1
PERFORMANCE AND ECONOMIC ANALYSIS OF A DIRECT INJECTION SPARK IGNITION
2
ENGINE FUELED WITH WET ETHANOL
3
4
Thompson Diórdinis Metzka Lanzanova*, Macklini Dalla Nora, Hua Zhao
5
Brunel University London, Centre for Advanced Powertrain and Fuels Research (CAPF), Kingston Lane,
6
Uxbridge, Middlesex UB8 3PH, United Kingdom
7
8
* Corresponding author at: Brunel University London, Centre for Advanced Powertrain and Fuels Research
9
(CAPF), Kingston Lane, Uxbridge, Middlesex UB8 3PH, United Kingdom. Tel.: +44 7477640793; fax: +44
10
11
1895266698.
E-mail address: [email protected], [email protected] (T.D.M. Lanzanova)
12
13
HIGHLIGHTS
14
15
-
Stable SI engine operation with 20% water-in-ethanol and λ=1.3;
16
-
Greater water-in-ethanol content reduced NOx emissions;
17
-
Operational cost reduction of up to 31% was achieved;
18
19
ABSTRACT
20
21
The use of wet ethanol with higher water content than the conventionally used in internal
22
combustion engines can reduce fuel production costs due to lower energy expense during the
23
distillation phase. However, during its combustion the extra water content may result in the
24
deterioration of fuel conversion efficiency and therefore a global energy evaluation should be
25
considered. This research investigated the operation of a single cylinder direct injected spark ignition
26
engine running with gasoline, anhydrous ethanol and several wet ethanol compositions (5% to 20%
27
of water-in-ethanol volumetric content) under stoichiometric and lean air/fuel ratios. Two part load
28
conditions of 3.1 bar and 6.1 bar indicated mean effective pressure were evaluated at 1500 RPM.
29
The impacts of increased water-in-ethanol content and lean operation on combustion and emissions
30
were discussed. Higher water content affected the heat release rate, which increased the
31
combustion duration and initial flame development phase. Lower nitrogen oxides emissions could be
32
achieved with higher water-content ethanol at the expense of higher unburned hydrocarbon
33
emission. An analysis of wet ethanol energy production costs and engine operation conditions was
34
carried out. The lean engine operation with 10% (v/v) water-in-ethanol fuel showed global energy
35
savings around 31% compared to anhydrous ethanol at stoichiometric conditions.
1
1. INTRODUCTION
2
3
In the last decades the growing concern on carbon dioxide emissions has increased the demand on
4
renewable biofuels in order to complement, or even substitute, fossil fuels for automotive
5
applications. More recently with the adoption of the Paris Protocol [1] several nations have agreed
6
to reduce global greenhouse gas emissions in order to hold the global average temperature below 2
7
°C above pre-industrial levels. In this scenario, bioethanol produced from fermented sugars from
8
various agricultural crops has been explored worldwide as an alternative to gasoline in spark ignition
9
(SI) internal combustion engines (ICE).
10
11
Ethanol production can be adapted according to the local crop availability, which does not only
12
reduce oil dependency and increases energy security but also stimulates the local agricultural,
13
industrial and commercial activities in emerging countries [2,3]. In a well-to-wheel analysis, when
14
land usage for ethanol crop production is in accordance with some policies, the greenhouse gas
15
(GHG) emission of ethanol is much lower than that of fossil fuels, as most of the GHG generated
16
during its combustion and industrialization is absorbed during the crop cultivation [4–6].
17
Nevertheless, ethanol usage is still linked to its production price, which is directly related to the
18
energy consumption during the whole biofuel production cycle.
19
20
The use of ethanol in SI engines has been explored both as an anti-knock additive to gasoline and a
21
dedicated fuel. The conventional water volumetric content is around 5% when used as dedicated
22
fuel. When mixed with gasoline, the water content is usually below 1% to avoid phase separation.
23
Compared to gasoline, ethanol presents higher knock resistance and higher latent heat of
24
vaporization (904 kJ/kg for ethanol against 350 kJ/kg for gasoline). The increased ethanol charge
25
cooling effect can lead to higher volumetric efficiency [7] and lower in-cylinder heat transfer [8].
26
Ethanol direct injection (DI) with concomitant gasoline port fuel injection has been also investigated
27
[9]. In order to take advantage of the greater cooling effect, ethanol DI must be controlled in order
28
to provide enough cooling effect without fuel impingement and cold start issues. Moreover,
29
ethanol’s lower heating value (LHV) is 37% lower than that of gasoline, which increases the
30
volumetric fuel consumption for the same energy substitution. It also presents corrosive effects in
31
some alloys [7].
32
33
The energy usage for ethanol production may vary from place to place due to the chosen crop [10–
34
12] and distinct industrial technologies. In most situations the net energy balance from ethanol
1
production cycle is positive. The main ethanol production steps from cereals are milling,
2
saccharification, fermentation, distillation and dehydration. If ethanol is produced from sugar syrups
3
(molasses), which is a by-product from sugar refining processes, only fermentation, distillation and
4
dehydration processes are needed. As ethanol and water are fully miscible and form an azeotrope
5
mixture, distillation cannot be used to achieve ethanol-in-water volumetric concentrations beyond
6
95.6%. As shown in some studies [13–15], the energy expense trend to achieve ethanol-in-water
7
volume fractions up to 80% increases in a linear trend. From 80% towards the azeotropic point, the
8
energy requirement trend for distillation becomes exponential. This fact highly reduces the net
9
energy balance of the bioethanol life cycle and consequently increases its final market price. To
10
achieve anhydrous ethanol, distinct dehydration processes are used. Although great energy
11
reduction has been achieved through the use of more sustainable dehydration techniques, such as
12
molecular sieves, the energy expense is still considerably high [16]. In most cases, the ratio of
13
gained energy of fuel LHV in MJ/L to the expended energy to dehydrate the same volume of ethanol
14
(99% of ethanol or more) is very low, which further reduces the bioethanol net energy balance.
15
16
Using distillation and dehydration energy requirement data presented elsewhere [13–15] and the
17
total energy expense to produce one litre of ethanol from distinct crops worldwide [11,12,17–19], it
18
is possible to estimate the ratio of gained energy per unit of volume of fuel LHV to the expended
19
energy Eind to produce the same volume for distinct water in ethanol volume fractions. This
20
calculation shows that LHV/Eind reaches its maximum value for mixtures containing between 80 and
21
90% of ethanol-in-water. These ethanol-water mixtures would provide the best net energy balance
22
compromise and the best monetary profit once the fuel conversion efficiency could be kept similar.
23
Nevertheless, a deeper analysis of using such fuels in current spark ignition engines has not been
24
fully proposed.
25
26
Previous studies using wet ethanol were carried out in different engines. The use of a catalytic
27
igniter has been explored to efficiently burn wet ethanol with up to 30% of water content [20,21].
28
Lower NOx emission and higher brake conversion efficiency were obtained compared to gasoline
29
operation. Homogeneous Charge Compression Ignition (HCCI) through intake air heating in high
30
compression ratio engines has also been extensively explored in an effort to reduce gaseous
31
emissions and achieve higher engine efficiencies while using wet ethanol (up to 40% of water
32
content) [22–24]. Negative valve overlap (NVO) has been explored to reach HCCI operation through
33
hot residuals trapping – achieved through early exhaust valve closure and late intake valve opening –
34
with up to 20% water wet ethanol for boosted operation. The water content showed a negative
1
effect on reducing the pressure rise rates and the maximum operational lambda [25,26]. Studies
2
conducted on a pre-chamber SI combustion engine concept have shown that it was possible to
3
maintain the hydrous ethanol (5% of water) engine operation efficiency at part load operation with
4
up to 30% of water in ethanol [27,28]. The simultaneous use of port fuel injection of hydrous ethanol
5
and direct injection of diesel has also been investigated in heavy duty engines using reactivity-
6
controlled compression ignition (RCCI) [29,30]. It has been shown that diesel like efficiencies can be
7
achieved with RCCI combustion for a wide range of loads. Diesel could be replaced by wet ethanol
8
(30% of water-in-ethanol in mass), whilst reducing NOx and soot emissions.
9
10
The expected effects of water addition in spark ignition engines at full load are lower flame growth
11
speed [31] and reduced heat release rate and peak in-cylinder pressure. The water dilution effect
12
reduces the peak temperatures and hence NOx emissions [32], although combustion efficiency is
13
penalized by higher aldehyde and total hydrocarbon (THC) emissions [33]. One of the main concerns
14
about wet ethanol operation in SI engines is the excessive wall wetting in port fuel injection engines
15
or fuel impingement in DI engines. Both events reduce the combustion efficiency and further
16
increase THC and aldehyde emissions.
17
18
Although some researches in SI wet ethanol operation have been carried out [27,28,31–33], none of
19
them presented an in-depth analysis of the part load operation. The lean SI operation with wet
20
ethanol has been not yet investigated. Therefore, this paper compares the combustion and emission
21
characteristics of a DI SI single cylinder engine operating with anhydrous ethanol (E100), hydrous
22
ethanol (E95W05), and two ethanol-in-water mixtures containing 10% and 20% of water in volume
23
(E90W10 and E80W20, respectively). Two operation loads of 3.1 bar and 6.1 bar indicated mean
24
effective pressure (IMEP) at 1500 rpm, with distinct air dilution strategies (lean operation), were
25
investigated. Herein, all ethanol-in-water mixtures will be called wet ethanol. Commercial RON 95
26
unleaded gasoline (GRON95) was also employed in the study. Finally, a brief energy cost analysis of
27
the wet ethanol production and engine operation efficiency was carried out so the economic impact
28
of wet ethanol could be accessed. As a result, the use of wet ethanol in a modern direct injection
29
spark ignition engine at part load under stoichiometric and lean operation was investigated. A
30
complete engine performance, combustion and emission analysis was carried out. The possible
31
economic benefits and challenges of using wet ethanol were also discussed.
32
33
34
1
2. EXPERIMENTAL SETUP
2
3
The engine used in this work is a Ricardo Hydra single cylinder direct injection engine. It has an
4
electro-hydraulic fully variable valve train (FVVT) system providing independent control over the
5
four valves. It also enables the operation in both two/four-stroke modes [34], though in this work
6
only the four-stroke cycle was explored. Table 1 provides the main engine specifications and Figure 1
7
presents the test cell setup, where symbols containing T and P represent temperature and pressure
8
transducers, respectively, and λ represents the wide band universal exhaust gas oxygen (UEGO)
9
sensor. Spark timing and fuel injection timing/quantity were assessed via an engine control unit. A
10
valve control unit managed the intake and exhaust valve opening and closure timings and the
11
maximum lifts. The engine load was manually controlled by an intake throttle and the speed was
12
kept constant at 1500 rpm by means of an active AC dynamometer
13
14
The fuel and air mass flow rates were measured through a Coriolis fuel flow meter and a laminar
15
flow meter, respectively. Two piezo-resistive pressure transducers were used to monitor the intake
16
and exhaust port instantaneous pressures, whilst a piezo-electric transducer coupled to a charge
17
amplifier recorded the in-cylinder pressure. K-type thermocouples were installed for measurements
18
of average temperatures in the intake and exhaust ports, oil and coolant galleries, etc. Engine
19
coolant and oil temperatures were kept constant at 363 K.
20
21
Table 1
Engine Specifications
Bore x Stroke
81.6 mm x 66.9 mm
Swept Volume
350 cm³
Compression Ratio
11.8 : 1
Combustion Chamber
Pent roof
Valve train
4
valves,
electro-hydraulic
actuation
Direct injection – side mounted
Fuelling method
Magneti
Marelli
solenoid type injector
Injection Pressure/Temperature
22
145±5 bar / 293±5 K
six-holes
1
2
3
Figure 1 - Engine test cell setup.
4
The engine-out emissions were measured with a Horiba MEXA 7170DEGR gas analyser system. This
5
equipment has a NDIR unit for CO and CO2 measurements, a paramagnetic based O2 measurement
6
unit, a Chemiluminescence unit for NOx measurements, and a Flame Ionization Detector model FIA-
7
720 for actual hydrocarbon measurement. As shown in some researches regarding ethanol SI
8
operation [33,35], a considerable part of the organic unburned emissions is constituted by aldehydes
9
and other oxygenated compounds. For more accurate estimation of total organic unburned emission
10
accounting for the oxygenated compounds when using a FID detector, a correction factor (𝑘𝐹𝐼𝐷 ) was
11
applied to the raw FID (𝐹𝐼𝐷𝑝𝑝𝑚 ) measurement depending on the ethanol volumetric content (𝑒) in
12
the fuel (which accounts only for fuels containing carbons) [36,37]. In this work the 𝐹𝐼𝐷𝑝𝑝𝑚 raw
13
measurement was corrected by the method presented in [36] using an updated factor of 0.68
14
presented in [37]. This way, an estimative of the oxygenated unburned compounds (excluding CO
15
and CO2) in the THC measurement was also considered. The corrected FID measurement (𝑇𝐻𝐶 𝑝𝑝𝑚 )
16
and the 𝑘𝐹𝐼𝐷 were calculated as
17
18
𝑇𝐻𝐶 𝑝𝑝𝑚 = 𝐹𝐼𝐷𝑝𝑝𝑚 ∗ 𝑘𝐹𝐼𝐷
(1)
19
20
𝑘𝐹𝐼𝐷 =
1
1−(1−0.64)(0.608𝑒 2 +0.092𝑒)
(2)
21
22
Finally, the indicated specific emissions were calculated following the procedures presented in [38]
23
on a wet basis. As wet ethanol contains high amount of water, its water content has been
24
introduced in the calculations of the dry-to-wet correction factor 𝑘𝑤 , adding the fuel water content
25
to the induced water due to air humidity. The specific NOx humidity correction factor has not been
26
used once the aim of such factor is to eliminate the charge humidity effect due to distinct day-to-day
1
and local-to-local temperature and humidity from the thermodynamic influence in the NOx
2
formation. The indicated specific gaseous emissions of each exhaust components evaluated (𝐼𝑆𝑔𝑎𝑠𝑖 )
3
were calculated as
4
5
𝐼𝑆𝑔𝑎𝑠𝑖 =
𝑢𝑖 [𝑥𝑖 ] 𝑘𝑤 𝑚̇𝑒𝑥ℎ
𝑃𝐼
(3)
6
7
where: 𝑢𝑖 and [𝑥𝑖 ] are the raw gas exhaust factor [38] and the concentration (in ppm) of the i gas in
8
the exhaust flow; 𝑘𝑤 is the dry to wet correction factor applied to CO and NOx; 𝑚̇𝑒𝑥ℎ is the exhaust
9
mass flow rate calculated as the sum of the instantaneous fuel and mass flow rates; 𝑃𝐼 is the
10
indicated power.
11
12
All measurements were acquired through a high speed data acquisition system, synchronized to the
13
crank position through a 720 pulse per revolution encoder. Engine operational parameters were
14
monitored and saved through in-house built Matlab based software. The results were averaged over
15
300 consecutive cycles.
16
17
The chosen valve train strategy for the present study was a positive valve overlap (PVO). This profile
18
was chosen to emulate the operation of conventional cam driven valve train. The exhaust valve
19
opens a few crank angle degrees (CAD) before the bottom dead centre (BDC) of the exhaust phase,
20
and closes after the top dead centre (TDC). The intake valve opens some degrees before the TDC and
21
closes after the BDC, as presented in Figure 2. As schematically represented in the figure, due to the
22
electro-pneumatic valve actuation, the valve lift profile is trapezoidal instead of the conventional
23
elliptical cam profile shape. The injection timing was kept constant at the centre of the intake stroke
24
at 450 CAD after the firing top dead centre. This injection timing has been chosen to promote good
25
charge homogeneity due to the side mounting injector positioning, between the intake and exhaust
26
valves. Injection quantity has been varied according to each fuel in order to result in the desired
27
load.
28
29
In this operation mode and during the valve overlapping period, there is backflow of burned gas into
30
the intake ports. During the start of the intake stroke, these gases return to the cylinder and some
31
exhaust gases from the exhaust ports can also be returned to the cylinder. As shown in [39], the
32
positive valve overlap can be used to obtain internal exhaust gas recycled (iEGR) fraction between
33
10 and 20%.
34
Inlet Valves
PVO Valve Profile
Exhaust Valves
Injection
Spark timing
0
TDC
180
360
540
720
TDC
1
2
3
Figure 2 – Positive valve overlap valve events, injection timing and spark timing representation.
4
The tests were conducted at the loads of 3.1±0.10 bar and 6.1±0.15 bar IMEP. The relative air/fuel
5
ratio (lambda λ) was varied in steps of approximately 0.1, starting from stoichiometric operation
6
until the leanest possible condition for most fuels when the cyclic variability, monitored by the
7
coefficient of variation of IMEP (COVimep) reached 5%. Lambda was monitored using an automotive
8
wide-band UEGO sensor. For unleaded gasoline, the leanest dilution tested was around λ = 1.3.
9
10
The load and air/fuel ratio were iteratively adjusted through throttle and injection quantity. To
11
enable comparison with previous work done in the same engine at similar loads to the 3.1 bar IMEP
12
load, the valve lift was set to 2 mm. As this low lift would excessively increase the pumping loop
13
work for the higher load, the lift was increased to 6mm for the 6.1 bar IMEP case. The spark timing
14
for each operating point was swept for the maximum indicated efficiency.
15
16
Commercial unleaded RON 95 UK standard Gasoline (herein named as GRON95) was used. According
17
to the UK fuel legislation, the maximum oxygen mass content in the fuel is 3%, which is the result of
18
approximately 8% of ethanol-in-gasoline volume fraction. Using the densities and LHV values of 44.0
19
and 26.9 MJ/kg for gasoline and E100 [39], respectively, the GRON95 fuel mixture’s LHV could be
20
calculated. Ethanol containing a maximum volumetric water content of 0.9% from Hayman Group
21
was used as the anhydrous ethanol (E100). The ethanol and water mixtures containing 5%, 10% and
22
20% of water volumetric content, herein named as E95W05, E90W10 and E80W20, respectively,
23
were prepared by splash-blending E100 with de-ionized water. A bulb alcoholmeter was used to
24
ensure the right water content. The LHV of wet ethanol was calculated according to the ethanol
25
mass fraction.
26
27
28
29
30
1
3. RESULTS
2
3.1.
Engine performance and combustion analysis
3
4
The load achieved in each operational point, spark timing, flame development angle (0-10% of mass
5
fraction burnt - MFB), combustion duration (10-90% MFB period), and COVimep are shown in Figure 3.
6
At both loads, the optimum spark timing near the stoichiometric operation (λ≈1.00) was advanced
7
as the water content in ethanol increased to achieve the best indicated efficiency. As the wet
8
ethanol’s water content acts as a diluent by reducing the charge temperature and decreasing its
9
reactivity, the initial flame development angle (FDA) increased. The higher in-cylinder temperature
10
for the 6.1 bar IMEP operation reduced the 0-10% MFB period as the temperature and pressure
11
were higher near stoichiometric operation for all tested fuels. Beyond λ≈1.2 the higher air mass
12
content seemed to affect the initial flame development in a higher degree and the initial FDA
13
increased for the higher load comparing to the lower load.
14
15
As water-in-fuel content increased, the combustion duration increased due to the higher charge
16
heat capacity and dilution effect. The same was valid for increased air dilution. The higher charge
17
heat capacity decreased the in-cylinder temperatures as the diluent absorbed the flame generated
18
heat and hampered the flame propagation process, further reducing the charge temperature. These
19
two combined effects resulted in lower flame propagation speeds with longer combustion durations,
20
which is in agreement with studies regarding laminar flames with higher water dilution [40–43]. The
21
increase in the combustion duration with the load was a result of higher in-cylinder charge
22
inhomogeneity and greater fueling rate. Gasoline low load combustion tended to be as fast as E100
23
combustion which also increased with the load.
24
25
The pressure and Heat Release Rate (HRR) trends at 3.1 bar presented in Figure 4 and Figure 5 show
26
that the highest peak of the heat release rate occurred for E100 and decreased almost linearly with
27
the increase in water content. At the same conditions, ethanol presents higher laminar flame speed
28
than gasoline. On the other hand, higher ethanol charge cooling effect decreased in-cylinder
29
temperatures which resultd in similar combustion periods for E100 and GRON95. As water content
30
increased, more expressive charge cooling effects were expected, reducing the in-cylinder
31
temperature and decreasing the heat release rate. The trend for the reduction in the peak HRR with
32
the increase in the water content (Figure 4) for the same air/fuel ratio was confirmed at both loads.
33
However, there was no clear correlation between the water content and the maximum in-cylinder
34
pressure shown in Figure 6. This fact occurred due to the distinct spark timing used for each fuel in
1
order to achieve the minimum spark advance for best torque (MBT), which affected the maximum
2
in-cylinder pressure.
3
7
IMEP (bar)
6
5
4
3
Spark Timing (CAD ATDCf)
-10
2
-17
-24
-31
-38
-45
30
FDA (CAD)
25
20
15
35
32
29
26
23
Combustion Duration (CAD)
10
4
COVimep (%)
20
3
2
1
0
0.90
4
Low Load
High Load
1.00
1.10
GRON95
GRON95
1.20
1.30
1.40
1.50
Lambda
E100
E95W05 E90W10
E100
E95W05 E90W10
1.60
E80W20
E80W20
5
Figure 3 – Operating conditions and combustion parameters. Filled symbols represent 3.1 bar IMEP
6
whilst hollow symbols represent 6.1 bar IMEP load.
7
24
21
20
18
15
15
10
12
9
5
6
0
3
-5
0
-20
-10
0
10
20
TDC
Crank Angle Degree
GRON95
E90W10
1
30
E100
E80W20
40
Heat Release Rate (J/CAD)
In-cylinder Pressure (bar)
25
E95W05
2
Figure 4 – Pressure and Heat Release Rate traces for 3.1 bar IMEP stoichiometric operation with
3
distinct fuels.
4
27
25
24
20
21
15
18
15
10
12
5
9
0
6
-5
3
-10
0
-30
-20
-10
0
10
20
30
40
Heat Release Rate (J/CAD)
In-cylinder Pressure (bar)
30
50
Crank Angle Degree
5
6
Lambda
1.0
1.2
1.4
1.5
Figure 5 – Pressure and Heat Release Rate traces of E100 at 3.1 bar IMEP.
7
8
There was a clear relationship between air dilution, HRR and in-cylinder peak pressure at both loads
9
and for all fuels. As shown in Figure 5 by increasing the λ for the same fuel, the peak HRR decreased
10
and combustion duration increased. The HRR peak was advanced towards TDC as the peak pressure
11
increased. The in-cylinder pressure increased due to the presence of more air in the beginning of the
12
compression phase, besides a higher ratio of specific heats which minimized heat losses. As water-
13
in-ethanol content increased, more advanced spark timings were necessary to account for slower
1
combustion as shown in Figure 3. Even then, the in-cylinder temperature (Figure 6) and combustion
2
efficiency (Figure 7) decreased. It has been shown [31] through OH Planar Laser Induced
3
Fluorescence (PLIF) images that increasing the water-in-ethanol content, for similar engine
4
operation, resulted in less flame wrinkle. It can be implied that the turbulent flame speed was
5
reduced in such situations.
6
7
The initial increase in air/fuel ratio led to a more homogeneous charge with better in-cylinder
8
conditions for the combustion process. Thus, for the lower load the COVimep decreased for initial
9
lambda increments. For higher air/fuel ratios (beyond lambda 1.2)?) the lower in-cylinder
10
temperature impaired the initial flame development process and increase the combustion cycle-to-
11
cycle variability, resulting in higher COVimep, until a point when misfire took place. A possible way to
12
reduce the COV would be to further advance the spark timing after the MBT is achieved. The result
13
would be a higher in-cylinder pressure and temperature during combustion which propitiates a more
14
stable combustion process in the penalty of lower engine indicated efficiency and higher NOx
15
emissions. At 6 bar IMEP there was higher in-cylinder inhomogeneity due to the higher injected
16
mass per cycle and the injector orientation (side mounted),. This fact resulted in higher combustion
17
variability with the increase in air/fuel ratio. The more pronounced COVimep of the 6.1 bar gasoline
18
cases seems to be the result of both poorer gasoline vaporization and in-cylinder mixture formation
19
process. The evidence was provided by emissions results discussed in a later section. Even then, the
20
COVimep values are between 2% and 3%, which can be considered stable operation. The optimization
21
of the injection timing for each load and fuel would possibly reduce the COVimep, but, would result in
22
distinct in-cylinder conditions and make the direct comparison of other parameters harder.
23
24
For all tested conditions the maximum pressure rise rate (PRRm) was kept below 3 bar/CA (Figure 6),
25
and there was no audible knocking noise. The PRRm seemed to be more directly related to the load
26
and spark timing than to the water-in-ethanol content. For all operating conditions, MBT could be
27
achieved at both loads with gasoline and all ethanol mixtures.
28
29
40
30
20
18
10
15
12
9
PRRm (dP/dCAD)
3
Max Pressure Angle (CAD ATDC)
Max Pressure (bar)
50
6
2
1
800
750
700
650
600
Max in-Cyl T (K)
1900
Exhaust Temperature (K)
0
550
1750
1600
1450
1300
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
Lambda
1
2
Low Load
High Load
GRON95
GRON95
E100
E100
E95W05
E95W05
E90W10
E90W10
E80W20
E80W20
Figure 6 – Pressure and temperature related parameters.
3
4
The reduction in exhaust temperature with the increase in water content and air dilution was
5
consistent with the in-cylinder pressure traces. The investigation of the pressure traces by the end of
6
combustion, for both Figure 4 and Figure 5, presented very similar pressure levels. It implied that the
7
temperatures in the expansion phase were lower for higher water content fuel mixtures (higher in-
1
cylinder mass), resulting in lower exhaust temperature. This trend agreed with the tendency shown
2
in [32]. The exhaust temperatures in the 3.1 bar at stoichiometric operation would be high enough
3
for the efficient use of a three-way catalytic converter. Previous studies [33] showed that this after-
4
treatment systems would be efficient enough to manage the engine-out emissions of wet ethanol.
5
The leaner the operation gets, the lower is the exhaust temperature and the use of three-way
6
catalysts is no longer possible. For the 3.1 bar IMEP load and conditions leaner than λ=1.3, exhaust
7
temperatures below 600 K would also impair the conversion efficiency of oxidation catalysts. Other
8
after-treatment systems as lean NOx trap would also need to be considered. The use of higher
9
internal and/or external residual gas recirculation (EGR) should also be considered for NOx
10
mitigation, but THC and CO would still be a challenge at lower loads.
11
12
The indicated efficiency presented in Figure 7 represented the relationship between the developed
13
work to the amount of energy delivered by the fuel per cycle. In the four-stroke SI throttled
14
operation the gas exchange and combustion efficiencies directly affect the indicated efficiency.
15
Throttled operation increases the pumping work during the intake stroke as a method to reduce the
16
amount of induced air, reducing the gas exchange efficiency. At low load as the water-in-ethanol
17
content increased, the charge cooling effect became more pronounced and the throttle needed to
18
be closed in order to keep the load, resulting in 10% difference between the low and high load
19
conditions. At higher load both throttling and charge cooling effect were less pronouced and the gas
20
exchange efficiency was virtually the same for all fuels. At the same load, the increase in the air/fuel
21
ratio resulted in reduced pumping loses and better thermodynamic characteristics (higher polytropic
22
coefficient), which increased the indicated efficiency. Considering only this effect, the lean SI
23
operation indicated efficiency would increase linearly with the increase in the gas exchange
24
efficiency, but the combustion effects on the indicated efficiency must also be considered.
25
26
The combustion efficiency was affected by the quantity of fuel injected per cycle, in-cylinder
27
temperature and homogeneity of the charge. As more fuel was injected (load increased), higher
28
charge stratification occurred and led to the formation of over- rich zones. It has been shown that
29
part of ethanol organic emissions is constituted by unburned ethanol [35]. While in DI gasoline
30
engines the fuel stratification may lead to soot formation and distinct unburnt hydrocarbon
31
components, DI ethanol operation produces mostly unburned ethanol emissions and aldehydes.
32
Therefore, the increased stratification at higher load led to lower combustion efficiency near
33
stoichiometric operation. As the mixture became globally leaner, the combustion efficiency
34
increased and reached its maximum around λ≈1.2 (for the high load operation), whilst the best
1
combustion efficiency for low load occurred at λ≈1.1. The water content decreased the
2
combustion efficiency not only due to the higher in-cylinder cooling effect, but also due to the
3
diluting effect. The combustion efficiency seemed to be more sensitive to water content at higher
4
loads, where the mass of water injected per cycle was higher and impaired the whole fuel
5
vaporization process. On the other hand, the initial increase in the air/fuel ratio raised the
6
combustion efficiency due to higher oxygen availability. For further increases in air/fuel ratio, the
7
average in-cylinder temperature during combustion decreased quickly, leading to partial oxidation.
8
Although gasoline combustion efficiency was lower, the trends were exactly the same as those of
9
alcohol fuels.
10
35
32
29
100
26
95
90
85
Combustion Efficiency (%)
97.5
80
Gas Exchange Eff (%)
Indicated Efficiency (%)
38
95.0
92.5
90.0
87.5
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60
Lambda
11
Low Load
High Load
GRON95
GRON95
E100
E100
E95W05
E95W05
E90W10
E90W10
E80W20
E80W20
12
13
Figure 7 – Efficiency related parameters.
14
The combined effect of gas exchange and combustion efficiency explained the initial fast increase
15
followed by slower change in the indicated efficiency as air/fuel ratios became higher. Considerable
16
increase in indicated efficiency could be reached when using lean combustion, although the water
17
addition decreased the indicated efficiency. Even then, at some operation conditions, the highest
1
water content wet ethanol operation reached the gasoline operation efficiency at the same air/fuel
2
ratio.
3
4
3.2.
Gaseous emissions
5
6
Figure 8 presents the engine-out emissions. The CO emissions were highly correlated to the air/fuel
7
ratio and in-cylinder homogeneity. During the tests at the low load it was noticed that from lambda
8
1.01 to lambda 0.99 the CO emission tended to increase in more than five times. For this reason, all
9
the tests were conducted approaching the desired stoichiometry from the leaner side (enriching the
10
mixture from 1.02 to 1.00). As the charge became leaner, CO emissions reduced slightly and then
11
remained almost constant. For the near stoichiometric operation, as water-in-ethanol content
12
increased, a higher fraction of fuel was left unburned or partially burnt and became organic
13
unburned compounds (treated here as THC) instead of CO. As the side mounted injector resulted in
14
a spray in the middle of the anti-tumble large scale motion (characteristic of the engine design), the
15
air-fuel mixing process worsened as the injected mass increased. Thus, as the single injection timing
16
was kept constant for both loads, higher in-cylinder inhomogeneity for the high load cases incurred
17
in higher CO emissions. It seemed that the gasoline inhomogeneity was higher than that in the case
18
of ethanol fuel mixtures. This fact led to less stable combustion resulting in a slightly higher COVimep
19
and reduced engine indicated efficiency.
20
21
THC emissions were believed to be caused by flame quenching, fuel impingement and crevices. Due
22
to the position of the fuel injector and the direction of the spray, some impingement was expected,
23
which was worsened by the extended injection duration with higher water content fuel. Both charge
24
cooling and increased air dilution reduced the combustion temperature and lowered post-flame THC
25
oxidation. At 6 bar IMEP, higher in-cylinder temperatures increased the conversion rates, resulting in
26
lower THC emissions compared to the low load case.
27
28
As the mixture became leaner, NOx emission at low and high loads exhibited opposite trends. NOx
29
formation was mainly dependent on temperature and oxygen availability. The expected trend of
30
reduction in NOx emissions with the increase of wet ethanol water content, for the same air/fuel
31
ratio, would happen if the maximum temperatures could be reduced. As the spark timing had to be
32
advanced for increased water-in-ethanol content fuels, similar in-cylinder peak pressure and
33
temperatures were produced for both E95W05 and E90W10. For E80W20 the peak pressure
1
reduction was more evident and the in-cylinder temperatures were much lower, which explained
2
the lower NOx formation.
90
80
70
60
50
40
30
20
10
0
45
40
35
30
25
20
15
10
5
0
12.0
ISTHC (g/kWh)
ISCO (g/kWh)
3
ISNOx (g/kWh)
10.0
8.0
6.0
4.0
2.0
0.0
0.90
Low Load
High Load
4
1.00
1.10
GRON95
GRON95
1.20
1.30
Lambda
E100
E100
1.40
E95W05
E95W05
1.50
E90W10
E90W10
1.60
E80W20
E80W20
5
6
Figure 8 – Engine-out emissions.
7
The increase in air/fuel ratio reduced the in-cylinder average temperature, which should have
8
reduced the NOx formation for a homogeneous mixture as occurring in the low load cases. The
9
higher stratification at higher loads increased the temperature in some flame reaction zones due to
10
stoichiometric to slightly rich mixture spots. Even with an average lower in-cylinder temperature, the
11
NOx formation increased due to the higher flame temperature achieved in these zones [44].
12
Gasoline NOx emissions were relatively higher due to the increased in-cylinder temperature resulted
13
from faster combustion.
14
15
16
17
1
4. COST-BENEFIT ANALYSIS
2
3
The practical use of wet ethanol in SI engines is directly connected to the ethanol price. Although
4
much has been said about the price reduction of the wet ethanol production process compared to
5
anhydrous (E100) or hydrous (E95W05) ethanol [14,20–23,27,28,33], an energy usage and
6
conversion based comparison has not been provided elsewhere. The data presented in [14] has been
7
considered regarding the energy expense for the production of one litre of ethanol from corn
8
(considering co-products). It has been assumed that around 68% of the total energy used during the
9
water removal processes is used for the distillation process to reach around 95% of ethanol-in-water
10
volumetric content. The 32% left is used in the dehydration process to reach 99.5% of ethanol-in-
11
water volumetric content. By using the distillation energy expense presented in [13], a normalized
12
water removal energy expense (NWREE) trend and a normalized energy expense in the production
13
of wet ethanol (NEEPWE) curve to produce one litre of different water-in-ethanol mixtures could be
14
obtained as shown in Figure 9. This figure presents the energy requirement for water removal during
15
the wet ethanol production, the normalized energy production costs of corn ethanol, the engine fuel
16
consumption, and energy based operational cost evaluation.
17
18
It is known that depending on the crop used and the possible co-products obtained, as well as the
19
ethanol production process, the energy fraction of the total production cost regarding the water
20
removal process varies and affects the final fuel cost. As the net-energy balance for ethanol is
21
positive for most of the production scenarios [17], the use of the NEEPWE instead of the absolute
22
monetary cost better illustrates the influence of water content in total energy expense for a more
23
general evaluation. The higher the energy fraction for the water removal process is (in the total
24
energy expensed during the ethanol production), the lower is the wet ethanol fuel cost and the
25
higher is the impact on the final engine operational costs.
26
27
When multiplying the NEEPWE by fuel consumption, for both loads tested and distinct fuel
28
compositions, the result is the normalized energy engine operational cost (NOpC), calculated as:
29
30
𝑚̇
𝑁𝑂𝑝𝐶 = ( 𝜌 𝑓) . NEEPWE
(4)
𝑓
31
32
where: 𝑚̇𝑓 is the engine mass fuel consumption, directly provided by the fuel flow meter; 𝜌𝑓 is fuel
33
density (ethanol-water mixtures) at 25 °C, calculated according to [45]. The ratio ( 𝜌 𝑓) is the
𝑚̇
𝑓
1
volumetric fuel consumption. The NOpC relates the normalized energy expense for wet ethanol
2
production and the engine volumetric fuel consumption based on the engine fuel conversion
3
efficiency. The lower the NOpC (looking at each load individually), the lower is the engine
4
operational cost on an energy bases and the lower is the real monetary operational cost reduction.
5
As shown in Figure 9, the very high energy expense for the production of anhydrous ethanol makes
6
the operational cost of such fuel the highest amongst the others. When using only the distillation to
7
reach around 95% of ethanol-in-water, the operational cost dropped considerably. The fitted curves
8
show that the best energy based operational cost would be achieved for water-in-ethanol mixtures
9
with water content between 85 and 90%. Although the production cost of one litre of E80W20
10
would be the cheapest amongst the fuels tested, the volumetric fuel consumption increased as a
11
consequence of the higher water content and lower engine efficiency. Comparable energy based
12
operational costs to the anhydrous ethanol can be expected for mixtures with more than 25% of
13
water in volume by analysing the extrapolated fitted curves.
14
15
Comparing the stoichiometric operational cost of anhydrous ethanol to the stoichiometric and lean
16
operational costs of wet ethanol (the average costs of both loads), the cost reduction in the
17
operational cost on an energy bases could be accessed. The highest operational cost reduction
18
occurred for E90W10 at lean conditions, according to the evaluated scenarios. The trend shows that
19
the lowest operational cost was achieved for wet ethanol containing around 12.5% of water content
20
in volume. It is also interesting to access the reduction in the operational cost of wet ethanol to
21
hydrous ethanol (E95W05), which is commercially available in some countries as Brazil. In this case
22
the operational cost reduction was lower than that compared to hydrous ethanol, but it was still
23
significant. Table 2 summarizes the operational cost reduction of wet ethanol compositions
24
compared to anhydrous (E100) and hydrous (E95W05) ethanol.
25
26
Table 2
Operational Cost Reduction
Compared Fuels
E90W10
E80W20
λ=1.0
λ=1.3
λ=1.0
λ=1.3
E100*
25%
31%
19%
25%
E95W05*
12%
19%
5%
11%
* λ=1.0
27
NWREE
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Distilation+
Dehydration++
1.0
0.8
0.7
NEEPWE
0.9
0.6
0.5
1.7
1.5
1.3
1.1
0.9
0.7
1.6
0.5
1.4
1.2
1
0.8
0.6
NOpC (l/h . MJ/MJ)
Fuel Consumption (l/h)
1.9
0.4
75
80
85
90
95
100
Ethanol-in-water volumetric content (%)
IMEP: 6.1 bar
3.1 bar
1
2
3
4
5
6
7
8
9
10
11
12
13
14
λ=1.0
λ=1.0
λ=1.3
λ=1.3
Figure 9 – Wet ethanol production and usage cost as function of the ethanol-in-water volumetric
content. +[13]; ++[14].
1
5. CONCLUSIONS
2
3
Tests were conducted in a direct injection single cylinder spark ignition engine with anhydrous
4
ethanol (E100), hydrous ethanol (E95W05), two wet ethanol blends (E90W10 and E80W20), and
5
unleaded UK gasoline (GRON95). Two part load operating conditions were tested, 3.1 and 6.1 bar
6
IMEP. The main findings regarding the engine operation can be summarized as follows:
7
8
9
- Stable engine operation could be achieved for lean mixture of λ=1.3 for all tested fuels at both
low and high loads;
10
- Flame development angle and combustion duration increased as the water-in-ethanol volumetric
11
content increased. This was a consequence of lower in-cylinder temperatures due to water
12
dilution, which lead to decreased heat release rate;
13
- In order to achieve maximum indicated efficiency through MBT operation, the location of the
14
peak pressure tended to advance towards TDC with the increase of the water-in-ethanol content.
15
The maximum in-cylinder pressure increased with the load but there was no clear trend between
16
water content and maximum pressure;
17
- The indicated efficiency increased for lean operation due to lower pumping loses and better
18
mixture characteristics. Combustion efficiency was initially improved by increasing the air/fuel
19
ratio until λ=1.2. For leaner mixtures, the lower in-cylinder temperature increased THC and CO
20
emissions and decreased the combustion efficiency.
21
- THC and CO engine-out emissions trend for all ethanol fuels were similar to gasoline operation. In
22
general as the water-in-ethanol content increased, CO engine-out emissions dropped whilst THC
23
increased. Gasoline THC emission was comparable to the anhydrous ethanol, whilst CO emissions
24
were the highest, attributed to the injection timing and poor mixing process;
25
- Low load NOx emissions with E80W20 were almost half of all other ethanol fuels for all air/fuel
26
ratios. E95W05 and E90W10 presented similar NOx emissions to E100. Gasoline presented the
27
highest NOx emissions amongst all cases tested (almost three times higher than the E80W20);
28
- Improvements in the fuel injection system are required to improve the SI engine efficiency and
29
combustion process. Port fuel injection should also be evaluated in future studies.
30
31
Regarding the engine operating parameters and engine-out gaseous emissions, it could be
32
concluded that the water-in-ethanol content diluent effect was more pronounced at the lowest load
33
than at the highest load. Indicated and combustion efficiencies were proportionally more impaired
34
and the effects on NOx emissions were more pronounced. For a real engine application it can be
1
expected that its operation will be slightly less efficient for wet ethanol with up to 10% volumetric
2
content than anhydrous ethanol, but similar to the gasoline operation regarding combustion
3
processes and CO and THC emissions.
4
5
Comparing E95W05 with E90W10 the impact on real engine operating conditions would be minor,
6
but the expected operational cost reduction would be in the order of 10%. Regarding their
7
application in flexible fuel cars, Gasoline and E90W10 miscibility problems would occur at low
8
temperatures (below 10 °C) and some gasoline additive to support higher water content in the
9
ternary mixture (gasoline-ethanol-water) are required. E80W20 would be practical only in dedicated
10
ethanol engines. Due to the lower LHV of water-ethanol mixtures, volumetric fuel consumption
11
would increase compared to anhydrous ethanol. Oil contamination and engine corrosion need to be
12
further investigated.
13
14
The comparison of NOpC showed that the most profitable scenario is the lean operation with
15
E90W10. It provided a reduction on the engine operational cost around 31% and 19% with E100 and
16
E95W05, respectively, when compared to the conventional stoichiometric operation. Although it has
17
been already shown in the literature that a three-way catalyst is effective to manage wet ethanol
18
stoichiometric emissions [33], after treatment systems for lean-burn operation is still a costly
19
challenge.
20
21
Finally, the use of wet ethanol reduced the energy requirement during the whole ethanol life cycle.
22
The saved energy in the production process can help to further reduce fossil fuel dependency whilst
23
mitigating greenhouse gas emissions.
24
25
26
6. ACKNOWLEDGEMENTS
27
28
The authors would like to acknowledge the Brazilian Council for Scientific and Technological
29
Development (CNPq – Brasil) for supporting the PhD studies of Mr. Lanzanova and Mr. Dalla Nora at
30
Brunel University London.
31
32
33
34
1
7. REFERENCES
2
3
[1]
United Nations. Adoption of the Paris Agreement. Framew. Conv. Clim. Chang., 2015, p. 32.
4
5
[2]
(EREC) EREC. Renewable Energy in Europe: Markets, Trends, and Technologies. Second Ed. London:
Earthscan Ltd; 2010.
6
7
8
[3]
Maroun MR, La Rovere EL. Ethanol and food production by family smallholdings in rural Brazil:
Economic and socio-environmental analysis of micro distilleries in the State of Rio Grande do Sul.
Biomass and Bioenergy 2014;63:140–55. doi:10.1016/j.biombioe.2014.02.023.
9
[4]
The European Union Parliament. L 140/88 2009:88–113.
10
11
12
[5]
Walter A, Dolzan P, Quilodrán O, de Oliveira JG, da Silva C, Piacente F, et al. Sustainability assessment
of bio-ethanol production in Brazil considering land use change, GHG emissions and socio-economic
aspects. Energy Policy 2011;39:5703–16. doi:10.1016/j.enpol.2010.07.043.
13
14
[6]
Elbehri A, Segerstedt A, Liu P. Biofuels and the sustainability challenge: A global assessment of
sustainability issues, trends and policies for biofuels and related feedstocks. Rome: 2013.
15
16
[7]
Jeuland N, Montagne X, Gautrot X. Potentiality of Ethanol As a Fuel for Dedicated Engine. Oil Gas Sci
Technol 2004;59:559–70. doi:10.2516/ogst:2004040.
17
18
[8]
PANA C, NEGURESCU N, POPA, M. G, CERNAT A, SOARE D. ASPECTS OF THE USE OF ETHANOL IN SPARK
IGNITION ENGINE. JSAE Tech Pap 2007. doi:10.4271/2007-01-2040.
19
20
21
[9]
Huang Y, Hong G, Huang R. Investigation to charge cooling effect and combustion characteristics of
ethanol direct injection in a gasoline port injection engine. Appl Energy 2015;160:244–54.
doi:10.1016/j.apenergy.2015.09.059.
22
23
[10]
Shapouri H, Duffield J, Wang M. The energy balance of corn ethanol revisited. Trans Soc Agric Eng
2003;46:959–68.
24
25
[11]
Dai D, Hu Z, Pu G, Li H, Wang C. Energy efficiency and potentials of cassava fuel ethanol in Guangxi
region of China. Energy Convers Manag 2006;47:1686–99. doi:10.1016/j.enconman.2005.10.019.
26
27
[12]
Salla DA, Furlaneto F de PB, Cabello C, Kanthack RAD. Avaliação energética da produção de etanol
utilizando como matéria-prima a cana-de-açúcar. Ciência Rural 2009;39:2516–20.
28
29
[13]
Ladisch MR, Dyck K. Dehydration of ethanol: new approach gives positive energy balance. Science
1979;205:898–900. doi:10.1126/science.205.4409.898.
30
31
[14]
Flowers DL, Aceves SM, Frias JM. Improving Ethanol Life Cycle Energy Efficiency by Direct Utilization of
Wet Ethanol in HCCI Engines. SAE Tech. Pap., 2007, p. 1070–8. doi:10.4271/2007-01-1867.
32
33
34
[15]
López-Plaza EL, Hernández S, Barroso-Muñoz FO, Segovia-Hernández JG, Aceves SM, Martínez-Frías J,
et al. Experimental and Theoretical Study of the Energy Savings from Wet Ethanol Production and
Utilization. Energy Technol 2014;2:440–5. doi:10.1002/ente.201300180.
35
36
37
[16]
Chen W-C, Sheng C-T, Liu Y-C, Chen W-J, Huang W-L, Chang S-H, et al. Optimizing the efficiency of
anhydrous ethanol purification via regenerable molecular sieve. Appl Energy 2014;135:483–9.
doi:10.1016/j.apenergy.2014.08.112.
1
2
[17]
Lorenz D, Morris D. How much energy does it take to make a gallon of ethanol? Institute for Local SelfReliance; 1995.
3
4
5
[18]
Salla DA, Furlaneto F de PB, Cabello C, Kanthack RAD. Estudo energético da produção de
biocombustível a partir do milho Energetic study of ethanol production from the corn crops. Ciência
Rural 2010;40:2017–22.
6
7
[19]
Nguyen TLT, Gheewala SH, Garivait S. Energy balance and GHG-abatement cost of cassava utilization
for fuel ethanol in Thailand. Energy Policy 2007;35:4585–96. doi:10.1016/j.enpol.2007.03.012.
8
9
[20]
Cordon D, Clarke E, Beyerlein S, Steciak J, Cherry M. Catalytic Igniter to Support Combustion of
Ethanol-Water/Air Mixtures in Internal Combustion Engines 2002. doi:10.4271/2002-01-2863.
10
11
[21]
Olberding J, Beyerlein DCS, Steciak J, Cherry M. Dynamometer Testing of an Ethanol-Water Fueled
Transit Van. SAE Tech. Pap., 2005. doi:10.4271/2005-01-3706.
12
13
[22]
Mack JH, Aceves SM, Dibble RW. Demonstrating direct use of wet ethanol in a homogeneous charge
compression ignition (HCCI) engine. Energy 2009;34:782–7. doi:10.1016/j.energy.2009.02.010.
14
15
16
[23]
Saxena S, Schneider S, Aceves S, Dibble R. Wet ethanol in HCCI engines with exhaust heat recovery to
improve
the
energy
balance
of
ethanol
fuels.
Appl
Energy
2012;98:448–57.
doi:10.1016/j.apenergy.2012.04.007.
17
18
19
[24]
Saxena S, Vuilleumier D, Kozarac D, Krieck M, Dibble R, Aceves S. Optimal operating conditions for wet
ethanol in a HCCI engine using exhaust gas heat recovery. Appl Energy 2014;116:269–77.
doi:10.1016/j.apenergy.2013.11.033.
20
21
22
[25]
Megaritis a., Yap D, Wyszynski ML. Effect of water blending on bioethanol HCCI combustion with
forced
induction
and
residual
gas
trapping.
Energy
2007;32:2396–400.
doi:10.1016/j.energy.2007.05.010.
23
24
25
[26]
Megaritis a., Yap D, Wyszynski ML. Effect of inlet valve timing and water blending on bioethanol HCCI
combustion using forced induction and residual gas trapping. Fuel 2008;87:732–9.
doi:10.1016/j.fuel.2007.05.007.
26
27
28
[27]
Lanzanova TDM, Vielmo HA, Sari RL, Dornelles HM, Tatsch GA, Martins MES, et al. Performance
Analysis of a Spark Ignited Engine Running on Different Water-in-Ethanol Mixtures. SAE Tech. Pap.,
2013. doi:10.4271/2013-36-0202.
29
30
[28]
Martins M, Lanzanova T, Sari R. Low Cost Wet Ethanol for Spark-Ignited Engines: Further
Investigations. SAE Int J Fuels Lubr 2015;8:2015–01 – 0954. doi:10.4271/2015-01-0954.
31
32
33
[29]
Dempsey AB, Das Adhikary B, Viswanathan S, Reitz RD. Reactivity Controlled Compression Ignition
Using Premixed Hydrated Ethanol and Direct Injection Diesel. Proc. ASME. 44427; ASME 2011 Intern.
Combust. Engine Div. Fall Tech. Conf., vol. 134, 2012, p. 963–75. doi:10.1115/1.4006703.
34
35
36
[30]
Fang W, Kittelson DB, Northrop WF. An Experimental Investigation of Reactivity-Controlled
Compression Ignition Combustion in a Single-Cylinder Diesel Engine Using Hydrous Ethanol. Proc ASME
2013 Intern Combust Engine Div Fall Tech Conf 2013;137:1–9. doi:10.1115/ICEF2013-19209.
37
38
39
[31]
Augoye A, Aleiferis P. Characterization of Flame Development with Hydrous and Anhydrous Ethanol
Fuels in a Spark-Ignition Engine with Direct Injection and Port Injection Systems. SAE Tech. Pap., 2014.
doi:10.4271/2014-01-2623.
1
2
[32]
Brewster S, Railton D, Maisey M, Frew R. The Effect of E100 Water Content on High Load Performance
of a Spray Guide Direct Injection Boosted Engine. SAE Tech. Pap., 2007. doi:10.4271/2007-01-2648.
3
4
5
[33]
Munsin R, Laoonual Y, Jugjai S, Imai Y. An experimental study on performance and emissions of a small
SI engine generator set fuelled by hydrous ethanol with high water contents up to 40%. Fuel
2013;106:586–92. doi:10.1016/j.fuel.2012.12.079.
6
7
8
[34]
Zhang Y, Zhao H. Investigation of combustion, performance and emission characteristics of 2-stroke
and 4-stroke spark ignition and CAI/HCCI operations in a DI gasoline. Appl Energy 2014;130:244–55.
doi:10.1016/j.apenergy.2014.05.036.
9
10
[35]
Melo T de, Brito M de, Machado G, Paiva C. Procedure for Uncertainty of Measurement Determination
of Spark Ignition Engine Emission Tests. SAE Tech Pap 2012.
11
12
13
[36]
Kar K, Tharp R, Radovanovic M, Dimou I, Cheng WK. Organic gas emissions from a stoichiometric direct
injection spark ignition engine operating on ethanol/gasoline blends. Int J Engine Res 2010;11:499–
513. doi:10.1243/14680874JER610.
14
15
16
[37]
Wallner T. Correlation Between Speciated Hydrocarbon Emissions and Flame Ionization Detector
Response for Gasoline/Alcohol Blends. J Eng Gas Turbines Power 2011;133:082801.
doi:10.1115/1.4002893.
17
18
[38]
Economic Commission for Europe of the United Nations. Regulation No 49 of the Economic
Commission for Europe of the United Nations (UN/ECE). Off J Eur Union 2013:1–390.
19
[39]
Heywood JB. Internal Combustion Engine Fundamentals. vol. 21. 1st ed. McGraw-Hil; 1988.
20
21
22
[40]
23
24
[41]
Parag S, Raghavan V. Experimental investigation of burning rates of pure ethanol and ethanol blended
fuels. Combust Flame 2009;156:997–1005. doi:10.1016/j.combustflame.2008.10.011.
25
26
27
[42]
Mazas a. N, Fiorina B, Lacoste D a., Schuller T. Effects of water vapor addition on the laminar burning
velocity
of
oxygen-enriched
methane
flames.
Combust
Flame
2011;158:2428–40.
doi:10.1016/j.combustflame.2011.05.014.
28
29
30
[43]
Rahman KM, Kawahara N, Tsuboi K, Tomita E. Combustion characteristics of wet ethanol ignited using
a
focused
Q-switched
Nd:YAG
nanosecond
laser.
Fuel
2015;165:331–40.
doi:10.1016/j.fuel.2015.10.067.
31
32
[44]
Zhao F, Lai M-C, Harrington D. Automotive spark-ignited direct-injection gasoline engines. Prog Energy
Combust Sci 1999;25:437–562. doi:10.1016/S0360-1285(99)00004-0.
33
34
35
[45]
Khattab IS, Bandarkar F, Fakhree MAA, Jouyban A. Density, viscosity, and surface tension of
water+ethanol mixtures from 293 to 323K. Korean J Chem Eng 2012;29:812–7. doi:10.1007/s11814011-0239-6.
36
37
38
a constant volume combustion vessel. Combust Flame 1982;46:17–28. doi:10.1016/00102180(82)90004-9.
1
8. NOMENCLATURE
2
ICE
BDC
CAD
CAI
CO
CO2
COVimep
DI
𝑒
E100
Eind
EVC
EVO
ExxWyy
FDA
FID
𝐹𝐼𝐷𝑝𝑝𝑚
FVVT
GHG
GRON95
HCCI
HRR
IMEP
in-Cyl T
ISCO
𝐼𝑆𝑔𝑎𝑠𝑖
ISNOx
ISTHC
IVC
IVO
𝑘𝐹𝐼𝐷
𝑘𝑤
LHV
MBT
MFB
NOpC
NEEPWE
NOx
NVO
NWREE
OH
𝑃𝐼
Internal combustion engine
Bottom dead centre
Crank angle degree
Controlled auto ignition
Carbon monoxide
Carbon dioxide
Coefficient of variation of IMEP
Direct Injection
Ethanol volumetric content
Anhydrous Ethanol
Expended Energy in fuel production
Exhaust Valve Closure
Exhaust Valve Opening
Mixture of xx% ethanol and yy% Water (v/v)
Flame Development Angle
Flame Ionization Detector
Raw FID measurement
Fully Variable Valve Train
Green House Gases
95 RON United Kingdom standard unleaded
gasoline
Homogeneous Charge Compression Ignition
Heat Release Rate
Indicated Mean Effective Pressure
In-cylinder Temperature
Indicated Specific CO emission
Indicated Specific gas emission
Indicated Specific NOx emission
Indicated Specific THC emission
Inlet Valve Closure
Inlet Valve Opening
FID correction factor
Dry-to-wet correction factor
Lower Heating Value
Minimum spark advance for best torque
Mass Fraction Burned
Normalized Energy Engine Operational Cost
Normalized Energy Expense in the Production of
Wet Ethanol
Nitrogen Oxides
Negative valve overlap
Normalized water removal energy expense
Hydroxyl
Indicated power
PLIF
PRRm
PVO
𝑞̇ 𝑒𝑥ℎ
RCCI
RON
rpm
SI
TDC
TDCf
THC
𝑇𝐻𝐶 𝑝𝑝𝑚
UEGO
𝑢𝑖
v/v
λ
[𝑥𝑖 ]
1
Planar Laser Fluorescence
Maximum Pressure Rise Rate
Positive Valve Overlap
Exhaust mass flow rate
Reactivity-controlled compression ignition
Research Octane Number
Revolution per minute
Spark Ignition
Top dead centre
Firing top dead centre
Total Hydrocarbon (used in this work as a total
unburned organic emission estimative)
Corrected FID measurement
Universal Exhaust Gas Oxygen
Raw gas exhaust factor
volume/volume
Excess of air factor – Lambda
gas concentration in ppm