Reduction of Pollutants Emissions on SI Engines …
J. of the Braz. Soc. of Mech. Sci. & Eng. Copyright © 2005 by ABCM July-September 2005, Vol. XXVII, No. 3 /
217
J. N. de S. Vianna
A. do V. Reis
and A. B. de S. Oliveira
Universidade de Brasília
Departament de Engenharia Mecânica
79910-900 Brasília, DF. Brazil
vianna@unb.br, alessandro.reis@bol.com.br
abso@unb.br
A. G. Fraga
Petrobrás
Rua Hilton Rodrigues, 71
41830-635 Salvador, BA. Brazil
andrei@petrobras.com.br
M. T. de Sousa
Volkswagen
Rua Volkswagen, 100
27000-000 Rezende, RJ. Brazil
marcelo.sousa@volkswagen.com.br
Reduction of Pollutants Emissions on
SI Engines - Accomplishments With
Efficiency Increase
This paper presents an experimental study aiming to identify the means to minimize the
reduction of the overall performance of a gasoline engine when employing the Exhaust-
Gas Recirculation (EGR) technique that reduces NOx emissions. The increase of the
compression ratio and turbocharging was evaluated as a mean to recover the original
performance. The formation of pollutants and the engine performance were verified at full
and partial loads. The results show that the combination of exhaust gas recirculation with
turbocharger or through an increase of the compression ratio enhance the relation
between the engine performance and the emission of NO. However, the turbocharger
seemed to be more sensitive to the negative effects of the EGR technology.
Keywords: EGR, NOx, pollutant emissions control
Introduction
The methods and techniques to reduce emission of pollutants
from internal combustion engines usually decrease its performance.
Considering the impossibility of a short term modification in the
current standards of energy consumption, the most effective way for
reducing environmental impacts relies on increasing the efficiency
of the thermal engines. In other words, research should be carried
out on development of more efficient engines or to apply means, for
the current level of technology, to minimize entropy generation.
Specifically, for internal combustion engines, a reasonable solution
is the reduction on pollutant formation by controlling some
combustion parameters in such way that engine performance is kept
unaltered. An effective way for reducing nitrous oxide (NO
x
)
emissions may be accomplished by changing the engine combustion
process through the recycling of exhausted gases. This process is
accomplished by adding combustion products to the fresh fuel-air
mixture during the intake process. This technology is known as
Exhaust Gas Recirculation (EGR) and has been applied in both
spark ignition engines and compression ignition engines. The
presence of inert molecules reduces the temperature and the
combustion pressure inhibiting the formation of NO, by the thermal
mechanism, as well as increases the detonation tolerance, (Heyhood,
1998). This method, however, while effective in reducing NO
x
emissions, may lead to considerable losses in engine performance.
Several authors, Sato et al (1997), Sousa (2000), Kohketsu (1997),
Han, S., and Cheng, W.K.(1998) amongst others, have discussed the
advantages and disadvantages of the EGR technology. They have,
also, proposed ways to minimize the drawbacks when applying
EGR technology in different types of diesel and gas engines.
1
Bortolet et al. (1999), presented a fuzzy modeling method to an
engine air inlet that operates on EGR. Abd-Alla and co-workers
(2001), investigated the effects of diluent admissions and intake air
Paper accepted June, 2005. Technical Editor: Atila P. Silva Freire.
temperature in EGR on the emissions of an indirect injection dual
fuel engine. They found that diluent addition decreased NOx
emissions. Even larger reduction was observed when carbon dioxide
was added to the inlet gaseous fuel air charge. Abd-Alla (2002),
presented a review on exhaust gas recirculation applied to internal
combustion engines. The aim of the work was to review the
potential of EGR to reduce exhaust emissions, mostly NO
x
, and to
delimit the application range of the technology. Zheng et al. (2004),
reviewed the advanced and novel concepts in diesel engine exhaust
gas recirculation. They claimed that EGR is effective to reduce
nitrogen oxides from diesel engines while increasing particulate
matter. Power losses are significant when EGR is further increased.
Different ways to implement EGR are outlined in the paper. In
addition, new concepts regarding EGR stream treatment and EGR
hydrogen reforming are proposed. More recently, Lü and co-
workers (2005) conducted a fundamental study on the control of the
HCCI combustion and emissions by fuel design concepts combined
with controllable EGR. Cooling EGR prolonged the time for
combustion. EGR had little effects on CO and UHC emissions on
HCCI engines.
Despite the number of works on EGR technology, there is still a
room for further investigation in SI engines when combining EGR,
for pollutant reduction, and turbo charging for performance
recovery. The lack of information on this specific subject has,
therefore, motivated this work.
Inert gases have a combined effect of reducing NO
x
and increase
the knocking tolerance, (Heyhood, 1998). Therefore, it is possible to
increase the compression ratio as well as to apply turbocharging
without reaching self-ignition levels that may jeopardize engine
integrity. It is well known that increasing compression ratio or turbo
charging the engine favor the NO formation because combustion
peak temperature is augmented. Conversely, inert gases mixed with
the intake charge, inhibit the formation of NO
x
, during the
combustion process. In this work, we searched for the best
combination on EGR application and engine performance recover.
J. N. de S. Vianna e al
/ Vol. XXVII, No. 3, July-September 2005 ABCM 218
The results are presented and discussed only for a 3000 rpm, since
the trends remain for a broad range of engine speeds.
Experimental Apparatus and Methodology
The experimental apparatus comprised a hydraulic
dynamometer with the auxiliary instruments allowing a complete
monitoring of the main engine parameters, such as, torque, power,
fuel consumption, air consumption, temperature and related
pressures. A gas analyzer was used for measuring the concentrations
of CO
2
, CO, O
2
, NO
x
and unburned hydrocarbon, in the combustion
products. This analyzer also provided the air-fuel ratio based on the
concentration of some specific gases in the exhaust system.
The dynamic pressure inside the cylinder was also measured.
The sensor was installed in the cylinder head as a mean to track
knocking occurrence. The primary element was a piezoelectric
sensor with an operational band from 0 to 250 bar. This element was
associated to a system set for acquisition and data treatment.. The
signal obtained by the sensor was amplified and processed by a
dynamic signal analyzer. All this system was in compliance with the
ISO TAG4/WG3 (1999) and the Vianna et al. (1999), procedures for
data acquisition. For frequencies, the maximum uncertainty was
1.96% of the measured value up to 1.2 kHz and of 3.2% for
frequencies ranging between 1.2 and 1.6 kHz.
The indication of the top dead center (TDC) was done through
an optical sensor and a perforated disk installed at the extremity of
the dynamometer shaft. The deviation of this measurement, in
relation to the geometric measuring, presented an error of 0.3
o
regardless of the speed, as observed by Oliveira et al. (1996).
It was used, for the investigation, a 4-cylinder engine, 1927 cm
3
displacement, 8.2:1 compression ratio, single-point fuel injection.
Several experiments were conducted in order to set reference
parameters, necessary for further comparisons. Volumetric fractions
of CO
2
, CO, O
2
, NO
x
,
and unburned hydrocarbons were also
obtained as a reference. Commercial gasoline with 23% of
anhydrous alcohol was used and the tests were conducted at 2500,
3000 and 4000 rpm. The dynamic pressure inside the cylinder was
measured for speeds. Test planning is shown in Table 1. Test 1 was
set as reference.
The investigation started with the EGR valve installed between
the intake and exhaust manifolds (Test 2). This valve was specially
designed to vary the amount of exhaust gases that were added to the
air-fuel mixture in the intake manifold. For each engine speed, the
amount of exhaust gases added to the fresh mixture was varied. The
runs were then conducted, at full load, by measuring all the relevant
parameters.
Test 3 combined EGR application along with turbo charging but
keeping compression ratio the same as in Tests 1 and 2. This set up
was replicated in Test 4 though at partial loads, 50 and 70%.
The compression ratio was finally altered to 8.9:1 (Tests 5 and
6). EGR was applied in Test 6 and set off in Test 5. Compression
ratio was increases using a cylinder head with a smaller combustion
chamber, but with the same geometry.
When the compression ratio was 8.2:1, the engine operated with
turbocharger. In this case, the EGR system was installed before the
compressor inlet.
In Fig. 1 the EGR assembling is shown where it can be seen the
details and the accessories for a naturally aspirated and
supercharged engine configurations.
In order to ensure consistency of the data, stoichiometric air-fuel
ratio was set, to all speeds of the engine. In two configurations, for
naturally aspirated engine, the ignition angle was adjusted to the
maximum torque in the dynamometer, for each speed. When
operating with turbocharger, the optimization of the ignition timing,
based on the torque, was limited by the presence of detonation,
checked through the dynamic pressure curves.
Table 1. Test Planning.
Test Compression
Ratio
EGR Turbocharge Load
1 8.2 uninstalled uninstalled Full
2 8.2 installed uninstalled Full
3 8.2 installed installed Full
4 8.2 installed installed 50 and 70%
5 8.9 uninstalled uninstalled Full
6 8.9 installed uninstalled Full
(a) (b)
Figure 1. Exhaust gases recirculation system assembly for naturally
aspirated (a) and turbocharged (b) – System components: 1 humidity
separator; 2 booster; 3 EGR valve; 4 single point injection; 5 heat
exchanger; 6 compressor; 7 single point injection and equalization box; 8
turbine.
The tests were divided in three stages. The first stage conducted
in a naturally aspirated engine, for which the compression ratios
were set to both 8.12:1 and 8.9:1. Tests were also conducted with
turbocharging, Test 3 and 4. These tests, under full load, allowed
the assessment of the effect of the exhaust-gas recirculation in the
global performance of the engine, for every planned configuration.
These experiments also allowed a better understanding of the effects
of the EGR technology on the operation of the engine. As a first
conclusion, the results showed that the engine presented better
overall performance including reduction of the emissions when
operating under supercharging. The second stage was conducted
with the turbocharger, still under full load, and aimed to infer
whether the positive effects of the recirculation remained when
some operational parameters were optimized. In the third and last
stage, experiments were conducted under partial loads, keeping the
supercharging and the optimized configuration.
The following procedures were adopted for each of the three
stages:
Procedure 1
: this procedure was employed for all three engine
configurations running at 2500, 3000 and 4000 rpm. For each of
these configurations, the spark timing was adjusted for the highest
torque without the presence of knocking. The stoichiometric air-fuel
ratio was set the same as the engine without recirculation.
For every speed, the recirculation ratio of the exhaust gases was
increased, without altering the load or the ignition timing. This
would result in a fall of the speed, and, as the fuel flow remained the
same, the air-fuel mixture kept enriching as the recirculation
increased. This procedure allowed the assessment of the effects of
applying EGR technology, for the different operating regimes. Also,
after the experiments, it was clear which configuration was more
sensitive to the negative effects of the EGR. The results clearly
indicated that turbocharging is the configuration that associates the
best performance to the highest reduction on the emissions.
Reduction of Pollutants Emissions on SI Engines …
J. of the Braz. Soc. of Mech. Sci. & Eng. Copyright © 2005 by ABCM July-September 2005, Vol. XXVII, No. 3 /
219
Procedure 2: based on the results achieved in procedure 1, the
engine with supercharging was tested for the same speeds, however,
the air-fuel ratio was kept stoichiometric and the ignition timing
adjusted for the maximum torque, for each recirculation ratio.
Procedure 3: again with turbocharging and the engine optimized
to full load, the investigation was also conducted with 75% and 50%
of the full load, but varying the recirculation ratio.
The degree of recirculation (percentage) was defined by relating
the amount of CO
2
in the intake manifold with that in the exhaust
pipe, through the Eq. (1).
%100
COCO
COCO
(%)EGR
amb2ex2
amb2adm2
⋅
−
−
=
, (1)
where the carbon dioxide concentrations used in the expression are
such that CO
2amb
was measured in the environment, CO
2adm
was
checked in the intake manifold and CO
2ex
was measured in the
exhaust gases.
Results and Discussion
The experimental results obtained in Procedure 1 allowed the
investigation of, both, engine performance and emissions levels. In
addition, it was possible to infer the effects of the recirculation in
the progress of the flame front, for the engine with compression
ratios of 8.2:1 and 8.9:1. When applying turbo charge the
compression ratio was fixed at 8.2:1.
The tests were conducted under full load, varying the degree of
recirculation (EGR) as well as engine speed. Here, only the results
at 3000 rpm are shown and discussed since for the remaining speeds
the trends were the same.
Figures 2, 3 and 4 show the dynamic pressure against crank
angle for different degree of gas recirculation as well as ignition
timing, for three configurations, Test 1, 6 and 3, respectively.
It can be observed in Fig. 2 that, keeping the amount of fuel and
the spark timing, while increasing the presence of inert gases in the
combustion chamber, the flame front was decelerated resulting a
displacement of the pressure curve in relation to the Top Dead
Center (TDC). The maximum pressure falls from 38 MPa, without
recirculation, to 30 MPa with 8.11% of EGR. One of the
consequences is the reduction of the net work of the cycle,
considering that the curves, in the compression stage, are coincident
until very close to the TDC. The practical results were, then, lower
torque and power, which, in turn, decreases the engine global
performance.
Figure 2. Pressure curves in the interior of the combustion chamber as a
result of the crankshaft angle and the percentage of recirculation for a
3000 rpm for the naturally aspirated engine with a compression ratio of
8.2:1 and ignition angle of 35°.
Figure 3. Pressure curves in the interior of the combustion chamber as a
result of the crankshaft angle and the percentage of recirculation at 3000
rpm for a naturally aspirated engine with compression ratio of 8.9:1 and
ignition angle of 37°.
Figure 3 presents the results in which the engine compression
ratio was increased to 8.9:1 and the spark timing adjusted to
maximum torque without the presence of detonation. Two cases can
be seen, first with the EGR valve deactivated and the second where
EGR was operational - Test 5 and 6, respectively. Due to the
increase in the compression ratio, it was observed an increase in the
maximum pressure up to 42 MPa, no recirculation, and as high as 35
MPa with 7.88% in gas recirculation. The trends here, regarding
engine performance were about the same as discussed in Fig. 2, Test
2.
Figure 4. Pressure curves in the interior of the combustion chamber as a
result of the crankshaft angle and the percentage of recirculation at 3000
rpm for the turbocharged engine, with a compression ratio of 8.2:1 and
ignition angle of 30°.
Figure 4 shows that, some recirculation resulted in a drastic
reduction of the maximum pressure of the cycle. However, with the
same degree of recirculation the value of the maximum pressure of
the cycles is quite similar, in all three configurations.
Also, in Fig. 2, 3 and 4 it possible to see the effect of the EGR
on the combustion ratio as well as on the maximum cycle pressures.
The reduction in the peak of the pressure is 20, 15 and 40% for the
engine operating with compression ratio of 8.12:1, 8.9:1 and
supercharged, respectively. Under such conditions, neither the spark
timing nor the air-fuel ratio were adjusted. The results show that
with turbocharger the engine was more sensitive to the increase in
gas recirculation compared to that operating in a naturally aspirated
mode. That resulted in a great loss of the net work of the cycle,
which strongly affected the effective power of the engine. This was
all confirmed by the dynamometric tests.
Figure 5 shows the results of the dynamometer tests, under full
load, for a fixed ignition angle, while varying gas recirculation
through the EGR valve. It is possible to notice that for a naturally
aspirated engine, with compression ratios of 8.2:1 and 8.9:1 did not
show great variations of power output with the increase of the
degree of exhaust gas recirculation However, for the supercharged
engine, recirculation progressively reduces the power. This loss,
J. N. de S. Vianna e al
/ Vol. XXVII, No. 3, July-September 2005 ABCM 220
when the EGR was set beyond 8%, went down to the point where
the engine operated as it were naturally aspirated, It can be seen, in
Fig. 5 that the error bars are mixed with the experimental points.
Figure 5. Power as result of the recirculation of the exhaust gases for the
engine with compression ratios of 8,2:1; 8,9:1 and turbocharged at 3000
rpm.
Figure 6. Specific Fuel Consumption as a result of the recirculation of the
exhaust gases for the engine with compression ratios of 8.2:1; 8.9:1 and
turbocharged at 3000 rpm.
Figure 7. NOx emissions as a result of the recirculation of the exhaust
gases for the engine under full load with compression ratios 8.2:1; 8.9:1
and turbocharged at 3000 rpm.
The global performance of the engine, supercharged, and
consequently, its specific fuel consumption (SFC) are strongly
affected by the recirculation. Figure 6 presents the results of the
experiments with all three configurations. The curves show a small
variation in the SFC, for naturally aspirated engine, which was not
observed when operating with turbocharger.
The NO
x
, CO and HC volume fractions were estmated in all the
tests reported. The efficiency of the EGR in inhibiting the formation
of NO and, consequently, reducing the emission of NO
x
with the
engine under full load can be observed in Fig. 7. Under these
circumstances, the recirculation allowed the increase in the
compression ratio, while keeping lower emissions of NO
x
. Higher
levels of NO
x
emissions were not observed after increasing the
compression ratio, which, in turn, enhanced engine performance.
Figure 7 also indicates that the efficiency of the recirculation, for the
turbocharged engine, is much higher than that observed for the
naturally aspirated engine. The trends are the same, regardless of
the compression ratio, even though the engine performance showed
a negative sensibility to the increase of the recirculation.
Figure 8. CO emissions as a result of the recirculation of the exhaust
gases for the engine under full load with a compression ratio 8,2:1; 8,9:1
and turbocharged at 3000 rpm.
As regarded to the emissions of other pollutant gases, the EGR
seemed to perform unfavorably. Figure 8 and 9 show that the
emissions of CO and HC increased, for 8% recirculation, about 75
and 67%, respectively,. Since EGR has a direct interference in the
combustion process this might explain the increment in the emission
of HC. The higher emissions of CO are due to the enrichment of the
mixture, as pointed out by Sousa (2000).
Figure 9. Emissions of HC as a result of the recirculaion of the exhaust
gases for the engine under full load with a compression ratio 8,2:1; 8,9:1
and turbocharged at 3000 rpm.
These preliminary results indicate that, as regarded to emissions
and engine performance, operation with supercharger seem more
appropriate than a naturally aspirated configuration. Therefore,
Procedure 2 was carried out only for the engine with turbocharger.
In this procedure the tests were done by adjusting the ignition angle
for the maximum torque while keeping stoichiometry for the air-fuel
ratio. The results of this procedure are presented in the following.
In Fig. 10 it is possible to observe the influence of the EGR on
the dynamic pressure of the combustion chamber, Test 3. It can be
seen that, by adjusting the spark angle and the speed, the loss of
output power with 0 and 4.2% of recirculation is much lower than
the one found for the same conditions of Procedure 1, fixed spark
angle. This loss of output power is observed more clearly in Fig.
11, 3.5% reduction in the power for a recirculation ratio of 4.2%.
Reduction of Pollutants Emissions on SI Engines …
J. of the Braz. Soc. of Mech. Sci. & Eng. Copyright © 2005 by ABCM July-September 2005, Vol. XXVII, No. 3 /
221
Figure 10. Dynamic pressure curves for two recirculation ratios of the
turbocharged engines, under full load, with the adjustment of the spark
angle for maximum torque, at 3000 rpm.
Figure 11. Power curve as a result of the recirculation of the turbocharged
engine, under full load, with the adjustment of the spark angle for
maximum torque, at 3000 rpm.
Figure 12. Specific Fuel Consumption (SFC) as a result of the recirculation
of the turbocharged engine, under full load, with the djustment of the
spark angle for the maximum torque, at 3000 rpm.
The loss in power output, showed in Fig. 11, was followed by a
slight reduction in the specific fuel consumption (SFC), as Fig. 12
shows. Even though the maximum recirculation ratio achieved was
not as high as that when the spark angle was fixed (Procedure 1), a
0.1% reduction in the specific fuel consumption was observed in
relation to the original engine, for the same degree of recirculation.
Figures 13, 14 and 15 present the emission of pollutant, with the
ignition timing adjusted for maximum torque under full load with
turbocharging. These results show that, under these conditions, the
turbocharger worked as a power output recover without increasing
emissions. Although the emissions of NO
x
was reduced only 22%,
with 4% recirculation, compared to a reduction of 54% achieved
with Procedure 1, HC emissions were kept relatively stable. As
regarded to CO emissions, the adjustment of the spark angle along
with constant stoichiometric air-fuel ratio was extremely effective. It
is important to mention that, under constant regime, the temperature
of the exaust gases is quite high, and therefore, emission values of
CO and HC, might be masked.
Figure 13. NOx emission curves as a result of the recirculation of the
turbocharged engine, under full load, with the adjustment of the spark
angle to full torque, at 3000 rpm.
The results under partial loads, following Procedure 3, are
presented in Fig. 16 and 17. It is possible to observe that
recirculation had positive and negative effects under partial loads.
Figure 14. CO Emission curves as a result of the recirculation of the
turbocharged engine, under full load, with the adjustment of the spark
angle to full torque, at 3000 rpm.
Figure 15. HC Emission curves as a result of the recirculation of the
turbocharged engine, under full load, with the adjustment of the spark
angle to full torque, at 3000 rpm.
