Journal of the Air & Waste Management Association

ISSN: 1096-2247 (Print) 2162-2906 (Online) Journal homepage: http://www.tandfonline.com/loi/uawm20

Bioethanolgasoline fuel blends: Exhaust

emissions and morphological characterization of
particulate from a moped engine

Maurizia Seggiani , M. Vittoria Prati , M. Antonietta Costagliola , Monica

Puccini & Sandra Vitolo

To cite this article: Maurizia Seggiani , M. Vittoria Prati , M. Antonietta Costagliola , Monica
Puccini & Sandra Vitolo (2012) Bioethanolgasoline fuel blends: Exhaust emissions and
morphological characterization of particulate from a moped engine, Journal of the Air & Waste
Management Association, 62:8, 888-897, DOI: 10.1080/10962247.2012.671793

To link to this article: http://dx.doi.org/10.1080/10962247.2012.671793

Accepted author version posted online: 03

Apr 2012.
Published online: 03 Apr 2012.

Submit your article to this journal

Article views: 443

View related articles

Citing articles: 6 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=uawm20

Download by: [80.82.77.83] Date: 02 March 2017, At: 00:24

TECHNICAL PAPER

Bioethanolgasoline fuel blends: Exhaust emissions and morphological

characterization of particulate from a moped engine
Maurizia Seggiani,1, M. Vittoria Prati,2 M. Antonietta Costagliola,2 Monica Puccini,1
and
1
Sandra Vitolo1
Department of Chemical Engineering, Industrial Chemistry and Material Science, University of Pisa, Pisa, Italy
2
Istituto Motori CNR, Napoli, Italy
Please address correspondence to: Maurizia Seggiani, Department of Chemical Engineering, Industrial Chemistry and Material Science,
University of Pisa, Largo L. Lazzarino 1, 56126 Pisa, Italy; e-mail: m.seggiani@diccism.unipi.it

This study was aimed at evaluating the effects of gasolineethanol blends on the exhaust emissions in a catalyst-equipped four-
stroke moped engine. The ethanol was blended with unleaded gasoline in at percentages (10, 15, and 20% v/v). The regulated
pollutants and the particulate matter emissions were evaluated over the European ECE R47 driving cycle on the chassis
dynamometer bench. Particulate matter was characterized in terms of total mass collected on filters and total number of particles
in the range 7 nm10 mm measured by electrical low-pressure impactor (ELPI). In addition, particle-phase polycyclic aromatic
hydrocarbons (PAHs) emissions were evaluated to assess the health impact of the emitted particulate. Finally, an accurate
morphological analysis was performed on the particulate by high-resolution transmission electron microscope (TEM) equipped
with a digital image-processing/data-acquisition system. In general, CO emission reductions of 6070% were obtained with 15 and
20% v/v ethanol blends, while the ethanol use did not reduce hydrocarbon (HC) and NOx emissions. No evident effect of ethanol on
the particulate mass emissions and associated PAHs emissions was observed. Twenty-one PAHs were quantified in the particulate
phase with emissions ranging from 26 to 35 mg/km and benzo[a]pyrene equivalent (BaPeq) emission factors from 2.2 to 4.1 mg/km.
Both particulate matter and associated PAHs with higher carcinogenic risk were mainly emitted in the submicrometer size range
(<0.1 mm). On the basis of the TEM observations, no relevant effect of the ethanol use on the particulate morphology was evidenced,
showing aggregates composed of primary particles with mean diameters in the range 17.532.5 nm.

Implications: Mopeds may contribute in a large share to air pollution in some major urban areas; therefore, efforts may be taken
to reduce their emissions. The addition of ethanol into gasoline (up 20% v/v) leads to a significant reduction of CO emission at the
exhaust of a catalyzed 4-S moped, whereas this does not reduce HC and NOx. No clear effects on particulate matter and associated
PAHs emissions have been observed with ethanol addition, but particulate-phase PAHs were mainly found on submicrometer-size
particles (<0.1 mm), which penetrate deeply into the human respiratory tract.

Introduction In recent years the application of alternative fuels has

Powered two-wheel vehicles are a widespread means of trans-
port in some regions such as southern Europe and in countries of
East and South Asia. In southern Europe a large proportion of in-
use powered two-wheelers are mopeds, that is, motorcycles with a
small combustion engine with displacement of not more than
50 cc and a maximum design speed of 45 km/hr. In Italy the
total population of about 51 million of vehicles included
36,371,790 cars, 4,584,210 trucks, 6,118,098 motorcycles (>50
cc) and 98,724 buses in 2009 (Automobile Club dItalia, 2009).
The contribute of mopeds is estimated at 3,321,000 vehicles
(Associazione Nazionale Ciclo Motociclo Accessori, 2011).
Therefore, urban air pollution can be strongly influenced by
two-wheeler emissions, which are not negligible when compared
with those of modern passenger cars.
received considerable attention due to increased fuel prices
as well as stringent emission restrictions. An increasing use
of biofuels for transport is emerging as an important policy
strategy to replace petroleum fuels. The European Union (EU)
aimed at achieving a 5.75% target of biofuels by 2010 (calcu-
lated on the basis of energy content), set by the EU biofuels
directive and adopted by most Member States in their national
biofuel objectives (Europan Commission [EC], 2003; 2009).
Among alternative fuels, ethanol is one of the fuels employed
most widely, for the following reasons. First, bioethanol can be
produced from various kinds of biomass such as corn, sugar-
cane, sugar beet, cassava, red seaweed, and so on. Second, the
oxygen content of this fuel favors the further combustion of
gasoline. Besides, gasoline blends well with ethanol, com-
pared to diesel, resulting in lower sulfur and aromatics content,

888
Journal of the Air & Waste Management Association, 62(8):888897, 2012. Copyright 2012 A&WMA. ISSN: 1096-2247 print
DOI: 10.1080/10962247.2012.671793
 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897 889

higher octane number, and higher vapor pressure compared to Table 1. Properties of tested fuels
the base fuel.
The quantification of pollutant emissions using ethanol- Gasoline Bioethanol
gasoline blends has mostly been investigated for passenger
Density (kg/m3 at 15
C) 747.8 791.6
cars. Information related to ethanol blended fuel on the emis-
Ethanol content (%vol ) n.a. 99.89
sions of air pollutants from two-wheelers is rather limited
Oxygen content (%wt) 1.46 34.7
(Larsen et al., 2009). Reductions in CO and hydrocarbon (HC)
Total aromatics content (%wt) 33.3
emissions and insignificant effects on NOx emission using a
MTBE (% vol) 8.01
10% v/v ethanolgasoline blend have been observed in a
Water content (%wt) n.a. 0.059
catalyst-free 4-S motorcycle (Jia et al., 2005). The influence of
Sulphur content (mg/kg) 15.3 0.8
ethanolgasoline blends (up to 20% v/v ethanol) at constant
Carbonyl content as acetone (ppmw) n.a. 1177
research octane number, RON (95), has been also studied in a
Vaporization latent heat (kJ/kg) 340 850
catalyst-free 4-S motorcycle (Yao et al., 2009). The outcomes
Lower heating value (kJ/kg) 41,500 26,900
show that CO and NOx emissions decrease with increasing
Reid pressure (kPa) 59.8 16
oxygen content in fuels, while no reduction is observed for the
Research Octane number, RON 97.9 110
HC emissions. Makkee investigated the effect of blends with
Motor Octane number, MON 86.5 92
pure ethanol or hydrous ethanol on a modern 2-S and a typical 4-
Stoichiometric air/fuel ratio 14.6 9
S scooter (Makkee eet al., 2009). He observed with the 4-S
scooter a lowering of CO and HC emissions and fuel consump- Note: n.a., Not available.
tion, an increase of NOx, no effect on particulate matter emitted,
and a reduction of nanoparticles count concentrations, especially
at transient operations. round-robin exercise with two motorcycles, was used as base
There are very few publications on the effects of bioethanol fuel (E0) for the preparation of the gasoline/ethanol splash
gasoline blends on the morphology and size distribution of the blends. Bioethanol, obtained from grape pomace produced dur-
particulate emissions (He et al., 2010; Larsen et al., 2009; Lee ing traditional wine processing, was provided by I.M.A. srl
et al., 2009). Morphological properties of particulate matter are (Trapani, Italy). Selected fuel properties are presented in
important parameters for assessment of its health impacts. The Table 1. Test methods used for gasoline and bioethanol, as a
sizes of both primary and aggregate particles need to be mea- blending component in gasoline, are those specified in EN 228
sured, because the growth of individual primary particles is 2008 and EN 15376, respectively. Three bioethanolgasoline
responsible for the total particulate mass emissions, whereas fuel blends were prepared with 10, 15, and 20% ethanol by
the aggregate particles formed by agglomeration of primary volume in gasoline, named E10, E15, and E20, respectively.
particles or smaller agglomerates that are emitted to the environ-
ment are closely related to human health. However, there is lack Equipment and analysis procedure
of information on the particulate emitted from four-stroke two-
wheelers, especially concerning morphology and size. A brand new four-stroke two-wheeler moped with catalytic
In this study, the effects of ethanol addition (10, 15, and 20% converter was used in this study. The rich air/fuel calibration of
v/v) to commercial gasoline on the regulated and unregulated four-stroke engines may limit the availability of oxygen for
emissions were evaluated in a catalyst equipped four-stroke postcombustion oxidation of HC and CO (Gong et al., 2011;
moped engine. Particulate matter was characterized in terms of Graham et al., 2008; Yao et al., 2011). For this reason a simple
mass (PM) and particle number (PN) emissions, and size dis- passive secondary air injection system (SAI), with a reed valve,
tribution by an electrical low-pressure impactor (ELPI) sampling has been adopted as a standard component upstream of the
system. The emissions of particulate-phase polycyclic aromatic catalyst to provide excess air to the catalyst and to further oxidize
hydrocarbons (PAHs) were also evaluated. Attention was engine-out gases. Table 2 shows the main characteristics of the
focused on the heavier, less volatile PAHs, which are essentially vehicle.
associated with the particulate matter, because they are of greater The schematic layout of the experimental setup is reported in
concern (IARC, 2010). Lighter PAHs, which display a signifi- Figure 1. Moped tailpipe emissions were characterized during
cant gas-phase component, are of lesser toxicological impor- driving cycles performed on a two-wheeler chassis dynam-
tance. Additionally, investigation on the particulate ometer (20-inch single-roll Zollner) with electrical simulation
morphology was performed by a high-resolution transmission of road load and inertia. Regulated pollutants were sampled in
electron microscope (TEM) coupled with a digital image-pro- diluted exhaust gas of a constant-volume sampling system (AVL
cessing/data-acquisition system. Mobile CFV-CVS) positioned downstream of a dilution tunnel.
Sampling bags were filled during driving cycles with diluted
exhaust; at the end of the test they were analyzed to evaluate
Experimental Methods CO, HC, NOx, and CO2. An AVL AMA4000 analyzer bench
Test fuels equipped with a nondispersive infrared detector for CO and CO2
quantification, a chemiluminescence detector for NOx, and a
Four test fuels were used in this study. A commercial gaso- flame ionization detector for HC were used. The particulate
line, provided by ENI S.p.A. in the framework of an Italian mass was determined by gravimetric analysis of a sampling filter
890 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897

Table 2. General parameters of the test vehicle indeno[1,2,3,-cd]pyrene, dibenzo[a,h]anthracene, benzo[g,h,i]

perylene, dibenzo[a,e]pyrene, coronene, dibenzo[a,h]pyrene,
Items Parameters dibenzo[a,i]pyrene, dibenzo[a,l]pyrene.
The particulate morphology was analyzed by a high-
Overall length 1755 mm
resolution TEM (Philips CM12). Three-millimeter TEM grids
Width 740 mm
were located on the four ELPI impaction stages 1, 3, 5, and 7
Wheelbase 1290 mm
having cutpoints (50%) at 0.02, 0.07, 0.2, and 0.5 m, respec-
Engine type Single cylinder, four-stroke, four valves
tively. Copper 300-mesh grids coated with carbon films were
Bore/stroke 39/41.8 mm
used as substrates. The TEM photographs were randomly taken
Displacement 49.9 cc
at various locations across the grid surface and digitized with
Compression ratio 12  0.5:1
image-processing software (Image ProPlus 6.0, Media
Maximum power 3.2 kW at 8250 rpm
Cybernetics, USA) to analyze the mean primary particle dia-
Carburetor Electronic Keihin NCV 20
meter and agglomerate dimensions.
Ignition Electronic inductive discharge, variable
spark advance
Starter Electric, kick starter Driving cycles
Cooling system Forced air
Exhaust system SAI (secondary air injection) two-way Tailpipe emission measurements were carried out according
catalytic converter to the European driving cycle known as ECE R47 provided for
Homologation Euro 2 type-approval of mopeds. The cycle is composed by eight ele-
Mileage 1500 km mentary modules with maximum speed of 45 km/hr: The first
four modules are considered as the warming-up phase and the
last four as the sampling phase. Euro 2 legislation provides a
limit for CO (1 g/km) and for the sum of HC and NOx (1.2 g/km),
both evaluated during the hot phase.
(47-mm Pallflex T60A20) following the same legislative proce-
The regulated emissions and particulate matter were evalu-
dure prescribed for diesel vehicles. An ELPI was used for mea-
ated during both the warming-up phase and the hot phase of the
suring particle number distribution, over 12 collecting stages, in
ECE-47 cycle to evaluate the cold-start influence on moped
the range 7 nm10 mm. The ELPI was positioned downstream of
emissions. At least three replicate tests were performed for
a further double-stage partial sampling system (Dekati FPS),
each fuel to evaluate the determination variability of the regu-
which worked with a dilution factor of 10. After each test, the
lated emissions, fuel consumption, PM, and PN. No repetition
12 collecting filters were divided in three groups characterized
was conducted for determination of particulate-phase PAHs and
by the following dimensions: Group 1 (7 nm0.12 mm), Group 2
TEM evaluations, which were carried out on particulate samples
(0.12 mm7.6 mm), and Group 3 (7.6 mm10 mm). These three
collected in a single test for each tested fuel.
fractions were separately extracted in a Soxhlet extractor using
cyclohexane in order to remove the organic fraction. The
extracted phase was concentrated by a rotary evaporator and
analyzed for PAHs quantification using an HP 5890 gas chro-
Results and Discussion
matograph coupled to an HP 5971 mass spectrometer operated in Regulated emissions
selected ion monitoring (SIM) mode. The PAH species deter-
mined were the following: phenanthrene, anthracene, fluor- Table 3 shows the mean values and the relative standard
anthene, pyrene, 3,4-benzofluorene, benz[a]anthracene, deviations (SD) of the emission factors of regulated pollutants
chrysene, 5-methylchrysene, benzo[b k]fluoranthene, benzo and CO2, expressed as mass emitted per kilometer traveled, and
[j]fluoranthene, benzo[e]pyrene, benzo[a]pyrene, perylene, the fuel consumptions during both phases of the driving cycle.

Figure 1. Schematic of experimental setup.

 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897 891

Table 3 reports also the sum of HC and NOx in order to compare

5.23E11
2.08E12
3.39E12
1.40E12
7.92E12
5.43E12
2.49E12
5.89E11
the experimental data with the Euro 2 standard limits.

SD
For each tested fuel the emissions of CO and HC during the
cold phase were markedly higher than those observed during the
PN
Table 3. Mean emissions of regulated pollutants (g/km), fuel consumption (mL/km), PM (mg/km), and PN (1/km) during the cold and hot phase of the ECE R47 driving cycle for the tested fuels
hot phase. This is due to the fact that during the cold start the

1.03E13
5.96E12
1.14E13
3.80E12
1.26E13
8.67E12
1.30E13
6.64E12
engine and the catalytic converter were not at their optimal
Mean operating conditions and higher emissions of unburned products
occurred. The NOx emissions were slightly higher during the hot
phase, and this is attributable to the higher temperatures in the
combustion chamber that facilitated the formation of this
3.76
2.07
4.70
0.62
1.74
0.74
2.44
1.52
SD

pollutant.
As shown, only with E15 and E20 did the hot CO emissions
PM

comply with the Euro 2 mass limit of 1 g/km, whereas the

Mean
11.41
3.16
17.46
2.51
9.20
4.76
7.17
3.61
combined hot mass levels of HC and NOx were below the Euro
2 limit of 1.2 g/km for all tested fuels with a reasonable margin.
According to the driver, there was no problem of drivability with
Fuel consumption

E10 and E15, whereas with E20 the engine showed an instable
SD
1.0
1.9
1.6
3.0
1.4
1.5
0.8
0.8

idle quality. Hence it must be mentioned that before testing E20,

the engine was tuned in a slightly richer condition by turning the
idle mixture adjustment screw on the carburetor.
Mean
31.1
28.3
30.4
27.5
26.9
25.0
25.5
24.9

In terms of the effect of ethanol on the regulated emissions, a

sensible reduction of CO was visible over both phases of driving
cycle and for almost all the fuels. With the exception of cold start
data of E10, which gave CO emissions almost 10% higher than
0.25
0.05
0.25
0.04
0.10
0.01
0.07
0.06
SD
HC NOx

gasoline, in all the other conditions the reduction ranged between

10 to 70%. The highest CO reductions obtained with E15 and
Mean

E20 were statistically significant according to one-way analysis

1.28
0.65
1.68
0.77
1.45
0.87
1.47
0.78

of variance (F-test) at 95% confidence. The CO emission reduc-

tions are due to the fact that ethanol has less carbon and more
oxygen than gasoline. allowing an increase of the oxygen/fuel
0.03
0.04
0.06
0.04
0.08
0.09
0.01
0.01
SD

ratio and, consequently, the combustion process is tuned to be

NOx

completed (Bayraktar, 2005; Hsieh et al., 2002; Larsen et al.,

Mean

2009; Najafi et al., 2009; Wu et al., 2004). On the contrary, there

0.29
0.25
0.34
0.31
0.34
0.33
0.34
0.30

was an increase of HC and NOx emissions with ethanol use. This

increase ranged between 10 and 35% for HC and between 20 and
30% for NOx. Despite almost all these variations not being
0.22
0.02
0.21
0.04
0.07
0.08
0.07
0.07
SD

statistically significant due to the high dispersion of data, the

HC

increase of emissions related to the bioethanol blends use is

Mean
0.99
0.40
1.34
0.46
1.10
0.53
1.12
0.48

indisputable, especially for NOx.

The explanation of the different influence of the ethanol
addition on CO and HC emissions can be found in the lowering
2.33
4.60
4.31
4.73
3.48
3.16
1.99
2.26

of exhaust temperature due to high latent heat of vaporization of

Note: Underlined CO values exceed Euro 2 standard limits.
SD

ethanol. Figure 2 reports the temperatures monitored upstream of

CO2

the converter for all tested fuels during the ECE R47 cycle. The
Mean
65.59
63.23
62.67
62.08
59.19
57.69
57.11
58.16

vehicle is equipped with a Pt-based catalyst coupled with an SAI

system, which provides additional oxygen for CO and HC com-
bustion over the catalyst. So, despite the higher content of oxy-
0.38
0.12
0.52
0.47
0.11
0.05
0.16
0.21

gen in the exhaust when the gasolineethanol blends were used,

SD

the consequent decreasing of the exhaust temperatures disadvan-

CO

taged the subsequent oxidation reactions at the converter.

Mean
3.53
2.04
3.97
1.79
1.39
0.56
1.25
0.70

As shown in Figure 2, the measured temperatures for E10,

E15, and E20 were lower than those for E0 (<30
C for E10 and
E15 and almost <50
C for E20). Consequently, the lower tem-
Phase

peratures attained by the converter with gasolineethanol blends

Cold

Cold

Cold

Cold
Hot

Hot

Hot

Hot

were sufficiently high for the oxidation of CO but not high

enough for the oxidation of HC, which needs higher tempera-
tures to be promoted (Gong et al., 2011). In addition, the test
Fuel

E10

E15

E20
E0

vehicle is equipped with a carburetor as the fuel injection system

892 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897

Table 3. The cold-start particulate matter emissions were mark-

edly higher than for hot emissions for all the tested fuels for
various reasons: incomplete combustion due to low combustion
cylinder temperatures, the catalyst not yet fully heated and lit off,
and the engine running a fuel-rich mixture to assist combustion
during cold engine conditions. Moving from the cold to the hot
phase the PM emissions decreased by 5085% and the PN
emissions by 3065%. Running with E15 and E20, during the
hot phase, seems slightly to increase both PM and PN emissions
compared to gasoline even if the variations among the different
fuel formulations are not statistically significant according to
one-way analysis of variance (F-test) at 95% confidence.
However, one explanation of the observed behavior could be
found in the lower temperatures reached by the engine and,
consequently, by the catalyst (Figure 2) with addition of ethanol
with respect to the gasoline, which led to reducing the combus-
tion and catalyst conversion efficiencies, and consequently
Figure 2. Temperature profiles upstream of the catalytic converter.
higher amounts of unburned carbonaceous particles were
emitted. As an example, the profiles of PN emitted during an
entire driving cycle are reported in Figure 3 for E0 and E20.
The number reported is the instantaneous sum of all particles
and the engine is more sensitive to changes that affect the air/fuel
measured at 12 ELPI collecting stages. As shown, the PN emis-
ratio compared to the newer scooters equipped with closed-loop
sion profile followed the speed profile with peaks that occurred
fuel injection systems and with adaptive learning functions.
during the acceleration phases of driving cycle. Power require-
Consequently, the amount of fuel supply cannot be adjusted
ment during acceleration involves the enrichment of fuel mixture
immediately according to the combustion conditions. This also
that produces an increasing number of carbonaceous particles,
resulted in an incomplete combustion and partially contributed
whereas a lean combustion mixture (e.g., during deceleration)
to higher HC emissions.
reduces soot formation (Prati et al., 2009). Representative size
The higher NOx emissions obtained with gasolineethanol
distributions of the particulate emitted during the first, third,
blends can be explained by the particular exhaust after-treatment
sixth, and eighth modules of the driving ECE R47 cycle are
configuration of the tested moped, which is designed to complete
reported in Figure 4 for the four tested fuels.
the oxidation reactions. Reduction reaction for NOx abatement
In these figures the number of particles normalized respect to
is, indeed, not controlled at the exhaust; NOx should be reduced
the particle dimensional range (dN/dlogDp) is reported versus
only during rich operative engine conditions by taking place in
the aerodynamic diameter. The aerodynamic diameter is defined
the oxidation reaction of CO and HC as a reducing agent. When
as the diameter of a unit density sphere that has the same settling
the engine was fueled with ethanol blends, the major content of
velocity as the particle in question (Baron and Willeke, 1993).
oxygen coming from the fuel enhanced the pollutant oxidation
As shown, the number of particles emitted in the first module
process but disadvantaged NOx reduction.
was markedly higher than those obtained during the hot modules
As shown in Table 3, the volumetric fuel consumption, esti-
mated by carbon balance, decreased as the ethanol content
increased (up to 15% with E20). The reduction was observed
during both phases of the driving cycle. The fuel consumption
reduction is a consequence of the increasing of the engine
thermal efficiency with the increasing ethanol percentage in
gasoline (Al-Hasan, 2003). Considering the CO2 emission,
Table 3 shows that the CO2 emission decreased by 2.3, 9.3,
and 10.5% with E10, E15, and E20, respectively, in comparison
to gasoline. The CO2 reduction is due to the reduction of the
brake-specific fuel consumption discussed earlier and to the
lower carbon content of ethanol than gasoline.

Particulate emissions
The mean values with relative SDs of the total PM mass
collected on filters, expressed as milligrams per kilometer, and
of the total number of particles (PN) with diameter in the range
7 nm10 mm, expressed as number per kilometer, measured
during both phases of the ECE R47 cycle are reported in Figure 3. Particle number emissions during the test driving cycle for E0 and E20.
 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897
Figure 4. Particle size distribution during the 1st, 3rd, 6th and 8th module of the ECE R47 driving cycle.
for the reasons discussed earlier. For all the tested fuels, more
than 98% of the particles emitted had an equivalent aerodynamic
diameter less than 0.5 mm during the cold phase and lower than
0.1 mm during the hot phase. Nucleation mode (710 nm) was
not visible in these experiments due to high and hot dilution of
sampled flow (Prati et al., 2009).
PAH emissions
The distribution of particulate-phase PAH emissions,
expressed as mass emitted per kilometer, in the three selected
dimensional groups is reported in Table 4 over an entire driving
cycle. The results related to E10 are not reported since for this
blend some of the samples were lost. As shown, PAHs were
mainly found on submicrometer-size particles (<0.1 mm). The
observed PAH size distribution is justified by the exhaust PAH
source, mainly originating frp, the unburned fuel and lubricating
oil occurring during incomplete combustion process.
Thus, both particulate matter and associated PAHs were
mainly emitted in the respirable size range in accordance with
previous studies concerning the particle-phase PAHs distribution
in the exhaust of mopeds (Spezzano et al., 2008; Yang et al.,
2005).
The use of ethanol demonstrates a positive effect on PAH
emissions mainly related to lower content of aromatics in the
injected fuel (de Abrantes et al., 2009). Experimental data of
893
this work showed a reduction by about 17% of total PAHs when
comparing E15 with E0, while the total PAH emission related
to E20 was comparable with that of gasoline, probably due to
the light enrichment combustion tuning operated on the engine
moped before starting E20 tests. Low-molecular-weight PAHs,
especially phenanthrene, were dominant in all the three dimen-
sional groups. This aspect was already observed at the exhaust
of a 4stroke motorcycle by Yang et al. (2005).
In order to evaluate the potential carcinogenicity of the ana-
lyzed PAHs for various size ranges the toxic equivalent factors
(TEFs) were used. TEFs were devised to compare the carcino-
genic potency of the individual PAH with the carcinogenicity of
benzo[a]pyrene (expressed as BaPeq), which is considered to be
one of the most toxic of the PAHs and is often used as an
indicator for this group of compounds. To date, only a few
proposals for the TEFs for PAHs are available; a good summary
of TEFs reported in literature is provided by Law et al. (2002).
Among them, the list of the TEFs proposed by Nisbet and LaGoy
(1992) seems to better reflect the actual state of knowledge on
the human health risk associated with inhalatory exposure to
PAH mixtures. Thus, the TEFs reported in Nisbet and LaGoy
(1992) were used in this study: BaPeq emission rates for each
individual PAH were computed by multiplying the emission
factor with the corresponding TEF value (Lee et al., 2002).
Table 4 shows the total BaPeq emission factors for various size
ranges and for each fuel. The BaPeq emission factors for all the
tested fuels were much higher in the finer size particles
 894

Table 4. Particulate-phase PHAs and BaPeq emission factors in mg/km for size ranges during the overall ECE R47 driving cycle

E0 E15 E20
PHA Group 1 Group 2 Group 3 Total Group 1 Group 2 Group 3 Total Group 1 Group 2 Group 3 Total IARC classa
Phenanthrene 2.56 1.52 0.83 4.92 0.51 1.93 1.87 4.31 0.57 1.26 1.74 3.56 3
Anthracene 0.77 0.35 0.14 1.26 0.14 0.35 1.14 1.62 0.02 0.11 0.21 0.33 3
Fluoranthene 0.85 0.76 0.73 2.33 0.43 0.66 0.78 1.87 0.26 0.40 0.58 1.24 3
Pyrene 0.86 0.74 0.49 2.09 0.41 0.59 0.66 1.65 0.37 0.56 0.73 1.67 3
3,4-benzofluorene 0.00 0.00 0.00 0.00 0.00 0.00 0.11 0.11 0.15 0.11 0.05 0.31 3
Benzo(a)anthracene 1.36 0.29 0.25 1.91 1.11 0.40 0.24 1.75 1.39 0.39 0.64 2.42 2A
Chrysene 1.38 0.21 0.06 1.65 0.76 0.53 0.21 1.51 1.35 0.45 0.69 2.49 3
5-methylchrysene 0.41 0.15 0.00 0.56 0.44 0.00 0.00 0.44 0.33 0.07 0.21 0.61 3
Benzo(b k)fluoranthene 3.38 0.62 0.07 4.07 2.12 0.61 0.27 2.99 5.17 2.13 1.08 8.38 2B
Benzo(j)fluoranthene 1.06 0.25 0.08 1.38 0.61 0.28 0.03 0.92 0.25 0.00 0.00 0.25 2B
Benzo(e)pyrene 1.20 0.40 0.27 1.87 1.01 0.56 0.32 1.89 2.41 0.74 0.41 3.55 3
Benzo(a)pyrene 1.85 0.39 0.09 2.33 1.16 0.29 0.14 1.58 0.65 0.39 0.31 1.35 2A
Perylene 1.37 0.43 0.32 2.12 1.06 0.55 0.41 2.02 0.60 0.20 0.14 0.94 3
Indeno(1,2,3,c,d)pyrene 1.50 0.32 0.00 1.82 0.87 0.25 0.00 1.11 2.25 0.40 0.13 2.79 2B
Dibenzo(a,h)anthracene 0.45 0.00 0.00 0.45 0.00 0.00 0.00 0.00 0.44 0.00 0.00 0.44 2A
Benzo(g,h,i)perylene 0.29 0.43 0.00 0.71 1.35 0.38 0.00 1.73 2.41 0.30 0.19 2.91 3
Dibenzo(a,e)pyrene 0.70 0.00 0.00 0.70 0.00 0.00 0.00 0.00 0.70 0.24 0.00 0.94 2B
Coronene 1.12 0.18 0.00 1.30 0.78 0.00 0.00 0.78 0.90 0.15 0.13 1.18 3
Dibenzo(a,h)pyrene 0.35 0.00 0.00 0.35 0.00 0.00 0.00 0.00 0.48 0.00 0.00 0.48 2B
Dibenzo(a,i)pyrene 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2B
Dibenzo(a,l)pyrene 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 2B
Total PHAs 21.45 7.04 3.33 31.81 12.74 7.38 6.18 26.30 20.70 7.90 7.23 35.83
Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897

Total BaPeq 3.44 0.53 0.12 4.09 1.59 0.43 0.21 2.22 2.63 0.74 0.51 3.88
Notes: The PHAs probably carcinogenic to humans are reported in bold.
a
Classification for carcinogenic risk by IARC (2010): 1: carcinogenic to humans; 2A: probably carcinogenic to humans; 2B: possibly carcinogenic to humans; 3: not classifiable.
 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897 895

(<0.1 mm), with increasing toxicity to humans. These results are
in accordance with earlier studies (Spezzano et al., 2008; Yang
et al., 2005). The total BaPeq emission factors obtained in this
study ranged from 2.2 to 4.1 mg/km.
TEM investigation
TEM analysis was performed on the particles samples collected
during an entire driving cycle at the four ELPI impaction stages: 1,
3, 5, and 7. Based on the TEM observations, no characteristic
morphological difference was detected between the particulates
collected for the different fuels. Representative TEM images of
particulate sampled for E0 and E15 are shown in Figure 5.
The aggregates were clearly nonspherical and demonstrated a
wide variety of chain and grape-like agglomerated structures of
small spherical primary particles (spherules). Magnifications of
75,000 and 13,000 were used to evaluate the size character-
istics of spherules and aggregates, respectively. For each sample
collected at the four ELPI stages, about 100 primary particles,
clearly distinguishable in the agglomerates, were randomly
selected and their diameters were measured with the aid of
Image ProPlus software. As examples, Figure 6 show the pri-
mary particle diameter distributions for E0 and E15 relative to
particulates collected at ELPI stage 3.
For all the tested fuels the spherule diameter distributions
were relatively narrow with mean diameters, dp, in the range of
17.5 and 32.5 nm. This is the first direct measurement of primary
particle sizes for particulates from gasoline and bioethanol
gasoline blends. These diameters are in good agreement with
those measured for diesel particulates (1935 nm) (Chandler
et al., 2007; Lee at al., 2002; Park et al., 2004; Zhu et al.,
2005). For the measurement of agglomerate sizes, 6070
agglomerates were randomly selected for each sample and their
size was quantified on the basis of mean diameter automatically
calculated as the average length of diameters measured at 2-
degree intervals and passing through the aggregates centroid.

Figure 5. Typical TEM images of particles collected at the ELPI impaction stage 3 collected during an entire driving cycle with E0 (gasoline) and E15 (75,000x).

Figure 6. Primary particle size distributions for particulates collected at stage 3 with E0 and E15.
896 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897
In contrast to primary particles, the aggregates composed of tens
to thousands of spherules showed generally broader size distri-
butions (polydisperse), which generally didnt correspond to the
relative impaction stage cutpoints. This is attributable to both
fragmentation and agglomeration phenomena occurring upon
impact, suggesting that inertial sampling by the low-pressure
impactor causes these artifacts (Park et al., 2004).
Conclusions
This study investigated the effects of ethanol as a gasoline
replacement component, at different levels up to 20% by volume
(E10, E15, and E20), on regulated and unregulated emissions in
a catalyzed 4-S moped, commonly used in urban areas.
There was no problem of drivability with E10 and E15,
whereas with E20 the engine showed an unstable idle quality.
Cold emissions for regulated and unregulated compounds were
markedly higher than those obtained in hot driving conditions,
except for NOx. The ethanol added into gasoline led to a sig-
nificant reduction of CO emission (6070% with E15 and E20)
and moderate diminutions of CO2 (<10%) and fuel consumption
(<15%) during the complete ECE R47 driving cycle. In contrast,
ethanol/gasoline blends did not reduce HC and NOx emissions.
No clear and unique effect on PM and PN emissions was
observed with ethanol addition. For all the tested fuel formula-
tions, solid particle numbers emitted fall in the range 8  1012 to
1  1013/km. Particulate emissions occur mainly during the
acceleration phase, which corresponds to a very rich combustion
mixture. Aerodynamic diameter mass distribution derived from
the 12 ELPI impaction stages showed that more than 95% of the
total collected mass had an equivalent aerodynamic diameter
<0.1 mm, and the hot emissions showed generally lower mean
aerodynamic diameters than those relative to the cold emissions.
A reduction by about 17% in the total particulate-phase PAH
emission was observed comparing E15 with E0, while E20
showed values comparable with those of gasoline. This can be
attributed to the changes in combustion conditions in the car-
buretor engine when E20 was used. Particulate-phase PAHs were
mainly found on submicrometer-size particles (<0.1 mm), which
penetrate deeply into the human respiratory tract.
The TEM analysis showed no relevant effect of the ethanol
use on the particulate morphology, showing chain or grape-like
aggregates composed of primary spherical particles with mean
diameters in the range 17.532.5 nm.
Acknowledgments
The authors thank Drs. Laura Ghetti, Mattia Di Benedetto,
and Antimo Buccino for their support in carrying out the tests.
References
Al-Hasan, M. 2003. Effect of ethanolunleaded gasoline blends on engine perfor-
mance and exhaust emission. Energy Conversion Manage.44:154761.
doi:10.1016/S0196-8904(02)00166-8
Associazione Nazionale Ciclo Motociclo Accessori (ANCMA). 2011. Statistics of
in-use mopeds and motorcycles 2011. http://www.ancma.it/it/publishing.asp.
Automobile Club dItalia (ACI). 2009. Annuario statistico 2009. http://www.aci.
it/sezione-istituzionale/studi-e-ricerche/dati-e-statistiche.html.
Baron, P. A., and K. Willeke, eds. 1993. Aerosol Measurement: Principles,
Techniques, and Applications. New York: Van Nostrand Reinhold.
Bayraktar, H. 2005. Experimental and theoretical investigation of using gasoline
ethanol blends in spark-ignition engines. Renew. Energy 30:173347.
doi:10.1016/j.renene.2005.01.006
Chandler, M.F.; Teng, Y.; Koylu, U.O. 2007. Diesel engine particulate emissions:
A comparison of mobility and microscopy size measurements. Proc. Comb.
Inst. 31:297179. doi:10.1016/j.proci.2006.07.200
de Abrantes, R., J. V. de Assunao, C. R. Pesquero, R. E. Bruns, and R. P. Nobrega.
2009. Emission of polycyclic aromatic hydrocarbons from gasohol and ethanol
vehicles. Atmos. Environ. 43:64854. doi:10.1016/j.atmosenv.2008.10.014
European Commission. 2003. Directive 2003/30/EC on the Promotion of the Use
of Biofuels or Other Renewable Fuels for Transport. http://eur-lex.europa.eu/
JOHtml.do?uri=OJ:L:2003:123:SoM:EN:HTML
European Commission. 2009. Directive 2009/28/EC of the European Parliament
and of the Council of 23 April 2009 on the Promotion of the Use of Energy
From Renewable Sources and Amending and Subsequently Repealing
Directives 2001/77/EC and 2003/30/EC. http://eur-lex.europa.eu/JOHtml.
do?uri=OJ:L:2009:140:SOM:EN:HTML
Gong, C., K. Huang, B. Deng, and X. Liu. 2011. Catalyst light-off behavior of a
spark-ignition LPG (liquefied petroleum gas) engine during cold start.
Energy 36: 5359. doi:10.1016/j.energy.2010.11.026
Graham, L. A., S. L. Belisle, and C.-L. Baas. 2008. Emissions from light duty
gasoline vehicles operating on low blend ethanol gasoline and E85. Atmos.
Environ. 42: 4498516. doi:10.1016/j.atmosenv.2008.01.061
He, X., J. C. Ireland, B. T. Zigler, M. A. Ratcliff, K. E. Knoll, T. L. Alleman, and
J. T. Tester. 2010. The Impacts of Mid-Level Biofuel Content in Gasoline on
SIDI Engine-Out and Tailpipe Particulate Matter Emissions. SAE Paper
2010-01-2125. Troy, MI: Society of Automotive Engineers.
Hsieh, W. D., R. H. Chen, T. L. Wu, and T.-H. Lin. 2002. Engine performance and
pollutant emission of an SI engine using ethanolgasoline blended fuels.
Atmos. Environ. 36: 40310. doi:10.1016/S1352-2310(01)00508-8
IARC. 2010. Some Non-Heterocyclic Polycyclic Aromatic Hydrocarbons and
Some Related Exposures. IARC Monogr. Vol. 92. Lyon: IARC.
Jia, L. W., M. Q. Shen, J. Wang, and M. Q. Lin. 2005. Influence of ethanolgasoline
blended fuel on emission characteristics from a four-stroke motorcycle engine.
J. Hazard. Mater. 123:2934. doi:10.1016/j.jhazmat.2005.03.046
Larsen, U., J. Troels, and J. Schramm. 2009. Ethanol as a Fuel for Road
Transportation. Main Report EFP06. IEA Implementing Agreement on
Advanced Motor Fuels. Technical University of Denmark. May.
Law, R. J., C. Kelly, K. Baker, J. Jones, A. D. McIntosh, and C. F. Moffat. 2002.
Toxic Equivalency Factors for PAH and Their Applicability in Shellfish
Pollution Monitoring Studies. J. Environ. Monit. 4:38388. doi:10.1039/
B200633M
Lee, K. O., R. Cole, R. Sekar, M. Y. Choi, J. S. Kang, C. S. Bae, and H. D. Shin.
2002. Morphological investigation of the microstructure, dimensions, and
fractal geometry of diesel particulates. Proc. Combust. Inst. 29:64753.
doi:10.1016/S1540-7489(02)80083-9
Lee, H., C. L. Myung, and S. Park. 2009. Time-resolved particle emission and
size distribution characteristics during dynamic engine operation conditions
with ethanol-blended fuels. Fuel 88(9): 16801686. doi:10.1016/j.
fuel.2009.03.007
Makkee, M., J. Czerwinski, and P. Comte. 2009. The Effect of (Hydrous) Ethanol on
the Emission and Performances of 2- and 4-Stroke Scooters. (Particle) Emissions
of 2-Stroke ScootersScience, Problems, Solutions & Perspectives. SAEEC
JRC Joint Conference. Monza, Milano, Italy, June 1112.
Najafi, G., B. Ghobadian, T. Tavakoli, D. R. Buttsworth, T. F. Yusaf, and M.
Faizollahnejad. 2009. Performance and exhaust emissions of a gasoline
engine with ethanol blended gasoline fuels using artificial neural network.
Appl. Energy 86:63039. doi:10.1016/j.apenergy.2008.09.017
Nisbet, C., and P. LaGoy. 1992. Toxic equivalency factors (TEFs) for polycyclic
aromatic hydrocarbons (PAHs). Regul. Toxicol. Pharmacol. 16:290300.
doi:10.1016/0273-2300(92)90009-X
Park, K., D. B. Kittelson, and P. H. McMurry. 2004. Structural properties of
diesel exhaust particles measured by transmission electron microscopy
(TEM): Relationships to particle mass and mobility. Aerosol Sci. Technol.
38:88189. doi:10.1080/027868290505189
 Seggiani et al. / Journal of the Air & Waste Management Association 62 (2012) 888897
Prati, M. V., and M. A. Costagliola. 2009. Emissions of fine particles and organic
compounds from mopeds. Environ. Eng. Sci. 26:11122. doi:10.1089/
ees.2007.0206
Spezzano, P., P. Picini, D. Cataldi, F. Messale, and C. Manni. 2008. Particle- and gas-
phase emissions of polycyclic aromatic hydrocarbons from two-stroke, 50 cm3
mopeds. Atmos. Environ. 42:433244. doi:10.1016/j.atmosenv.2008.01.008
Wu, C. W., R. H. Chen, J. Y. Pu, and T.-H. Lin. 2004. The influence of airfuel
ratio on engine performance and pollutant emission of an SI engine using
ethanolgasoline-blended fuels. Atmos. Environ. 38:7093100. doi:10.1016/
j.atmosenv.2004.01.058
Yang, H. H., S. M. Chien, M. R. Chao, and C. C. Lin. 2005. Particle size
distribution of polycyclic aromatic hydrocarbons in motorcycle exhaust emis-
sions. J. Hazard Mater. 125, 15459. doi:10.1016/j.jhazmat.2005.05.019
Yao, Y. C., J. H. Tsai, and H. L. Chiang. 2009. Effects of ethanol blended gasoline
on air pollutant emissions from motorcycle. Sci. Total Environ. 407(19):
525762. doi:10.1016/j.scitotenv.2009.06.017
Yao, Y.-C., J.-H. Tsai, and H.-L. Chiang. 2011. Emissions of organic air toxics
from a four-stroke motorcycle using ethanol-blended gasoline. Environ. Eng.
Sci. 28: 14758. doi:1089/ees.2010.0255
897
Zhu, J., K. O. Lee, A. Yozgatligil, and M. Y. Choi. 2005. Effects of engine
operating conditions on morphology, microstructure, and fractal geometry
of light-duty diesel engine particulates. Proc. Comb. Inst. 30:278189.
doi:10.1016/j.proci.2004.08.232
About the Authors
Maurizia Seggiani is research assistant in industrial and technological chemistry
at the Department of Chemical Engineering of the University of Pisa (Italy). Her
main research areas are waste management, renewable energy sources, biomass,
biofuels, and modeling/simulation of chemical processes.
M. Vittoria Prati and M. Antonietta Costagliola are affiliated with the Istituto
Motori CNR, Napoli, Italy.
Monica Puccini and Sandra Vitolo are members of the Department of Chemical
Engineering, Industrial Chemistry and Material Science, University of Pisa, Pisa,
Italy.