Renewable and Sustainable Energy Reviews 47 (2015) 162172

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Effects of using renewable fuels on vehicle emissions

Larry G. Anderson 1
Department of Chemistry, University of Colorado Denver, Denver, CO 80217-3364, USA

art ic l e i nf o

a b s t r a c t

Article history:
Received 10 September 2014
Accepted 1 March 2015
Available online 21 March 2015

In an effort to reduce greenhouse gas emissions, renewable fuels (biofuels) are being developed and used
in motor vehicles. The most common of these are bioethanol and biodiesel that are blended with
petroleum based gasoline and diesel fuels at varying percentages. This provides some reduction in the
fossil carbon emissions from these fuels. Biomethane (produced from biogas) can and is used as a vehicle
fuel to a much more limited extent. Signicant modications are required for the motor vehicle to use
biomethane.
The use of bioethanol blended fuels leads to increases in emissions of formaldehyde and
acetaldehyde, with an accompanying decrease in benzene emissions. Nitrogen oxides (NOx) emissions
are often increased with the use of bioethanol. The use of biodiesel blended fuels generally leads to
increased NOx emissions, and decreases in particulate matter (PM) emissions. The use of biomethane is
expected to have little adverse impact on the emissions from the combustion process, but it will lead to
increased emissions of methane due to leakage and unburned fuel emissions. Methane is a stronger
greenhouse gas than CO2, which is the greenhouse gas emitted in largest quantities from all combustion
processes. With the appropriate time and effort, operating and emission control technologies can be
developed to minimize the adverse effects that biofuel use will have on vehicle emissions. But this will
require that vehicles be properly designed for the use of specic biofuels or biofuel blends. Renewable
fuels may lead to decreases in greenhouse gas emissions, even with increasing use of transport fuels. No
improvement, and more likely deteriorating air quality, is expected in the future with increasing
renewable fuel use.
& 2015 Elsevier Ltd. All rights reserved.

Keywords:
Transport
Biofuels
Bioethanol
Biodiesel
Biomethane
Compressed natural gas

Contents
1.
2.
3.
4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bioethanol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Biodiesel. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Biomethane (natural gas) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.
Buses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.1.
Exhaust after-treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.2.
Bus test cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.3.
Meeting current and future emissions standards for transit buses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2.
Diesel and CNG trucks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.
Light-duty CNG vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4.
Natural gas composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

E-mail address: Larry.Anderson@ucdenver.edu

Tel.: 1 303 556 2963.

http://dx.doi.org/10.1016/j.rser.2015.03.011
1364-0321/& 2015 Elsevier Ltd. All rights reserved.

163
163
164
164
164
166
167
168
168
169
169
170
170

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

1. Introduction
Currently there are about one billion light-duty motor vehicles in
operation throughout the world, with approximately 75% in the
developed, OECD nations. The number of light-duty, personal vehicles
is expected to grow from 0.8 billion in 2010 to 1.7 billion by 2040;
approximately 80% of this growth is expected to occur in the
developing, non-OECD nations. This growth in the vehicle population
is expected to be largely offset by the improved vehicle fuel economy,
so that future light-duty vehicle fuel demand is only expected to
increase slightly (about 5%) [1]. The demand for commercial transport
is also expected to substantially increase by 2040, and the fuel
demand for heavy-duty vehicles (trucks and buses) is expected to
increase by about 70%. Currently, heavy-duty vehicle fuel demand is
about 80% of that for light-duty vehicles, however by 2040 it is
expected to be about 130% [1].
Transportation sources currently constitute about 15% of the
global greenhouse gas emissions, although it is a more important
contributor in many areas of the world. In an effort to reduce
greenhouse gas emissions, renewable fuels (biofuels) are being
developed and used in motor vehicles; the most common are
bioethanol and biodiesel, which are blended with petroleum based
gasoline and diesel fuels, at varying percentages. Use of biofuels
provides some reduction in the fossil carbon emissions from
vehicles. At this time, biofuel use accounts for only about 3.4% of
the on-road transportation fuel used globally. About 80% of the
transportation biofuel is bioethanol, and about 20% is biodiesel;
total biofuel production actually declined slightly during 2012 [2].
Small percentage biofuel blends have been used in motor vehicles,
with little modication to the vehicle. The assumption has been
that the use of these fuels would have little adverse effects on air
quality.
Natural gas used as a vehicle fuel may have slightly lower
greenhouse gas emissions than gasoline or diesel fuel [3], and may
be useful for initiating development of infrastructure, which could
ultimately serve for biomethane distribution as a vehicle fuel. Fossil
natural gas is clearly not a sustainable vehicle fuel. As biomethane
(from biogas) production increases, it can be blended in increasing
quantities with fossil natural gas to supply gas needs. Natural gas is
used extensively as a vehicle fuel in a few countries, and is used to a
small extent in many more. For example, Sweden had over 38,000
natural gas vehicles (NGVs) in 2012, and biomethane accounted for
over 60% of the fuel used in these vehicles [4].
This review discusses vehicle emissions and air quality impacts
for bioethanol, biodiesel and biomethane vehicle fuels. The discussion of bioethanol and biodiesel will be based on previous
publications [5,6], and will be supplemented by more recent
information. The effects of biomethane use will be assessed by
reviewing the effects of compressed natural gas (CNG) and liquid
natural gas (LNG) on vehicle emissions.

2. Bioethanol
Ethanol blended fuels have been used extensively in Brazil as a
vehicle fuel since the 1980s [5]. By 1983, most new light-duty (nondiesel) vehicles produced in Brazil were built to run on hydrous
ethanol, rather than gasoline or gasohol (a gasolineethanol blend).
By 1990, sugar prices had increased and fuel ethanol production had
dropped to the point Brazil was forced to import ethanol for fuel use.
The high cost and availability issues for this ethanol fuel ended the
production of ethanol-only vehicles by the mid-1990s. In 1993, Brazil
enacted a law that required all gasoline be blended with at least 20
25% ethanol. Around 2003, Brazil started signicant production of
ex-fuel vehicles that could operate on the E20 blend (gasohol),
ethanol, or any mix of these fuels. This permits these newer vehicles

163

to use the least expensive fuel available at the time. The vehicle
emissions data from Brazil generally showed that nitrogen oxides
(NOx) emissions were higher from gasohol than ethanol fueled
vehicles, and the aldehyde emissions were higher from ethanol than
gasohol fueled vehicles [5]. The emissions of acetaldehyde are higher
than formaldehyde with all ethanol blended fuels.
The resulting concentrations of both formaldehyde and acetaldehyde observed in Brazil are generally higher than observed in
other countries [5]; acetaldehyde, formaldehyde and nitrogen
oxides are all important initiators of photochemical air pollution
problems such as ozone. Severe ozone air quality problems have
been observed in a number of cities in Brazil. The concentrations
of ozone precursors and ozone have decreased in recent years
from more stringent vehicle emissions standards, but ozone issues
persist in Brazil.
Anderson and Wilkes [7] analyzed the effects of using 10% ethanol
blended (E10) fuel use on vehicle emissions in a very large data set.
Beginning in 1995, the state of Colorado started an advanced
inspection and maintenance (I/M) program. This program required
that 1982 and newer light-duty vehicles undergo a centralized IM240
dynamometer emissions test. From November through February,
Colorado required that all gasoline sold in the area be blended with
an oxygenated compound (either ethanol or methyl tertiary butyl
ether [MTBE]), this was intended to reduce carbon monoxide (CO)
concentrations during winter periods. Over 80% of the fuel sold
during the winter periods of 1995 and 1996 was an E10 blend,
supplies of fuel ethanol were sufciently limited that only a small
fraction of the fuel sold in other months of the year were blended
with ethanol. The vehicle emissions data collected during the winter
months of a calendar year (January, February, November and December) were compared with the emissions data collected during the
months of April through September. Data collected in March and
October were not used, since fuel with variable ethanol content was
sold. Table 1 shows a summary of the effects of this program on
vehicle emissions, and because of the large number of vehicles tested,
each of the percent differences are statistically signicant. Hydrocarbon (HC) emissions were found to decrease by 1.3% and 0.2%
during the E10 fuel use periods of 1995 and 1996. Similarly, CO
emissions were found to be 7.0% and 4.9% lower, and NOx emissions
were found to be 13.5% and 13.1% higher during the E10 periods of
1995 and 1996. The emissions reductions for CO were lower than
expected, while the increases in NOx emissions were higher.
Jacobsen [8] used a nested global-urban air pollution/weather
forecast model, combined with high-resolution future emission
inventories, population data, and health effects data, to examine
the effect of converting from gasoline to E85, on cancer, mortality,
and hospitalization in the United States. The results suggested that
E85 (85% ethanol, 15% gasoline) may increase ozone-related
mortality, hospitalization, and asthma by about 9% in the Los
Angeles area and 4% on average in the United States. The model
predicted ozone increases in Los Angeles and the northeast US,
and decreases in the southeast; it also predicted increases in
peroxyacetyl nitrate (PAN) concentrations in the US.
Lpez-Aparicio and Hak [9] evaluated the effects of use of bioethanol (E9595% ethanol) fueled buses in Oslo, Norway. Aldehydes were
Table 1
Results of the analysis of I/M240 emissions data for 1995 and 1996 [7].

Number of vehicles tested

HC emissions
CO emissions
NOx emissions

1995

1996

721,942
 1.3%a
 7.0%
13.5%

651,223
 0.2%
 4.9%
13.1%

a
The table shows the % difference between the winter (E10) months and the
summer (non-oxygenated fuel) months.

164

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

measured in the ambient air, at locations exposed and not exposed to

the bioethanol buses. The acetaldehyde concentration tended to be
higher and the formaldehyde/acetaldehyde ratio was lower at the sites
exposed to the bioethanol buses. Ethanol and acetaldehyde were
measured in the exhaust plume during on-road operating conditions
for an E95 bus. Both ethanol and acetaldehyde concentrations in the
exhaust plume rose to concentrations of approximately 100 ppm
during idling periods, and acetaldehyde occasionally rose to 200 ppm
when stopped for a trafc signal.
Salvo and Geiger [10,11] analyzed temporally and spatially resolved
trafc data, meteorology, and pollutant concentrations in Sao Paulo,
Brazil (between 2009 and 2011) to identify changes in pollutant
concentrations during periods of substantial change in the fuel used
(ethanol or gasohol). The analysis showed that as ethanol fuel use
decreased, ozone (O3) concentrations decreased, and nitric oxide (NO)
and CO concentrations increased. It is likely that the decrease in ozone
concentration is due to the reaction of O3 with the increased emissions
of NO from the increased gasohol use. Madronich [11] suggests that the
production of secondary organic aerosols might be stimulated during
the periods of decreased use of ethanol and increased gasohol. This
analysis clearly shows changes in fuels used affect important aspects of
ambient air quality.
Nogueira et al. [12] have reported formaldehyde, acetaldehyde,
nitrogen oxides and ozone concentrations in Sao Paulo, Brazil,
from June 2012 through May 2013. They found formaldehyde and
acetaldehyde concentrations were near the low end, and the
formaldehyde/acetaldehyde ratio was near the high end of the
values reported over the previous thirty years, in spite of the
increased number of vehicles and fuel use in the area. They
suggest that the newer ex-fuel vehicles are meeting more
stringent emissions standards (including standards for aldehyde
emissions), allowing the ambient concentrations to decrease in
spite of increasing numbers of vehicles. Ozone in Sao Paulo
remains a serious problem, with the ozone air quality standard
being exceeded during eight of the twelve months studied.
There is still considerable uncertainty on the effects of ethanol
fuel use on vehicle emissions. The data suggests that ethanol and
acetaldehyde emissions will increase, formaldehyde emissions will
increase slightly, and benzene and 1,3-butadiene emissions will
decrease. The effects on the emissions of NOx and PM is likely to be
fairly small, but is uncertain. It is widely thought that urban ozone
will increase in many areas. With improved vehicle emissions
control technology, it is expected that the effects of ethanol fuel
use on vehicle emissions will decrease. In conclusion, increased
use of ethanol blended fuels is unlikely to improve air quality over
future gasoline vehicles. Little data is available that allows assessment of bioethanol blends on compliance with newer, more
stringent emissions standards such as US EPA Tier 3 or Euro VI.

3. Biodiesel
Recent reviews have shown that increasing percentages of
biodiesel blended with petroleum diesel generally decrease CO, HC
and PM emissions, while NOx emissions generally increase [6,13].
These studies have also identied many other factors that affect the
change in emissions: LD and HD vehicles, engine size, the feedstock
from which the biodiesel was made, the model year of the engine,
the type of emissions controls used, and the transient cycle used in
the emissions test. For the two more environmentally important
regulated emissions - PM and NOx, there seems to be a trade-off
shown by the test data, where a larger increase in NOx will occur
with a larger decrease in PM emissions.
Limited data is available to assess the effects of biodiesel fuel
blends on the emissions of unregulated pollutants. This data
suggests the emissions of polycyclic aromatic (PAH) compounds

may increase with higher levels of biodiesel blends [6]. The data
consistently shows an increase in formaldehyde (HCHO) and
acetaldehyde (CH3CHO) emissions, and more limited data suggest
that acrolein and other aldehydes also increase with the use of
biodiesel blends [6,14]. There is insufcient data available to assess
the effects of biodiesel fuel use on the emissions of other
hazardous air pollutants or on ultrane particulate emissions,
especially particle number and size distributions.
It is likely that by properly adjusting the diesel engine, and
using the appropriate emissions controls, the adverse impacts of
biodiesel blended fuels could be largely eliminated. These adjustments and controls are very likely to depend strongly on the
percentage of biodiesel used, and on the feedstock for the
production of the biodiesel. Little data is available that allows
assessment of biodiesel blends on compliance with newer, more
stringent emissions standards such as US EPA 2010 or Euro VI.

4. Biomethane (natural gas)

Biomethane is methane derived from biogas, an entirely renewable low-carbon alternative fuel that can be produced locally from
organic waste. Chemically the same as conventional natural gas,
biomethane is produced from the anaerobic digestion of organic
matter such as animal manure, sewage, and municipal solid waste.
After it is processed to required standards of purity, biomethane
becomes a renewable substitute for natural gas and, once compressed
or liqueed, can be used to fuel natural gas vehicles. Both compressed
natural gas (CNG) and liqueed natural gas (LNG) can be used as a
gasoline or diesel substitute in an array of vehicles: light-duty
vehicles, transit buses, solid waste trucks, heavy-duty trucks and off
road heavy-duty equipment. Use of biomethane as a vehicle fuel is
expected to reduce greenhouse gas emissions substantially, compared
to conventional transportation fuels. It can be an important domestic
resource, reducing dependence on foreign oil.
This review will explore the potential effects of biomethane as
a transportation fuel on air pollutant emissions. Emissions data for
CNG fueled vehicles, as well as data for comparable conventionally
fueled (diesel or gasoline) vehicles will be included. This review
will not include exhaustive data for conventionally fueled vehicles.
Results will only be presented when a minimum of ten sets of
emissions data are available in each category being compared.
More extensive data is available for CNG fueled transit buses
which will be discussed rst.
4.1. Buses
Hesterberg et al. [15] have reviewed emissions data from
vehicles fueled with CNG. Much of the data discussed compares
emissions of CNG and diesel fueled transit buses. The results of
that review will be highlighted along with updated information.
In transit buses, diesel fueled vehicles had lower carbon monoxide
(CO) and hydrocarbon (HC) emissions than their CNG fueled counterparts [15]. Non-methane hydrocarbon (NMHC) and methane (CH4)
emissions from CNG fueled buses were also higher than those from
the diesel buses. Exhaust after-treatment lowered emissions for both
diesel and CNG fueled buses. Due to the limited emissions test data
and its variability, few of these differences were statistically signicant
(Po0.05). CO and HC emissions from CNG fueled transit buses were
signicantly higher than the same emissions from untreated diesel
buses, and NMHC emitted from the CNG buses was reduced signicantly by treatment with a three-way catalyst (TWC). Diesel fueled
transit buses had the highest particulate matter (PM) emissions. CNG
buses with and without exhaust after treatment, and diesel buses
with traps had low emissions that were not signicantly different.
Transit bus diesels with and without after-treatment and CNG

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

vehicles without after-treatment had similar, relatively high nitrogen

oxides (NOx) emissions. These emissions are high compared to the US
emissions standards for NOx, which were fully phased in by 2010, and
the Euro VI emissions standard for NOx, which took effect in 2013.
Table 2 shows a summary of the emissions data that is available
for transit buses that use diesel or CNG fuel [1641]. The diesel bus
category also separates diesel-hybrid buses from non-hybrid
buses. Data in this table includes emissions data for buses tested
with and without exhaust after-treatment and for all vehicle
driving cycles. The effects of exhaust after-treatment and vehicle
driving cycles on emissions will be discussed in Sections 4.1.1
and 4.1.2.
The largest quantity of emissions data is available for regulated
pollutants and related compounds (HC, NOx, CO, PM, and carbon
dioxide [CO2]). Total hydrocarbon emissions (Table 2) are signicantly lower for hybrid diesel transit buses than for non-hybrid
buses, and are signicantly higher for CNG fueled buses than for
diesel hybrid and non-hybrid buses. Based on the data for the CNG
fueled buses, about 90% of the HC emissions are CH4; the CH4
emissions for diesel buses are near zero. NMHC emissions from
diesel buses are low and the emissions from CNG fueled buses are
also low, but signicantly higher than for both classes of the diesel
buses. NOx emissions for non-hybrid diesel buses and CNG fueled
buses are not signicantly different, while the emissions from the
hybrid diesel buses are signicantly lower than from the nonhybrid diesel and CNG buses. CO emissions from hybrid diesel
buses are signicantly lower than non-hybrid diesel buses, however CO emissions from CNG fueled buses are signicantly higher
than both non-hybrid and hybrid diesel buses. The data show that
CO2 emissions from the CNG fueled buses are signicantly lower
that the CO2 emissions only from non-hybrid diesel buses. The PM
emissions from hybrid diesel buses are signicantly lower than
those of non-hybrid diesel buses, and the PM emissions from the
CNG fueled buses are signicantly lower than those from both the
hybrid and non-hybrid diesel buses. This data shows that in
general diesel hybrid buses have lower emissions of these regulated pollutants than either diesel or CNG fueled transit buses, and
only CNG fueled buses emit signicantly lower PM than the hybrid
diesel buses. Combining CNG and hybrid technologies may lead to
even lower emissions.
A smaller quantity of data exists for unregulated and toxic
emissions from buses, however the effects of CNG use will also be
investigated. Hesterberg et al. [15] reported that nitrogen dioxide
(NO2) emissions from diesel buses with particulate traps had the
highest emissions and NO2/NOx ratios. Continuously regenerating
diesel particulate lter systems have been linked to increased emissions of NO2 from the tailpipe. NO2 is produced on the catalyst and is
the key component used to burn the soot collected in the lter, thus
leading to an increase in tailpipe NO2 over engine-out emissions. In
Europe, an increasing share of diesel passenger cars combined with

the introduction of exhaust after-treatment on diesel vehicles, has led

to a situation in which total NOx emissions have been reduced.
However, the NO2 levels in urban environments have not been
reduced. This suggests that the direct emission of NO2 has become
a problem [36]. Dieselnet.com [42] has indicated that a maximum
limit for the NO2 component of NOx emissions may be dened at a
later stage as a part of the Euro VI emission standard.
Based on the data analyzed in this study, the particle number
emissions for CNG fueled buses was not signicantly different
from the particle number emissions for non-hybrid diesel buses.
This along with previous studies have shown that PM mass
emissions from CNG fueled buses are generally lower than those
from diesel buses that are not equipped with particulate traps
[15,25]. Jayarante et al. [43] presented results of on-road and
dynamometer studies which show that particle number (PN)
emissions from CNG buses are signicantly lower than from diesel
buses at low engine power; they become comparable at high
power. PN emissions were not signicantly different between
acceleration and operation at steady maximum power for diesel
buses. However, the corresponding PN emissions from CNG buses
when accelerating can be an order of magnitude greater than
when operating at steady maximum power. During acceleration
under heavy load, PN emissions from CNG buses are an order of
magnitude higher than from diesel buses. Hallquist et al. [44]
reported particle number emissions for buses operating in an
accelerating mode of 2.5 71.0  1014 particles/km for diesel fueled
buses with diesel particulate lters, 8.6 74.8  1014 particles/km
for diesel fueled buses without diesel particulate lters, and
407 29  1014 particles/km for CNG fueled buses without particulate lters; these studies are based on on-road measurements. The
particle number data analyzed in this study are based on driving
cycles that are intended to be representative of real world bus
operations, including accelerating, decelerating and steady-state
operating modes. The impact of CNG fuel use on PN emissions
under real-world operating conditions is still not clear.
The data analyzed here show that the elemental carbon (EC)
emissions (0.83 70.14 mg/km) from CNG fueled transit buses are
signicantly lower than the organic carbon (OC) emissions
(7.46 72.02 mg/km). There was insufcient data for diesel fueled
buses to allow a comparison with the EC and OC emissions. A
previous study of diesel bus emissions equipped with a diesel
oxidation catalyst has shown that the total PM emissions were
about four times those from a CNG fueled bus without exhaust
after-treatment [45]. This previous study also showed that the EC
emissions from the diesel were about three times the OC emissions for the diesel bus. The OC emissions from the CNG bus were
about half those from the diesel bus and about six times the EC
emissions from the CNG bus.
Sufcient data was available to begin assessing the effects of
CNG use on the emissions of toxic air pollutants from transit buses.

Table 2
Comparison between diesel (non-hybrid) buses, diesel (hybrid) buses, and CNG buses. This comparison is
for all exhaust after treatment and driving cycles.
Emission
HC
NMHC
CH4
NOx
CO
CO2
PM

1-Diesel non-hybrid
0.317 0.04a (97) o 3
0.177 0.05 (45) o 3
0.007 0.00 (37) o 3
14.30 7 1.01 (130)
1.98 7 0.41 (114) o 3
14177 87 (58)
102.247 11.91 (131)

165

2-Diesel hybrid

3-CNG

0.11 70.05 (29) o 1, o 3

0.06 70.03 (24) o 3
0.00 70.00 (19) o 3
9.2471.23 (42) o1, o 3
0.73 70.15 (37) o 1, o 3
1302 7132 (29)
29.83 75.62 (38)o 1

11.87 7 0.98 (165)

0.80 7 0.12 (87)
8.177 1.07 (90)
12.39 7 0.89 (178)
4.127 0.53 (156)
1236 7 60 (98) o 1
15.687 1.51 (162) o 1, o 2

a
Data presented are mean 7standard deviation for the emission in g/km (except PM which is in
mg/km). Next is the (number of measurements) and o of data column when the measured value is
signicantly less than the column value at Po 0.05.

166

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

The emissions of benzene from CNG fueled buses was

1.50 70.30 mg/km (n 28), and those from diesel fueled buses
was 1.00 70.29 mg/km (n 17), which is not signicantly different. Emissions of 1,3-butadiene from CNG fueled buses was
1.217 0.42 mg/km (n 16), while the quantity of data was insufcient to assess differences for diesel fueled buses. The emissions of
formaldehyde from CNG fueled buses was 195 765 mg/km
(n 44), which was signicantly greater than those from diesel
fueled buses, 66.1 721.1 mg/km (n 18). Emissions of acetaldehyde from CNG fueled buses was 7.78 72.28 mg/km (n 45), and
was signicantly less than the emissions from diesel fueled buses
of 22.6777.33 mg/km (n 18).
The emissions of total polycyclic aromatic hydrocarbons from CNG
fueled transit buses averaged 135770 mg/km (n37), which was not
signicantly different from those of diesel fueled buses, 69.9735.3 mg/
km (n18). Sufcient data was available to allow the assessment of
CNG fuel use on mutagenicity of the emissions from transit buses. For
diesel fueled buses the mutagenicity for T98S9 was found to be
133772 revertants/km (n11); for CNG fueled bus emissions it was
16.973.4 revertants/km (n21), which is not signicantly different.
For T98 S9 the mutagenicity of the diesel bus emissions was found to
be 3207124 revertants/km (n13), while that for CNG fueled buses
was 101738 revertants/km (n18), which was, again, not signicantly
different.
4.1.1. Exhaust after-treatment
Sufcient data is available to only begin assessing the effects of
exhaust after-treatment techniques on diesel emissions from transit
buses for the regulated pollutants and CO2 (Table 3). The data for
the diesel fueled transit buses was divided into the following
groups: no exhaust treatment, oxidation catalyst (OxC), selective
catalytic reduction (SCR), diesel particulate lter (DPF), and combinations of exhaust gas recirculation (EGR) with DPF, and EGR with
SCR and DPF. HC emissions which are low in untreated diesel bus
exhaust are not changed signicantly by the addition of an OxC, but
are reduced signicantly by the addition of DPF. DPFs often include
a catalyst, which can oxidize some of the particulate matter and
some of the gaseous HC emissions. Not enough data exists in this
data set for evaluation of the effects of SCR or the EGR combinations
of treatment on HC emissions. NOx emissions were signicantly
reduced by the addition of an oxidation catalyst and the addition of
the SCR. SCR is a technique that is specically designed to reduce
NOx emissions by reduction of NOx to N2. NOx emissions were
lowest for SCR, signicantly lower than with no exhaust treatment
and with DPF, but not signicantly different from treatment with
the OxC or EGR combinations. NOx emissions were signicantly
higher with the DPF than all of the other exhaust after-treatment
options; the high temperatures and oxidizing catalysts in the DPF
lead to the increase NOx emissions. Insufcient data was available
for the EGRSCRDPF combination of treatments. This combination is expected to lead to lower NOx than the EGRDPF, which was
combined with this data (Table 3).

For CO emissions from diesel transit buses with no exhaust

after-treatment, OxC and SCR were not signicantly different. CO
emissions using the DPF and the EGR combinations (Table 3) were
signicantly lower than CO emissions without exhaust aftertreatment and with OxC. CO2 emissions with OxC were signicantly higher than no after-treatment, and CO2 emissions with SCR
were not signicantly different from no after-treatment or OxC.
Insufcient data was available for CO2 emissions with DPF and the
EGR combinations, to assess the effects of these after-treatments.
PM emissions with OxC after-treatment were not signicantly
different than no after-treatment. With SCR treatment, PM emissions were lower than no after-treatment and OxC after-treatment.
PM emissions with DPF were signicantly lower than no aftertreatment, OxC and SCR after-treatment; PM emissions with the
EGR combination of after-treatments were signicantly lower than
all four of the other subdivisions of treatment.
Sufcient data is available to only begin assessing the effects of
exhaust after-treatment techniques on emissions from CNG transit
buses for the regulated and related pollutants and CO2 (Table 4). The
data for the CNG fueled transit buses was divided into groups, with
different combustion characteristic and exhaust after-treatment: lean
burn, lean burn with oxidation catalyst (LeanOxC), stoichiometric
combustion with three-way catalystwithout or with exhaust gas
recirculation (StoichTWC or Stoich EGRTWC), and lean mix
combustion with three-way catalyst (Lean MixTWC). The last two
technologies are being developed to meet the newer, more stringent
emissions standards (US EPA 2010 and Euro VI) for transit vehicles.
HC emission is higher for CNG fueled than for diesel fueled buses. This
data set shows that 90712% (n90) of the HC emissions are CH4.
The lean burn combustion emits the highest HC, and lean burn with
an oxidation catalyst is only slightly lower. The oxidation catalyst does
not typically operate at a sufciently high temperature to oxidize the
CH4 efciently. Lean burn technology has been used extensively with
CNG fueled transit buses, because it allowed operation at higher
efciency and allowed NOx emissions to be reduced from stoichiometric combustion conditions. Stoichiometric combustion occurs at a
higher temperature, burning hydrocarbons (this case CH4) more
efciently; therefore HC emissions are signicantly lower than both
of the lean burn combustion categories. With the addition of the
three-way catalyst, the emissions of both CH4 and NOx can be
reduced. Three-way catalysts cannot be used directly with lean burn
combustion, since the oxygen content of the exhaust is too high to
allow the catalyst to work properly. A limited amount of data is
available for the lean-mix system, which combines the use of lean
burn at high torque and/or speed with stoichiometric combustion at
moderate load and speed. This combustion system can be used with a
TWC to further control emissions [25,39].
NMHC emissions (Table 4) are highest for the lean burn CNG
transit buses. These emissions are reduced signicantly with the
use of an OxC, and the emissions for the stoichiometric system
with TWC are signicantly less than both of the lean burn systems.
CH4 emissions are highest for the lean burn systems, and are not

Table 3
Comparison between non-hybrid diesel bus emissions with different exhaust after-treatment. This comparison is for all driving cycles.
Emission

1-Nonea

2-OxC

3-SCR

HC
NOx
CO
CO2
PM

0.46 70.08b(25)
15.85 7 1.58 (33) o 4
2.14 7 0.37(25)
1149v154(17) o 2
206.387 29.83(33)

0.45 70.07(28)
10.58 71.69(28) o 1, o 4
2.3570.33(28)
1684 7261(12)
154.25 721.97(32)

7.747 1.64(12) o 1, o 4
6.127 3.41(12)
1381 7 186(12)
76.25 7 28.29(12) o1, o 2

4-DPF

5-EGR DPF or EGR SCR DPF

0.09 7 0.02(16) o 1, o 2
21.09 7 2.36(24)
0.277 0.07(21) o 1, o 2

11.157 3.09 (15) o 4

0.277 0.08 (15) o 1, o 2

17.50 7 3.25(26) o1, o 2, o 3

8.747 2.41 (15) o 1, o 2, o3, o 4

a
Columns identify the exhaust after-treatment used 1-noneno after-treatment, 2-OxCoxidation catalyst, 3-SCRselective catalytic reduction, 4-DPFdiesel
particulate lter, and 5-EGR DPF or EGR SCR DPFexhaust gas recirculation and diesel particulate lter without or with selective catalytic reduction.
b
Data presented are mean 7standard deviation for the emission in g/km (except PM which is in mg/km). Next is the (number of measurements) and o of data column
when the measured value is signicantly less than the column value at P o0.05. If a cell contains no data fewer than 10 measurements were available.

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

167

Table 4
Comparison between CNG fueled buses with different combustion characteristics and exhaust after-treatment. This comparison is for all driving cycles.
Emission 6-Lean Burna
HC
NMHC
CH4
NOx
CO
CO2
PM

15.96 7 2.54b (30) 41,42, 4 4

1.40 7 0.32 (19) 41, 4 2, 44
14.96 7 3.34 (17)
15.96 7 2.23(35)4 2, 43, o4

7-Lean OxC
14.25 7 1.23(1 0 1) 4 1,42, 44
0.85 7 0.14(50) o 641,4 2, 44
9.067 1.27(51)
14.65 7 1.15(10 5) 42, 43, o 4

8-Stoich TWC or Stoich EGR TWC 9-Lean Mix TWC

2.30 70.45(28) o 6, o7 41,42, 4 4

0.13 70.08(13)o 6, o 74 1,42
2.48 70.62(17) o 6, o7
4.19 70.81(28) o 6, o 7o 1,o 2,
o 3, o 4, o 5
9.36 7 1.57(32)41,42, 43, 44, 4 5 1.60 7 0.41(90) o6, o 84 4, 45
6.74 71.35(28)41,4 2, 44, 45
13197 95(53)
1218 787(26)
21.727 3.19(37) o 1,o 2, o 3, 45
14.85 7 1.88(92) o 6o 1,o 2, o 3, 45 13.80 75.35(22) o1,o2, o 3

1.687 0.50(12) o 6, o 7,o 8o 1,o 2,

o3, o4, o 5
9777 136(12) o 7o 2, o 5
5.17 71.62(12) o 6, o 7o 1,o 2, o 3, o 4

a
Columns identify the combustion characteristics and exhaust after-treatment used 6-Lean Burnlean burn with no after-treatment, 7-Lean OxClean burn with
oxidation catalyst, 8-Stoich TWC or Stoich EGR TWCstoichiometric combustion with three-way catalyst without or with exhaust gas recirculation, 9-Lean Mix TWC
lean mix combustion with three-way catalyst.
b
Data presented are mean 7standard deviation for the emission in g/km (except PM which is in mg/km). Next is the (number of measurements) and o of data column
when the measured value is signicantly less than the column value at P o0.05. If a cell contains no data fewer than 10 measurements were available. The comparison also
identies the signicant differences between CNG and the diesel exhaust after-treatment techniques, o of data column when the measured value is signicantly less than
the column value and 4 of data column when the measured value is signicantly greater than the column value at Po 0.05.

reduced signicantly by the addition of an OxC, however the

stoichiometric combustion with TWC leads to signicantly lower
CH4 emissions. Lean burn without OxC has the highest NOx
emissions, but these are not reduced signicantly by addition of
an OxC. The stoichiometric system with TWC has signicantly
lower NOx emissions than both lean burn congurations, and the
lean mix system with TWC has signicantly lower NOx emissions
that all three of the other congurations.
CO emissions are highest with the lean burn buses, and slightly
but not signicantly lower using the stoichiometric system with
TWC. The lean burn system with OxC has signicantly lower CO
emissions than the other two congurations. CO2 emissions
decrease slightly, but not signicantly, when going from lean burn
with OC, to stoichiometric with TWC. The CO2 emissions for lean
mix with TWC is signicantly lower than for lean burn OxC.
Insufcient CO2 emissions data for lean burn buses with no
exhaust after-treatment was available for assessment. PM emissions were highest with the lean burn buses, and were signicantly lower for the lean burn buses with OxC. The lean mix buses
with TWC had the lowest emissions of PM, and these were
signicantly less than both of the lean burn congurations. The
PM emissions for the stoichiometric buses with TWC were
sufciently variable to not be statistically different than the other
three types of CNG buses.
Limited data was available for other pollutant emissions from
CNG buses with different exhaust after-treatment. Formaldehyde
emissions from lean burn buses averaged 546 7188 mg/km
(n 13) and signicantly lower 47.77 8.0 mg/km (n 27) from
lean burn bus with OxC. Acetaldehyde emissions from lean
burn OxC buses averaged 2.50 70.64 mg/km (n 24) and significantly lower 0.30 70.09 mg/km (n 10) from CNG buses with
stoichiometric combustion and TWC. It is expected that formaldehyde emissions would have been lowered further for stoichiometric combustion and TWC. Sufcient emissions data for total
PAHs existed only for lean burn and lean burn OxC CNG buses,
however, the variability in the data was sufciently high that no
signicant difference was found.
In a recent study, Yoon et al. [41] reported substantial reductions in emissions of gaseous unregulated pollutants (such as
carbonyl compounds, benzene and 1,3-butadiene) for CNG fueled
transit buses with stoichiometric combustion engines and TWC,
compared to lean-burn engines with or without an oxidation
catalyst (OxC). Similarly, OC, EC and PAH emissions were reduced
substantially with stoichiometric combustion and TWC. Potential
mutagenicity measured using a microsuspension modication of
the Salmonella/microsome assay was also lower by more than 99%

for buses with stoichiometric engines and TWC, compared to

buses with lean-burn engines and OxC.
The data in Tables 3 and 4 can be used to compare the emissions
from diesel and CNG fueled transit buses with different exhaust aftertreatment. For HC emissions, the emissions are signicantly higher for
the CNG buses with any exhaust after-treatment than for any diesel
exhaust after-treatment combination. Most of the HC emissions from
CNG buses are CH4 emissions. However, comparing NMHC emissions
for the CNG buses with HC emissions for the diesel buses, the same
situation holds. The exception is CNG with stoichiometric combustion
and TWC, which is not signicantly different than the diesel buses
with DPF. NOx emissions for CNG buses with stoichiometric combustion and TWC, and for lean mix combustion and TWC, are signicantly lower than from diesel buses with any exhaust after-treatment.
NOx emissions for CNG buses with lean-burn or lean-burn with OC
emit signicantly more NOx than diesel buses with OC or SCR; these
same lean-burn CNG buses emit signicantly less NOx than diesel
buses with DPF.
Lean-burn CNG buses emit signicantly more CO than each group
of diesel buses. Lean-burnOC CNG buses emit signicantly more CO
than dieselEGR combinations, and dieselDPF. CNG buses with
stoichiometric combustion and TWC emit signicantly more CO than
each group of diesel buses, except dieselSCR. CO2 emissions from
the CNG buses are not signicantly different than each group of diesel
buses, except lean mix buses with TWC, which emit signicantly less
CO2 than both dieselOC and dieselSCR. All CNG bus emission
after-treatment groups emit signicantly less PM than diesel with no
after-treatment, dieselOC and diesel SCR. Both CNG lean-burn and
lean-burn OC emit signicantly more PM than diesel EGR combinations. Only CNG buses with lean mixTWC emit less PM than
dieselDPF.

4.1.2. Bus test cycles

Several different test cycles were used during the collection of
the emission data. Five different test cycles had a minimum of 10
sets of emissions data for both non-hybrid diesel and CNG fueled
buses. The test cycles meeting the data requirements were the
Central Business District (CBD) cycle, the Orange County Transport
Authority (OCC) cycle, the Braunschweig (Braun) cycle, the New
York bus (NYBus) cycle, and the ADEME-RATP Paris bus (Paris)
cycle. Some of these chassis dynamometer driving cycles (NYBus
and Paris) are more demanding than others (OCC and Braun).
Typically, the emissions from vehicles operated on the more
demanding cycles are higher. Since the data shown in previous
Tables 24 are combined data for all of the driving cycles, the

THC
NOx
CO
PM

a
Data presented are mean 7 standard deviation for the emission in g/km (except PM which is in mg/km). Next is the (number of measurements) and in the CNG column the CNG value is indicated as being signicantly greater
than diesel ( 4D) or signicantly less than diesel ( o D) at Po 0.05.

16.197 3.06(18) 4 D
12.707 1.46(18) o D
5.127 1.88(17)
23.09 7 3.44(17) o D
0.64 7 0.13(13)
18.50 7 2.24(14)
2.87 7 0.77(14)
203.6 747.3(14)
39.23 7 3.18(16) 4D
28.737 4.39(19)
13.95 7 3.12(16)
51.187 6.47(16) o D
0.49 7 0.18(10)
33.78 7 2.63(18)
10.83 7 5.30(10)
199.6 7 45.7(13)
6.04 7 1.89(17) 4D
7.05 71.12(17)
0.83 70.23(16)
19.07 75.53(17) o D
0.12 7 0.04(14)
7.017 0.70(15)
0.79 7 0.25(15)
70.377 16.64(15)
7.32 7 1.52(18) 4 D
7.08 7 1.00(18)
2.08 7 0.65(17)
7.30 7 1.53(16) oD
0.187 0.05(11)
7.577 1.37(12)
0.89 7 0.30(12)
91.317 25.58(12)
8.96 7 0.45(50) 4D
15.28 7 1.52(52)
3.747 0.52(45) 4D
9.577 1.26(44) o D
0.33 70.09 (24)
17.26 7 0.60(27)
2.05 7 0.45(27)
106.47 20.0(29)

Diesel

CNG

Diesel

CNG

Diesel

CNG

Diesel

CNG

Paris
NYBus
Braunschweig

Diesel

The data for diesel and CNG truck emissions [17,39,4952] is much
more limited than data available for transit buses. Table 6 summarizes
the emissions of regulated pollutants for diesel and CNG fueled
medium- and heavy-duty trucks. This data includes data collected
with different exhaust after-treatment and for several different
emissions test cycles. HC emissions are signicantly lower for diesel
fueled trucks than for CNG fueled trucks; emissions of NHMC from
CNG trucks are not signicantly different from the HC emissions from
diesel trucks. The fraction of the total HC emissions that was CH4 for
CNG fueled trucks was 0.6770.20, which is lower than for CNG

OCC

4.2. Diesel and CNG trucks

CBD

4.1.3. Meeting current and future emissions standards for transit

buses
Current data suggests there are options for powering transit
buses into the future. Diesel hybrid buses have the potential to use
fuel more efciently that conventional buses, due to the hybrid
technology, so their emissions without added exhaust aftertreatment are lower for many pollutants from conventional buses.
Advance exhaust after-treatment technologies have not been
applied as widely to diesel hybrids, as they have been to either
diesel or CNG buses. The emissions from diesel hybrid buses can
be further reduced by the addition of these advanced exhaust
after-treatment techniques.
Diesel engines using combinations of EGR, SCR and DPF have
been certied as meeting the US EPA 2010 and Euro VI emissions
standards. CNG engines operated in a stoichiometric mode with
EGR and TWC readily meet US EPA 2010 standards. The Euro VI
standard has more rigorous emissions limits than the US 2010
standard for HC and CO, which requires greater care in the
implementation of these technologies. The lean mix combustion
with TWC is expected to be able to comply with lower emissions
limits for HC and CO more easily than stoichiometric combustion
with TWC.
Applying natural gas hybrid technologies to transit buses has
the potential to increase operating efciency and further reduce
vehicle emissions. Another important consideration for future
transit buses is greenhouse gas emissions. Biomethane has the
potential to be substituted for natural gas, greatly reducing fossil
fuel related greenhouse gas emissions [4648].

Emission

differences in driving cycles can account for some of the variability

in the data.
Table 5 shows data for the emissions of the regulated pollutants
from non-hybrid diesel and CNG buses for each of the ve different
driving cycles; this data includes all exhaust after-treatment technologies. It is clear that the NYBus and Paris cycles generally had the
highest emissions for a specic pollutant and fuel, while the OCC and
Braun cycles had the lowest emissions. For all ve driving cycles, the
HC emissions for the CNG fueled buses were higher than for diesel
fueled vehicles. NOx emissions were not signicantly different for diesel
and CNG fueled buses for all of the test cycles, with the exception of the
Paris cycle, while the CNG fueled emissions of NOx were signicantly
lower. CO emissions were not signicantly different for diesel and CNG
fueled buses for all of the test cycles, with the exception of the CBD
cycle, where the CNG fueled emissions of CO were signicantly greater.
PM emissions with CNG fuels were signicantly lower than those for
diesel fuels for each of the test cycles.
Ideally, it would be benecial to compare emissions for the
different exhaust after-treatment technologies for individual driving cycles, to allow a better determination of the effectiveness of
the exhaust treatment technologies. Sufcient data does not exist
for the different technologies using a single test cycle, to allow this
type of assessment.

CNG

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

Table 5
Comparison between non-hybrid diesel and CNG fueled buses with different driving cycles. This comparison is for all exhaust after-treatment techniques.

168

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

169

Table 6
Comparison of emissions for diesel trucks, and CNG trucks. This comparison is for all exhaust after treatment and driving cycles.
Emission

HC
NOx
CO
CO2
PM

Medium- and heavy-duty

Refuse trucks

Diesel

CNG

Diesel

CNG

0.20 7 0.07a(36)
5.917 0.52(37)
0.65 7 0.14(37)
8247 44(20)
82.87 7 14.32(36)

4.84 7 0.41(26)4 D
3.767 0.39(26) o D
0.357 0.15(23)
690 7 43(22) oD
12.92 7 2.83(27) o D

1.447 0.43(14)
37.36 7 6.25(15)
3.017 0.55(15)
3099 7 354(12)
568.27 163.9(15)

12.0771.24(53)4D
34.53 73.76(53)
6.13 71.26(53) 4D
1994 7220(33) o D
83.1 722.5(30) o D

a
Data presented are mean 7 standard deviation for the emission in g/km (except PM which is in mg/km). Next is the (number of measurements) and in the CNG column
the CNG value is indicated as being signicantly greater than diesel ( 4D) or signicantly less than diesel (o D) at Po 0.05.

transit buses. NOx, CO2 and PM emissions for the diesel fueled trucks
was signicantly higher than for the CNG fueled trucks, and CO
emissions were not signicantly different. In general, the emissions
for medium- and heavy-duty trucks shown in Table 6 are lower than
for the transit buses shown in Table 2. The NOx emissions for the CNG
trucks was signicantly lower than diesel, while the NOx emissions
were not signicantly different for CNG and diesel transit buses. The
CO emissions for the CNG and diesel trucks were not signicantly
different, while CO emissions for the CNG buses were signicantly
higher than diesel.
Table 6 also summarizes the emissions of regulated pollutants
for diesel and CNG fueled refuse trucks [5356]. These trucks are
often tested using a test cycle characteristic of refuse truck
operations, which include a large number of start and stop
operations, as well as idling and trash compaction operations.
The distance related emissions from these test cycles are generally
higher than for test cycles used for other trucks and buses. Table 6
includes data collected with different exhaust after-treatment and
for several different emissions test cycles. HC emissions are
signicantly lower for diesel fueled trucks than for CNG fueled
trucks. The emissions of NHMC from CNG trucks is not signicantly different from the HC emissions from diesel trucks. For HC
emissions, the fraction that was CH4 for these CNG fueled vehicles
was 0.86 70.05. For the refuse trucks, the NOx emissions were not
signicantly different for the diesel and CNG fueled trucks. The CO
emissions for CNG trucks were signicantly higher than for diesel
trucks; again, both CO2 and PM emissions for diesel trucks was
signicantly higher than for CNG trucks. The pollutants that are
signicantly different for diesel and CNG refuse trucks in Table 6,
are the same as for diesel and CNG transit buses in Table 2.
4.3. Light-duty CNG vehicles
Table 7 summarizes the emissions data for the regulated and
related pollutants for gasoline and CNG fueled light-duty vehicles
[5767]. Light-duty CNG vehicles will be compared with gasoline
fueled vehicles, rather than diesel vehicles, as was done by
Hesterberg et al. [15]. This includes data collected with different
exhaust after-treatment, some data from bi-fueled vehicles (that
can operate on either gasoline or CNG) and for different emissions
test cycles. The HC emissions are signicantly lower for gasoline
fueled light-duty than for CNG fueled vehicles. The emissions of
NHMC from CNG vehicles is signicantly lower than from gasoline
fueled vehicles. For HC emissions, the fraction that was CH4 for
these CNG fueled light-duty vehicles was 0.59 70.02. The NOx and
PM emissions from CNG fueled light-duty vehicles was not
signicantly different than from gasoline vehicles. However, both
CO and CO2 emissions from CNG vehicles were lower than from
the gasoline vehicles.
Limited additional data exists for some unregulated pollutant
emissions from gasoline and CNG fueled light-duty vehicles. Durbin
et al. [59] have reported data for organic carbon (OC) and elemental

Table 7
Comparison of emissions for light-duty gasoline and CNG fueled vehicles. This
comparison is for all exhaust after treatment and driving cycles.
Emission
HC
NMHC
NOx
CO
CO2
PM

Gasoline

CNG
a

0.217 0.03 (46)

0.187 0.03(47)
0.38 7 0.05(50)
3.577 0.97(50)
293.5 7 8.9(35)
3.067 1.18(13)

0.447 0.05(70)4G
0.05 70.01(72)oG
0.357 0.05(80)
1.29 70.19(80) o G
206.7 77.5(42) o G
4.15 71.90(13)

a
Data presented are mean7 standard deviation for the emission in g/km
(except PM which is in mg/km). Next is the (number of measurements) and in the
CNG column the CNG value is indicated as being signicantly greater than gasoline
( 4G) or signicantly less than gasoline (o G) at Po 0.05.

carbon (EC) emissions. The EC emissions for CNG fueled vehicles were
lower than those for gasoline fueled vehicles, however, the OC
emissions for CNG vehicles were higher than for gasoline vehicles.
The same paper also reported total polycyclic aromatic hydrocarbon
(PAH) measurements that were much lower for the CNG vehicles
than for the gasoline vehicles. Correa and Arbilla [68] reported
formaldehyde and acetaldehyde emissions from 20 bi-fueled vehicles
tested in Brazil, which compared emissions of the aldehydes for CNG
fueled vehicles with those from gasohol fueled vehicles (24% ethanol
with gasoline); the formaldehyde emissions for CNG fuel were almost
4 times those from gasohol. Gasohol formaldehyde emissions are
typically greater than emissions from gasoline fueled vehicles. Acetaldehyde emissions were about 25% of those for gasohol. Since most
of the formaldehyde and acetaldehyde emissions data from light duty
vehicles was from the Brazilian bi-fueled vehicle study, the results are
not likely to be representative of the effects of CNG fuel on light-duty
vehicles with gasoline as the reference fuel. Myung et al. [69] reported
data for particle number (PN) emissions from a bi-fueled vehicle,
using either gasoline or CNG. The PN emissions were lower for CNG
than gasoline for one driving cycle, but were higher for two other
driving cycles.
4.4. Natural gas composition
There are three recent papers that explore the effects of natural
gas composition on vehicle emissions. Hajbabaei et al. [37]
investigated these effects for two CNG buses equipped with lean
burn combustion and OxC, and one stoichiometric CNG bus
equipped with EGR and TWC. For the lean burn buses, CNG with
more of the heavier hydrocarbons exhibited higher NOx and
NMHC emissions, but lower emissions of HC, CH4, and formaldehyde. The newest technology bus with the stoichiometric combustion engine and TWC did not show any specic fuel composition
effects, it appears that newer technology, heavy-duty natural gas
engines can run on a wider range of natural gas fuels of varying
composition, without impacting emissions. Karavalakis et al. [56]
conducted a similar study on one CNG fueled refuse hauler with a

170

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

lean burn engine and OxC. NOx emissions tended to be higher for
natural gas mixtures with more of the heavier hydrocarbons
during all three segments of the test cycle. HC, CH4, CO, PM, PN,
formaldehyde and acetaldehyde all showed lower emissions for
blends with more of the heavier hydrocarbons. Karavalakis et al.
[64] tested the emissions for two different cars, each with four
different CNG fuel compositions. The trends were not consistent
between the vehicles, although more of the heavier hydrocarbons
in the blends did tend to lead to higher NHMC emissions for both
cars. Further studies would be needed to better understand the
impact of these factors on the trends in the HC, NOx, and CO
emissions for the different vehicles.

5. Conclusions
There is still considerable uncertainty on the effects of bioethanol fuel use on vehicle emissions. This is partially due to the
changes in vehicle engine and emissions control technologies. It is
generally believed that ethanol and acetaldehyde emissions will
increase signicantly, formaldehyde emissions will increase
slightly, and benzene and 1,3-butadiene emissions will decrease
signicantly. The effects on emissions of NOx and PM are likely to
be small, but that is uncertain. To minimize adverse effects on
vehicle emissions at higher levels of ethanol use, combustion and
emissions control systems must be optimized for the blend of fuel
used. Higher levels of ethanol should not be used in conventional
vehicles. It is generally believed that urban ozone will increase in
many areas, and that increased use of ethanol blended fuels is
unlikely to improve air quality over gasoline use in future vehicles.
The data that is available for biodiesel fuel use suggest that NOx
emissions generally increase, and a more limited quantity of data
suggests that aldehyde emissions, especially formaldehyde,
increase for biodiesel blends. Previous studies suggest that by
properly adjusting the diesel engine, and using appropriate emission controls, the adverse impacts of biodiesel blended fuels could
be largely eliminated [6]. The adjustments and controls that are
required to reduce the emissions are likely to depend strongly on
the level of biodiesel blended and the biodiesel feedstock. Higher
levels and/or variable feedstock for biodiesel should not be used
merely as a drop-in for petroleum diesel. Rather, a properly
designed engine system would be required to maintain low levels
of pollutant emissions.
If renewable liquid fuels are to be used in developed countries with
stringent new vehicle emissions standards, it seems clear that combustion characteristics will need to be adjusted and emissions controls
devised that will allow bioethanol and biodiesel to meet these
standards. Since it will be desirable to be able to use different
percentages of biofuels in the blends, to allow increasing use in the
future, and to adapt to changing costs and availability of the biofuel and
fossil fuel components, it will be necessary to develop sensors that can
detect the percent composition of the biofuel mix in use. Similarly, in
the case of biodiesel, it will be necessary to develop a sensor that can
identify the biodiesel feedstock. These parameters would be used to
select the optimized combustion conditions that will minimize pollutant emissions from the vehicle. The development and implementation
of these technologies will take a few years. In developing countries,
where implementation of emissions standards lag, the impacts of
increased biofuels use is expected to be more signicant.
Biomethane has an advantage as a renewable fuel, since the
vehicle technology for using it under low emission operating
conditions has already been fairly well developed for natural gas.
Additional efforts are necessary to further reduce the CH4 emissions from natural gas fueled vehicles. Light-duty vehicles fueled
by CNG have similar emissions to comparable gasoline fueled
vehicles. It is expected that CNG-hybrid technologies may lead to

further efciencies and emissions reductions. There is currently

some variability in the natural gas fuel composition, and there is
likely to be more in the future, as increasing quantities of
biomethane are introduced into the fuel system. Thus far, the
limited data available does not show much of an effect on natural
gas fueled vehicle emissions with the natural gas composition.
There are other signicant impediments to natural gas/biomethane use as a transportation fuel. The infrastructure for lling
vehicles with it will require signicant development. Much of the
current natural gas fuel use is for eets, which operate their own
fueling systems; public availability of fueling systems will be
necessary before natural gas use will expand signicantly. In areas
of the world that have natural gas pipelines, which is an obvious
means of distributing the product, signicant changes will still be
required there as well. The sulfur based odorant added to natural
gas to make it easy to detect leaks cannot be used in vehicle fuel,
since it will prevent the proper operation of catalysts systems on
the vehicle. The leakage of natural gas from the pipeline, and
vehicle fueling will need to be reduced, as will unburned methane
emissions. Natural gas composition standards would need to be
developed for the gas in the system that is consistent with the
current fossil origin natural gas in the system, and the biomethane
enrichment capabilities for biogas that will ultimately displace the
fossil methane. The biggest barrier in many countries will be the
development and expansion of biomethane production that could
ultimately displace fossil natural gas.
Renewable fuels may lead to decreases in greenhouse gas emissions, even with increasing use of transport fuels. No improvement,
and more likely deteriorating air quality, is expected in the future with
increasing renewable fuel use. This leads to an important question, is
the use of the internal combustion engine in motor vehicles sustainable, even with the use of renewable biofuels? Battery-electric vehicles
and fuel cell use in vehicles are much more likely to be a sustainable
means for powering the transport of people and goods on the roads
and highways of the world.
References
[1] Exxon Mobil. The outlook for energy: a view to 2040, http://cdn.exxonmobil.
com/  /media/Reports/Outlook%20For%20Energy/2014/2014-Outlook-for-E
nergy-print-resolution.pdf; 2014 (last accessed March 25, 2014). p. 58.
[2] Prugh T. Biofuel Production Declines. Worldwatch Institute 2014, http://
worldwatch.org/biofuel-production-declines-1; 2014 (last accessed June 25,
2014).
[3] Farrell AE, Sperling D. A low-carbon fuel standard for California, Part 1: Technical
analysis, UCD-ITS-RR-07-07 Davis, California, http://www.energy.ca.gov/low_car
bon_fuel_standard/UC_LCFS_study_Part_1-FINAL.pdf; 2007 (last accessed June 25,
2014). p. 181.
[4] Natural & bio Gas Vehicle Association Europe. Sweden, http://www.ngvaeu
rope.eu/sweden; 2014. July 1, 2014.
[5] Anderson LG. Ethanol fuel use in Brazil: air quality impacts. Energy Environ Sci
2009;2:101537.
[6] Anderson LG. Effects of biodiesel fuels use on vehicle emissions. J Sustainable
Energy Environ 2012;3:3547.
[7] Anderson LG, Wilkes EB. Effects of oxygenated fuels use on I/M240 emissions
test results in colorado. In: Air and waste management association annual
meeting; 1998.
[8] Jacobson MZ. Effects of ethanol (E85) versus gasoline vehicles on cancer and
mortality in the United States. Environ Sci Technol 2007;41:41507.
[9] Lpez-Aparicio S, Hak C. Evaluation of the use of bioethanol fuelled buses
based on ambient air pollution screening and on-road measurements. Sci Total
Environ 2013;452453:409.
[10] Salvo A, Geiger FM. Reduction in local ozone levels in urban Sao Paulo due to a
shift from ethanol to gasoline use. Nat Geosci 2014;7:4508.
[11] Madronich S. Atmospheric chemistry: ethanol and ozone. Nat Geosci
2014;7:3957.
[12] Nogueira T, Dominutti PA, de Carvalho LRF, Fornaro A, MdF. Andrade.
Formaldehyde and acetaldehyde measurements in urban atmosphere
impacted by the use of ethanol biofuel: metropolitan area of Sao Paulo
(MASP), 20122013. Fuel 2014;134:50513.
[13] Giakoumis EG. A statistical investigation of biodiesel effects on regulated
exhaust emissions during transient cycles. Appl Energy 2012;98:27391.
[14] Cahill TM, Okamoto RA. Emissions of acrolein and other aldehydes from
biodiesel-fueled heavy-duty vehicles. Environ Sci Technol 2012;46:83828.

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

[15] Hesterberg T, Lapin C, Bunn W. A comparison of emissions from vehicles

fueled with diesel or compressed natural gas. Environ Sci Technol
2008;42:643745.
[16] Wang WG, Clark NN, Lyons DW, Yang RM, Gautam M, Bata RM, et al.
Emissions comparisons from alternative fuel buses and diesel buses with a
chassis dynamometer testing facility. Environ Sci Technol 1997;31:31327.
[17] McCormick RL, Graboski MS, Alleman TL, Yanowitz J. Idle emissions from
heavy-duty diesel and natural gas vehicles at high altitude. J Air Waste
Manage 2000;50:19928.
[18] Chandler K, Norton P, Clark N. Dallas area rapid transits (DART) LNG bus eet:
nal results, NREL/BR-540-28739 Golden, Colorado, http://www.nrel.gov/
docs/fy01osti/28739.pdf; 2000 (last accessed March 25, 2014). p. 47.
[19] Northeast Advanced Vehicle Consortium, M. J. Bradley & Associates, West
Virginia University. Hybrid-electric drive heavy-duty vehicle testing project:
nal emissions report, NAVC1098-PG009837 Boston, MA, http://walshcar
lines.com/pdf/Navc9837.pdf; 2000 (last accessed April 7, 2014). p. 74.
[20] Holmen BA, Ayala A. Ultrane PM emissions from natural gas, oxidationcatalyst diesel, and particle-trap diesel heavy-duty transit buses. Environ Sci
Technol 2002;36:504150.
[21] Seguelong T, Coroller P, Plassat G. Comparative study on exhaust emissions
from diesel- and CNG-powered urban buses. In: Diesel engine emissions
reduction conference, Newport, Rhode Island, http://energy.gov/sites/prod/
les/2014/03/f9/deer_2003_seguelong.pdf; 2003 (last accessed April 7, 2014).
[22] Lowell DM, Parsley W, Bush C, Zupo D. Comparison of clean diesel buses to
CNG buses. Diesel engine emissions reduction (DEER) workshop Newport,
Rhode Island 2003, http://www.osti.gov/scitech/servlets/purl/829622; 2003
(last accessed April 7, 2014).
[23] Wayne WS, Clark NN, Nine RD, Elefante D. A comparison of emissions and fuel
economy from hybrid-electric and conventional-drive transit buses. Energy
Fuel 2004;18:25770.
[24] Tang S, Graham L, Shen L, Zhou X, Lani T. Simultaneous determination of
carbonyls and NO2 in exhausts of heavy-duty diesel trucks and transit buses
by HPLC following 2,4-dinitrophenylhydrazine cartridge collection. Environ
Sci Technol 2004;38:596876.
[25] Nylund N, Erkkil K, Lappi M, Ikonen M. Transit bus emission study:
comparison of emissions from diesel and natural gas buses, PRO3/P5150/04,
http://www.vtt./inf/pdf/jurelinkit/VTTNylund.pdf; 2004 (last accessed
March 25, 2014).p. 63.
[26] Kado NY, Okamoto RA, Kuzmicky PA, Kobayashi R, Ayala A, Gebel ME, et al.
Emissions of toxic pollutants from compressed natural gas and low sulfur
diesel-fueled heavy-duty transit buses tested over multiple driving cycles.
Environ Sci Technol 2005;39:763849.
[27] Melendez M, Taylor J, Zuboy J, Wayne WS, Smith D. Emission Testing of
Washington Metropolitan Area Transit Authority (WMATA) natural gas and
diesel transit buses, NREL/TP-540-36355 Golden, Colorado, http://www.afdc.
energy.gov/pdfs/36355.pdf; 2005 (last accessed March 25, 2014). p. 40.
[28] Okamoto RA, Kado NY, Kuzmicky PA, Ayala A, Kobayashi R. Unregulated
emissions from compressed natural gas (CNG) transit buses congured with
and without oxidation catalyst. Environ Sci Technol 2006;40:33241.
[29] Kappanna HK. Reduction of toxic air contaminants (TACs) and particulate
matter emissions from heavy-duty natural gas engines. Morgantown, West
Virginia: West Virginia University, http://wvuscholar.wvu.edu:8881//exlibris/
dtl/d3_1/apache_media/L2V4bGlicmlzL2R0bC9kM18xL2FwYWNoZV9tZWRpY
S8yMTAxNw==.pdf; 2006 (last accessed April 7, 2014).
[30] Nylund N-O, Erkkil K, Clark N, Rideout G. Evaluation of duty cycles for heavyduty urban vehicles, VTT research notes 2396, http://www.vtt./inf/pdf/
tiedotteet/2007/T2396.pdf; 2007 (last accessed March 25, 2014). p. 96.
[31] Hausberger S, Vuckovic T. Emissions and Fuel Consumption of Clean City Bus
Concepts, FVT-85/06 Graz, Austria 2007, http://www.bafu.admin.ch/climater
eporting/00545/01913/index.html?download=NHzLpZeg7t,lnp6I0N
TU042l2Z6ln1ad1IZn4Z2qZpnO2Yuq2Z6gpJCGe352fWym162epYbg2c_
JjKbNoKSn6A&lang=en; 2007 (last accessed April 7, 2014).
[32] Wayne WS, Clark NN, Khan AS, Gautam M, Thompson GJ, Lyons DW.
Regulated and non-regulated emissions and fuel economy from conventional
diesel, hybrid-electric diesel, and natural gas transit buses. J Transp Res Forum
2008;47:10525.
[33] Padmavathy A. Evaluation of exhaust after-treatment device effectiveness in
reducing regulated and unregulated emissions from natural gas fueled heavy
duty transit bus. Morgantown, West Virginia: West Virginia University; 2008.
[34] Jayaratne ER, Ristovski ZD, Meyer N, Morawska L. Particle and gaseous
emissions from compressed natural gas and ultralow sulphur diesel-fuelled
buses at four steady engine loads. Sci Total Environ 2009;407:284552.
[35] Gautam M, Thiruvengadam A, Carder D, Besch MC, Shade B, Thompson G,
et al. Testing of volatile and nonvolatile emissions from advanced technology
natural gas vehicles, 07-340 Morgantown, WV, http://www.arb.ca.gov/
research/apr/past/07-340.pdf; 2011 (last accessed April 7, 2014). p. 75.
[36] Nylund N-O, Koponen K. Fuel and technology alternatives for buses: overall
energy efciency and emission performance, T46, http://www.vtt./inf/pdf/
technology/2012/T46.pdf; 2012 (last accessed March 25, 2014). p. 402.
[37] Hajbabaei M, Karavalakis G, Johnson KC, Lee L, Durbin TD. Impact of natural
gas fuel composition on criteria, toxic, and particle emissions from transit
buses equipped with lean burn and stoichiometric engines. Energy
2013;62:42534.
[38] Yoon S, Collins J, Thuruvengadam A, Gautam M, Herner J, Ayala A. Criteria
pollutant and greenhouse gas emissions from CNG transit buses equipped

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]
[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

171

with three-way catalysts compared to lean-burn engines and oxidation

catalyst technologies. J Air Waste Manage 2013;63:92633.
Willner K. Testing unregulated emissions from heavy duty natural gas
vehicles, SGC Rapport 2013:289 Malmo, Sweden, http://www.erdgas-mobil.
de/leadmin/downloads/Presse/Studien_Artikel_Broschueren/Study_Emis
sions_HeavyDuty_NGVs.pdf; 2013 (last accessed April 7, 2014). p. 67.
M.J. Bradley & Associates . Comparison of modern CNG, diesel and diesel
hybrid-electric transit buses: efciency & environmental performance,
Concord, MA, http://www.mjbradley.com/sites/default/les/CNG%20Diesel%
20Hybrid%20Comparison%20FINAL%2005nov13.pdf; 2013 (last accessed
March 25, 2014). p. 23.
Yoon S, Hu S, Kado NY, Thiruvengadam A, Collins JF, Gautam M, et al.
Chemical and toxicological properties of emissions from CNG transit buses
equipped with three-way catalysts compared to lean-burn engines and
oxidation catalyst technologies. Atmos Environ 2014;83:2208.
Dieselnet.com. Emission Standards European Union heavy-duty truck and bus
engines, https://www.dieselnet.com/standards/eu/hd.php; 2014 (last accessed
June 21, 2014).
Jayaratne ER, Meyer NK, Ristovski ZD, Morawska L, Miljevic B. Critical analysis
of high particle number emissions from accelerating compressed natural gas
buses. Environ Sci Technol 2010;44:372431.
Hallquist , Jerksj M, Fallgren H, Westerlund J, Sjdin . Particle and gaseous
emissions from individual diesel and CNG buses. Atmos Chem Phys Discuss
2012;12:2773773.
Ayala A, Kado N, Okamoto R, Gebel M, Rieger P, Kobayashi R, et al. CNG and
diesel transit bus emissions in review. In: Diesel engine emissions reduction
conference Newport, Rhode Island, http://energy.gov/sites/prod/les/2014/
03/f9/deer_2003_ayala.pdf; 2003 (last accessed April 7, 2014).
Thamsiriroj T, Smyth H, Murphy JD. A roadmap for the introduction of gaseous
transport fuel: a case study for renewable natural gas in Ireland. Renewable
Sustainable Energy Rev 2011;15:464251.
hman M. Biomethane in the transport sectoran appraisal of the forgotten
option. Energy Policy 2010;38:20817.
Patterson T, Esteves S, Dinsdale R, Guwy A. An evaluation of the policy and
techno-economic factors affecting the potential for biogas upgrading for
transport fuel use in the UK. Energy Policy 2011;39:180616.
Chandler K, Walkowicz K, Clark N. United Parcel Service (UPS) CNG Truck Fleet:
Final Results, NREL/BR-540-31227 Golden, Colorado, http://www.nrel.gov/vehicle
sandfuels/eettest/pdfs/31227.pdf; 2002 (last accessed March 25, 2014). p. 42.
Lyford-Pike EJ. An emission and performance comparison of the natural gas Cgas plus engine in heavy-duty trucks, NREL/SR-540-32863 Golden, Colorado,
http://www.nrel.gov/docs/fy03osti/32863.pdf; 2003 (last accessed March 25,
2014). p. 40.
Herner JD, Hu S, Robertson WH, Huai T, Collins JF, Dwyer H, et al. Effect of
advanced aftertreatment for PM and NOx control on heavy-duty diesel truck
emissions. Environ Sci Technol 2009;43:592833.
Hu S, Herner JD, Robertson W, Kobayashi R, Chang M-CO, Huang S-m, et al.
Emissions of polycyclic aromatic hydrocarbons (PAHs) and nitro-PAHs from
heavy-duty diesel vehicles with DPF and SCR. J Air Waste Manage
2013;63:98496.
Clark NN, Rapp BL, Gautam M, Wang W, Lyons DW. A long term eld
emissions study of natural gas fueled refuse haulers in New York City SAE
international fall fuels and lubricants meeting, San Francisco, CA, http://www.
osti.gov/scitech/servlets/purl/771106; 1998 (last accessed April 9, 2014).
Chandler K, Norton P, Clark N. Waste managements LNG truck eet: nal
results, NREL/BR-540-29073 Golden, Colorado, http://www.afdc.energy.gov/
pdfs/waste_lng_nal.pdf; 2001 (last accessed March 25, 2014). p. 50.
M.J. Bradley & Associates A demonstration of diesel particulate lter emission
control technologies on refuse collection trucks and deployment of natural gas
powered street sweepers, Boston, MA, www.nescaum.org/documents/cum
mins_dsny_report_nal.pdf/; 2005 (last accessed April 9, 2014). p. 124.
Karavalakis G, Hajbabaei M, Durbin TD, Johnson KC, Zheng Z, Miller WJ. The
effect of natural gas composition on the regulated emissions, gaseous toxic
pollutants, and ultrane particle number emissions from a refuse hauler
vehicle. Energy 2013;50:28091.
Blazek CF, Grimes J, Freeman P, Bailey BK, Colucci C. Fuel composition effects on
natural gas vehicle emissions. In: 207th ACS national meeting, San Diego, CA,
https://web.anl.gov/PCS/acsfuel/preprint%20archive/Files/39_2_SAN%
20DIEGO_03-94_0476.pdf; 1994 (last accessed April 9, 2014).
Motta RC, Kelly KJ, Warnock WW. Compressed natural gas and liqueed
petroleum gas conversions: the national renewable energy laboratorys
experience, NREL/SP-425-20514 Golden, Colorado. http://www.afdc.energy.
gov/pdfs/conversion.pdf; 1996 (last access April 9, 2014). p. 19.
Durbin TD, Truex, T. J., Norbeck, J. M. Particulate measurements and emissions
characterization of alternative fuel vehicle exhaust, 98-VE-RT12-010-FR Riverside, California, http://cichlid.cert.ucr.edu/research/pubs/98-ve-rt12-010-fr.
pdf; 1998 (last accessed April 9, 2014). p. 69.
Norbeck JM, Truex TJ, Smith MR, T. D. Inventory of AFVs and AFV comparison:
OEM vs. retrots, 98-VE-RT2W-008-FR Riverside, California, http://cichlid.
cert.ucr.edu/research/pubs/98-ve-rt2w-008-fr.pdf; 1998 (last accessed April
9, 2014). p. 85.
Durbin TD, Norbeck JM, Huai T. Investigation of emission rates of ammonia
and other toxic and low-level compounds using FTIR, 01-VE-20903-001-FR
riverside, CA, http://cichlid.cert.ucr.edu/research/pubs/20903-ammonia.pdf;
2001 (last accessed April 17, 2014). p. 53.

172

L.G. Anderson / Renewable and Sustainable Energy Reviews 47 (2015) 162172

[62] Ristovski Z, Morawska L, Ayoko GA, Johnson G, Gilbert D, Greenaway C.

Emissions from a vehicle tted to operate on either petrol or compressed
natural gas. Sci Total Environ 2004;323:17994.
[63] Wright S, Pinkelman A. Natural gas internal combustion engine hybrid
passenger vehicle. Int J Energy Res 2008;32:61222.
[64] Karavalakis G, Durbin TD, Villela M, Miller JW. Air pollutant emissions of lightduty vehicles operating on various natural gas compositions. J Nat Gas Sci Eng
2012;4:816.
[65] Subramanian KA, Mathad VC, Vijay VK, Subbarao PMV. Comparative evaluation of emission and fuel economy of an automotive spark ignition vehicle
fuelled with methane enriched biogas and CNG using chassis dynamometer.
Appl Energy 2013;105:1729.

[66] Bielaczyc P, Woodburn J, Szczotka A. An assessment of regulated emissions

and CO2 emissions from a European light-duty CNG-fueled vehicle in the
context of Euro 6 emissions regulations. Appl Energy 2014;117:13441.
[67] Martins AA, Rocha RAD, Sodr JR. Cold start and full cycle emissions from a
exible fuel vehicle operating with natural gas, ethanol and gasoline. J Nat Gas
Sci Eng 2014;17:948.
[68] Correa SM, Arbilla G. Formaldehyde and acetaldehyde associated with the use
of natural gas as a fuel for light vehicles. Atmos Environ 2005;39:45138.
[69] Myung C, Lee H, Choi K, Lee Y, Park S. Effects of gasoline, diesel, LPG, and lowcarbon fuels and various certication modes on nanoparticle emission
characteristics in light-duty vehicles. Int J Automot Technol 2009;10:53744.