
not specifically related to the reduced CO cold emissions with
[16] S. Kumar, S. Neetu, P. Ram, Anhydrous ethanol: a renewable source of energy,
increasing ethanol percentage in the test fuels, since ethanol/gaso- Renew. Sustain. Energy Rev. 14 (2010) 1830–1844.
line blended fuels are characterized by a lower carbon number [17] M. Balat, Global bio-fuel processing and production trends, Energy Explor.
than pure gasoline. Hence CO2 cold emission factors were reduced Exploit. 25 (2007) 195–218.
[18] S.K. Thangavelu, A.S. Ahmed, F.N. Ani, Review on bioethanol as alternative fuel
with the addition of ethanol in the blends. With regard to fuel con- for spark ignition engines, Renew. Sustain. Energy Rev. 56 (2016) 820–835.
sumption during cold operating conditions, two reverse effects [19] A.M. Pourkhesalian et al., Performance and emission comparison and
were evaluated. First, owing to the lower heating value of ethanol investigation of alternative fuels in SI engines, SAE Technical Paper, 2009,
No. 2009-01-0936.
compared to gasoline, higher flow rates of ethanol/gasoline mix- [20] B.M. Masumn, H.H. Masjuki, M.A. Kalam, I.M. Rizwanul Fattah, S.M. Palash, M.J.
tures should be injected with increasing ethanol fractions in the Abedin, Effect of ethanol–gasoline blend on NOx emission in SI engine, Renew.
blends. Secondly, although during the cold transient phase the Sustain. Energy Rev. 24 (2013) 209–222.
[21] M.A. Costagliola, L. De Simio, S. Iannaccone, M.V. Prati, Combustion efficiency
electronic fuel injection strategy operates the engine with rich air/- and engine out emissions of a S.I. engine fuelled with alcohol/gasoline blends,
fuel ratios, high ethanol fractions added to gasoline fuel blends can Appl. Energy 111 (2013) 1162–1171.
produce a leaning effect, which improves the fuel combustion pro- [22] I. Gravalos, D. Moshou, T. Gialamas, P. Xyradakis, D. Kateris, Z. Tsiropoulos,
Emissions characteristics of spark ignition engine operating on lower–higher
cess. These two aspects compete with each other, resulting in little molecular mass alcohol blended gasoline fuels, Renew. Energy 50 (2013) 27–
difference in fuel consumption between using commercial gasoline 32.
and using ethanol/gasoline blends. [23] M.A. Ghadikolaei, Effect of alcohol blend and fumigation on regulated and
unregulated emissions of IC engines-a review, Renew. Sustain. Energy Rev. 57
(2016) 1440–1495.
Acknowledgements [24] H. Bang-Quan, W. Jian-Xin, H. Ji-Ming, Y. XiaoGuang, X. Jian-Hua, A study on
emission characteristics of an EFI engine with ethanol blended gasoline fuels,
Atmos. Environ. 37 (2003) 949–957.
The authors of this study thank researchers and colleagues at [25] M.A. Ceviz, F. Yüksel, Effects of ethanol unleaded gasoline blends on cyclic
the Istituto Motori of the National Research Council, Italy (CNR) variability and emissions in an SI engine, Appl. Therm. Eng. 25 (2005) 917–
for their support in performing experimental tests. 925.
[26] United Nations European Commission for Economics, Measurement procedure
for two wheeled motorcycles equipped with a positive or compression ignition
Appendix A. Supplementary material engine with regard to the emission of gaseous pollutants, CO2 emissions and
fuel consumption, Global Technical Regulation (GTR) N° 2 – (ECE/TRANS/180/
Add.2), 2005.
Supplementary data associated with this article can be found, in [27] A. Roberts, R. Brooks, P. Shipway, Internal combustion engine cold-start
the online version, at https://doi.org/10.1016/j.applthermaleng. efficiency: a review of the problem, causes and potential solutions, Energy
Convers. Manage. 82 (2014) 327–350.
2017.11.090.
[28] H. Li et al., Study of thermal characteristics and emissions during cold start
using an on-board measuring method for modern SI car real world urban
References driving, SAE Int. J. Engines 1 (2009) 804–819.
[29] J. Heywood, Internal Combustion Engine Fundamentals, McGraw-Hill, 1988.
[30] T.R. Aulich, X.M. He, A.A. Grisanti, C.L. Knudson, Gasoline evaporation-ethanol
[1] A. Amoresano, G. Langella, S. Sabino, Optimization of solar integration in
and nonethanol blends, J. Air Waste Manage. Assoc. 44 (1994) 1004–1009.
biomass fuelled steam plants, Energy Procedia 81 (2015) 390–398.
 P. Iodice et al. / Applied Thermal Engineering 130 (2018) 1081–1089 1089

[31] M.N.A.M. Yusoff, N.W.M. Zulkifli, B.M. Masum, H.H. Masjuki, Feasibility of [35] D. Turner et al., Combustion performance of bio-ethanol at various blend ratios
bioethanol and biobutanol as transportation fuel in spark-ignition engine: a in a gasoline direct injection engine, Fuel 90 (2011) 1999–2006.
review, RSC Adv. 5 (2015) 100184–100211. [36] M.K. Balki, C. Sayin, M. Canakci, The effect of different alcohol fuels on the
[32] H.S. Farkade, A.P. Panthre, Int. Conf. Adv. Nanomater. Emerg. Eng. Technol. 2 performance, emissions and combustion characteristic of a gasoline engine,
(2012) 205–215. Fuel 115 (2014) 901–906.
[33] B. Dogan, D. Erol, H. Yaman, E. Kodanli, The effect of ethanol-gasoline blends [37] F.T. Ansari, A.P. Verma, Int. J. Eng. 1 (2012) 1–10.
on performance and exhaust emissions of a spark ignition engine through [38] A. Gimelli, M. Muccillo, O. Pennacchia, Study of a new mechanical VVA system.
exergy analysis, Appl. Therm. Eng. 120 (2017) 433–443. Part II: Estimation of the actual fuel consumption improvement through 1D
[34] Y.C. Yao, J.H. Tsai, H.L. Chiang, Effects of ethanol-blended gasoline on air fluid dynamic analysis and valve train friction estimation, Int. J. Engine Res. 16
pollutant emissions from motorcycle, Sci. Total Environ. 407 (2009) 5257– (2015) 762–772.
5262.