Internal Combustion Engine

(ME60111)

Pabitra Badhuk
Assistant Professor
Department of Mechanical Engineering
IIT Kharagpur
Module 4 – Emissions
Overview of emissions and
regulations
Emissions categorization
• Gaseous Emissions (NOx, CO, UHC, CO2)
• Particulate Emissions (soot)

• Regulated emissions (UHC, CO, NOx, CO2, PM)

• Unregulated emissions (Formaldehyde, Aldehydes, SOx, Methane, PAHs)
Modified Indian Drive Cycle (MIDC)
Urban phase Extra-urban phase

The Indian Drive Cycle (IDC) is a standardized driving cycle developed to represent typical driving
conditions in India. It is mainly used for emission testing and fuel consumption measurement of
vehicles under controlled laboratory conditions. It provides a reproducible framework to test and
compare vehicle emissions and fuel economy.
Emission norms – light duty vehicles (Vehicle weight < 3500 kg)

Until Euro 5, there

were no PM limits
for petrol cars.

Bharat Stage (BS) 1 developed on Euro 1, was implemented nationwide in the year 2000.
Bharat Stage (BS) 2 developed on Euro 2, was implemented nationwide in the year 2005.
Bharat Stage (BS) 3 developed on Euro 3, was implemented nationwide in the year 2010.
Bharat Stage (BS) 4 developed on Euro 4, was implemented nationwide in the year 2017.
Bharat Stage (BS) 6 developed on Euro 6, was implemented nationwide in the year 2020.
Emission norms – heavy-duty vehicles (Vehicle weight > 3500 kg)

Bharat Stage (BS) 1 developed on Euro 1, was implemented nationwide in the year 2000.
Bharat Stage (BS) 2 developed on Euro 2, was implemented nationwide in the year 2005.
Bharat Stage (BS) 3 developed on Euro 3, was implemented nationwide in the year 2010.
Bharat Stage (BS) 4 developed on Euro 4, was implemented nationwide in the year 2017.
Bharat Stage (BS) 6 developed on Euro 6, was implemented nationwide in the year 2020.
Vehicular exhaust emissions
SI Engine exhaust gases before post-processing contain,
Oxides of nitrogen (collectively known as NOX) (of order: 500 to 1000 ppm or 20 g/kg fuel)
Carbon monoxide (CO) (of order: 1 per cent or 150 g/kg fuel)
Organic compounds which are unburned or partially Unburned hydrocarbons (HC) (of order:
2000 ppm or 15 g/kg fuel ).

Three-way catalytic converters reduce these emissions by over 90% before they are exhausted into the
ambient air.
CI Engine Exhaust before post-processing contain,
NOx emissions comparable to SI Engines.
Unburnt Hydrocarbons lower by a factor of 5 than typical SI engine levels.
Particulate emissions; between about 0.2 and 0.5 percent of the fuel mas is emitted as
small particulates (~ 10 to 500 nm diameter).

Diesel engines are not a significant source of carbon monoxide.

Effect of equivalence ratio on SI engine emissions
The figure shows how NO, CO, and HC exhaust emissions vary
with the equivalence ratio. The spark-ignition engine has
historically been operated close to stoichiometric. The figure
shows that leaner mixtures give lower CO and HC emissions
until the combustion quality becomes poor (and eventually
misfire occurs), when HC emissions rise sharply and engine
operation becomes erratic. However, NO emissions peak
about 10% lean of stoichiometric.

Exhaust gas recirculation (EGR) is used with stoichiometric

mixtures in many engine control systems. Note that the highest
power levels are obtained from the engine with slightly rich-of-
stoichiometric mixtures and no recycled exhaust to dilute the
incoming charge. As we will see, several emission control
techniques are required to reduce engine-out emissions of all
three pollutants across all engine operating modes.
Effect of equivalence ratio on CI engine emissions

The figure shows a qualitative picture of HC, CO and NOx

emissions with respect to equivalence ratio, φ, for a four-stroke
DI Diesel engine. As can be seen, HC will decrease slightly
with an increase in φ due to higher cylinder temperatures,
making it easier to burn up any over-mixed (very lean) or
under-mixed (rich) fuel-air mixture. At high loads, however,
HC may increase again if the amount of fuel in regions is too
rich to burn during the primary combustion process. CO
emissions will be very low at all equivalence ratios since excess
air is always available. NOx emission will steadily increase as
φ increases due to an increasing fraction of cylinder contents
being burnt gases close to stoichiometric during combustion,
and also due to higher peak temperatures and pressures.
Emission norms compliance enabling technologies
Petrol cars
Carburettor → MPFI → GDI
Variable valve timing
Exhaust gas recirculation (EGR) for NOx reduction
Three-way catalytic converter
Gasoline particulate filters (GPF) After-treatment devices

Diesel cars
Mechanical pump injection → common rail injection system
Indirect injection → Direct injection
Exhaust gas recirculation (EGR)
Split injection strategy
Diesel particulate filters (DPF) for soot reduction
Diesel oxidation catalyst (DOC)
Selective catalytic reduction (SCR) for NOx reduction After-treatment devices
Lean NOx trap (LNT)
NOx formation and control
NO and NO2 formation mechanisms
Kinetics of NO formation
𝑁2 + 𝑂 = 𝑁𝑂 + 𝑁 This mechanism of NO formation kicks in
𝑁 + 𝑂2 = 𝑁𝑂 + 𝑂 Thermal NO mechanism when gas temperature exceeds 1800 K.
Zel’dovich mechanism Nitrogen oxidation reactions are slower than
𝑁 + 𝑂𝐻 = 𝑁𝑂 + 𝐻
fuel oxidation reactions.

Kinetics of NO2 formation is given below. Conversion of NO2 to NO occurs as well. NO2/NO ratio is
negligibly small in SI engines. In diesel engines, NO2 can constitute up to 10-30% of the total NOx
concentration. NO2/NO ratio peaks in diesel engines in lighter loads, because the mixture is globally
leaner.
𝑁𝑂 + 𝐻𝑂2 = 𝑁𝑂2 + 𝑂𝐻
𝑁𝑂2 + 𝑂 = 𝑁𝑂 + 𝑂2
NO and NO2 formation mechanisms
The figure shows examples of NO and NO2 emissions data from a spark ignition and a diesel engine.
The maximum value for the ratio (NO2/NO) for the SI engine is 2 percent, at a relative air-fuel ratio
of about 1.2. For the diesel this ratio is higher, and is highest at light load, when the engine is
operating very lean. The ratio also depends on engine speed.

(a) NO and NO2

concentrations in SI
engine-out exhaust as
function of air/fuel
ratio, 1500 rev/min,
wide-open throttle;
(b) NO2 as percent of
total NOx in diesel
exhaust as function of
load and speed.
Engine variables controlling NO emission in SI engines
Equivalence ratio: Maximum burned gas temperatures occur at 𝜙 ≈ 1.1; however, at this relative
air/fuel ratio oxygen concentrations are low. As the mixture is further enriched, burned gas
temperatures fall. As the mixture is leaned out, increasing oxygen concentration initially offsets the
falling gas temperatures and NO emissions peak at 𝜙 ≈ 0.9.

NO mole fraction

𝜙 ≈ 0.9 Equivalence ratio

Exhaust gas recirculation (EGR)
In addition to controlling the air-fuel ratio, some engines
may also employ an exhaust gas recirculation (EGR)
system. The adjacent image shows a traditional EGR
system. A small fraction of the exhaust gases are directed
back into the intake manifold through a EGR cooler.

The purpose of EGR is to reduce NOx emission when the

engine is operating at cruise condition. NOx emission
increases exponentially with flame temperature. The
recirculated exhaust gases dilute the fresh reactants, and
hence reduce the combustion temperature and NOx
emission.

Two methods for achieving EGR are:

1. Traditional EGR (Shown in the adjacent image)
2. Internal EGR (Exhaust is drawn back into the
[Link]
combustor by simply optimizing the exhaust valve
exhaust-gas-recirculation
timing)
 Engine variables controlling NO emission in SI engines
Diluents: The diluents can be the residual gas from the previous
cycle and any exhaust gas recycled to the intake for NOx
emissions control. The residual gas fraction is influenced by load,
valve timing (especially the extent of valve overlap), and, to a
lesser degree, by speed, air/fuel ratio, and compression ratio. The
burned gases act as a diluent in the unburned mixture; hence,
increasing the burned gas fraction reduces NO emissions levels.
The primary effect of the
burned gas diluent is that it
reduces flame temperature
by increasing the heat
capacity of the cylinder
charge, per unit mass of
fuel. For example, CO2
has higher specific heat
capacity than He; hence,
CO2 is more effective for
NOx reduction.
Engine variables controlling NO emission in SI engines
Spark timing: Spark timing significantly affects NO emission levels. Advancing the timing so that
combustion occurs earlier in the cycle increases the peak cylinder pressure; retarding the timing
decreases the peak cylinder pressure. Higher peak cylinder pressures result in higher peak burned gas
temperatures, and hence higher NO formation rates. For lower peak cylinder pressures, lower NO
formation rates result. The figure shows typical NO engine-out emission data for a spark-ignition
engine as a function of spark timing. NO emission levels steadily decrease as spark timing is moved
closer to TC.
NOx formation in CI engines
NO formation is insignificant during the fuel-rich premixed
burning phase; almost all the NO will form in the mixing-
controlled diffusion-flame burning process. Chemical
reaction is centered where the mixture composition is close to
stoichiometric, producing the highest NO formation rates.
At high load, with higher peak P and T, and larger flame area,
NO levels increase. Both NO and NOx concentrations were
measured; NO2 is 10 to 20% of total Nox. Though NO levels
decrease with a decreasing overall equivalence ratio, they
do so much less rapidly than do spark-ignition engine NO
emissions since, in the diesel, the zones where NO
formation rates are highest are stoichiometric.
Though the amount of fuel injected decreases proportionally
as the overall equivalence ratio is decreased, almost all the
fuel still burns close to stoichiometric. Thus, NO emissions
should be roughly proportional to the mass of fuel injected,
provided burned gas pressures and temperatures in the
diffusion flame do not change greatly.
EGR in CI engines

The figure shows the effect of dilution of the intake

air with N2, CO2, and exhaust gas on NOx exhaust
levels. The heat capacity of CO2 (per mole) at the
temperatures relevant to diesel combustion is about
twice that of N2. That of exhaust gas is slightly
higher than that of N2. Thus, these data show that the
effect is primarily one of reduced burned gas
temperatures. Note that the composition of the
exhaust gas of a diesel varies with load. At idle, there
is little CO2 and H2O, and the composition does not
differ much from that of air. At high load the
diluent capacity increases since the concentrations
of CO2 and H2O in the recycled exhaust are
substantially higher.
Bore = 140 mm, stroke = 152 mm, rc = 14.3, speed
= 1300 rev/min, injection timing at 4° BTC.
EGR in CI engines

The engine can tolerate high EGR levels (50%) at

low load since at very lean operating condition (ϕ ∼
0.3), a major fraction of this EGR is effectively “air.”
As load and overall equivalence ratio increases, the
mixture becomes less lean, and a larger fraction of
the EGR is effectively “stoichiometric burned gas”
and less is “air.” Thus, EGR behavior in the diesel
combustion process is significantly different than
that in a spark-ignition engine. Only a portion of the
EGR acts as an inert diluent in the close-to-
stoichiometric region of the diffusion flame where
almost all of the NO is formed.

Effect of recycled exhaust gas (EGR) at light, mid, and

high loads: Heavy-duty diesel engine at 1800 rev/min.
CO/UHC formation and
control
CO emission from SI engines
Carbon monoxide emissions from internal combustion
engines are controlled primarily by the equivalence ratio. The

CO mole fraction
figure shows CO levels in the exhaust of a spark-ignition
engine. For fuel-lean mixtures, CO concentrations in the
exhaust is of order 10-3 mole fraction.

Since spark-ignition engines often operate close to

stoichiometric at part load and, depending on calibration, may
be fuel-rich at high load, CO emissions are significant and
need to be controlled. Diesels, however, always operate well
𝜙 ≈ 1.0 Equivalence
on the lean side of stoichiometric; CO emissions from diesels
are low enough to be unimportant, and will not be discussed ratio
further.
CO formation (faster): RH → R → RO2 → RCHO → RCO → CO (R stands for hydrocarbon radical)
CO oxidation (slower): CO + OH = CO2 + H
Since CO oxidation reactions are slower than the formation reactions, measured CO concentration at
exhaust is significantly higher than the equilibrium concentrations.
Sources of UHC emission in SI engines
Hydrocarbons are the consequence of incomplete combustion of the hydrocarbon fuel. The level of
unburned hydrocarbons (HC) in the exhaust gases of gasoline fueled spark ignition engines is
typically 1 to 2% of the fuel: in diesels it is much less. HC emissions rise steadily as the mixture
becomes richer than stoichiometric. With very lean mixtures, as combustion quality deteriorates, HC
emissions rise rapidly due to incomplete combustion or misfire in a fraction of the engine’s operating
cycles.
There are several sources that contribute to engine HC
emissions. By sources, we mean any HC escaping the

UHC mole fraction

primary combustion process.
UHC formation mechanisms:
Crevices
HC Absorption/desorption in engine oil
HC Absorption/desorption in engine depositions
Liquid fuel on combustor walls
Flame quenching on combustor walls
Incomplete combustion 𝜙 ≈ 1.0 Equivalence
Leakage of the gases through the exhaust valve ratio
Blowby losses (historically, now solved)
Sources of UHC emission in SI engines
Sources of UHC emission in SI engines: Crevices
Sources of UHC emission in SI engines: Crevices
The crevices in the combustion chamber walls—small volumes with narrow entrances—into which
the flame is unable to penetrate have been shown to be the largest source of unburned HC. Total
crevice volume is about 1.5 to 2.5% of the clearance volume and the piston ring crevices are the
dominant contributor.

Representative crevice volumes

in a typical gasoline engine.
Sources of UHC emission in SI engines: Crevices
As the cylinder pressure rises during
compression, unburned mixture is forced
into the crevice regions. Since these
volumes are thin, they have a large
surface/volume ratio so the gas flowing
into each crevice cools by heat transfer to
close to the wall temperature. During
combustion, while the pressure continues
to rise, unburned mixture continues to
flow into the crevice volumes. When the
flame arrives at each crevice, it can either
propagate into these crevices and partially burn the
fuel and air within the crevice, or it can quench at the
crevice entrance. Once the cylinder pressure reaches
its peak and starts to fall, gas flows back from each
crevice into the cylinder. The most important of these
crevices, the volume between the piston, piston rings,
and cylinder wall, is shown schematically.
Example 1
In a 4-stroke SI engine, the mass of the fuel-air mixture in the crevices is maximum when the piston is
at the TDC. The volume of the crevices is 2% of the cylinder clearance volume; the crevice gases and
the in-cylinder gases are at an average temperature of 150°C and 1950°C, respectively. Calculate the
mass of gas in the crevices relative to the total mass in the cylinder. Assume that the pressure inside the
crevices is equal to the cylinder pressure.

Answer: About 10%

Sources of UHC emission in SI engines: Lubrication oil film
An oil layer a few microns thick on the cylinder liner is uncovered by the descending piston during the
intake stroke. During this time the oil layer is exposed to unburned fuel-vapor, air, residual gas mixture.
As a consequence, such vapor components are transferred from the bulk gas into the oil layer. During
the expansion and exhaust strokes, the oil layer on the liner, uncovered by the downward moving piston,
has higher concentrations of gasoline compounds in the oil than correspond to equilibrium with the
close-to-zero levels of gasoline in the burned gases which fill the cylinder. Thus, hydrocarbons are
desorbed out of the oil film and diffuse into the cylinder gases. These contribute to the engine-out UHC
emissions. Oil layers within the cylinder are a source of some 10 to 25% of the UHC emissions.

Gasoline component vapor concentration profile in the (a) (b)

in-cylinder gases and in the lubricating oil on the
cylinder liner: (a) during intake and compression and
(b) post combustion and during exhaust.
Sources of UHC emission in SI engines: Wall deposition
Over extended run, deposits build up on the intake valves, and on the combustion chamber walls of
spark-ignition engines fueled with standard gasolines. Both these types of deposits affect
hydrocarbon emissions. SI engine deposits are porous in structure. The larger pores apparently act as
“distributed crevices,” and store HC during the combustion process. Substantial levels of deposits in
modern spark-ignition engines can cause steady-state engine-out hydrocarbon emissions to increase
by significant amounts.
Sources of UHC emission in SI engines: Liquid fuel on cylinder walls
In port-injected engines, the injector usually directs the liquid gasoline toward the back of the intake
valve and surrounding intake port surface area. With injection toward an open valve, the fuel will
enter the cylinder as liquid. The less volatile fuel components may form a liquid film, or may be
stored in the deposits, oil layers or crevices in the combustion chamber. The incomplete oxidation of
this fuel would contribute to the exhaust hydrocarbon emissions.

With GDI, fuel is injected directly into the combustion chamber through an injector located in each
cylinder head. GDI eliminates the liquid fuel films formed on the intake port and valve walls.
However, there will be much more liquid fuel within the cylinder, especially with a cold engine and
during warm up. The behavior of the DI fuel spray does need to be carefully managed to avoid any
substantial liquid wetting of the cylinder liner and piston crown surfaces.
UHC formation mechanism in SI engines: Flame quenching
The flame extinguishes at a short distance from the cooled combustion chamber walls leaving a thin
layer of unburned fuel-air mixture adjacent to the wall. This is called a quench layer. While
originally thought to be a major source of unburned HC emissions, it is now known that the mixture
in these quench layers diffuses into the hot combustion products outside the layer and burns up
during expansion, under most engine operating conditions. However, the HC burnout may remain
incomplete under high-EGR engine operation. Flame quenching at the combustion chamber walls
prevents a small fraction of the fuel (about 0.5%) from burning during the normal engine combustion
process. Quench layers thus constitute a small loss in engine performance and efficiency.

Head-on quenching Side-wall quenching

All bounding surfaces of the combustor (walls, piston crown, cylinder head) are also sites for flame
quenching. Typically, 𝛿𝑞 ȁ𝑆𝑊𝑄 > 𝛿𝑞 ȁ𝐻𝑂𝑄 .
UHC formation mechanism in SI engines: Blowby losses

Blowby is the gas that flows from the combustion chamber,

past the piston and rings, into the crankcase. It is forced
through any leakage paths afforded by the piston-liner-ring
assembly in response to high in-cylinder gas pressures
during compression, combustion, and expansion. Blowby of
gases from the cylinder to the crankcase removes gas from
this crevice region and thereby prevents some of the crevice
gases from returning to the cylinder. Crankcase blowby
gases were once vented directly to the atmosphere and
constituted a significant source of HC emissions. The
engine’s crankcase is now vented to the intake system, the
blowby gases are recycled, and this source of HC
emissions is now fully controlled.

[Link]
Summary: SI engines
UHC formation in CI engines
The typical sources of UHC in SI engines, such as gas mixture trapped in crevices or fuel
absorption/desorption from oil layer/deposits, is not significant in CI engines. Although crevices exist,
they contain air or very lean mixtures. Hence, the unburned trapped fuel is smaller.

In SI engines, the absorption typically happens during compression, and desorption during exhaust.
These stages do not contain fuel in CI engines. Hence, this is insignificant.
UHC formation in CI engines
There are two primary paths by which hydrocarbons can
escape this diesel combustion process unburned: Overleaning
and Undermixing.

Overleaning: As soon as fuel injection into the cylinder

commences, a distribution in fuel/air equivalence ratio across
each fuel spray develops. The amount of fuel that is mixed
leaner than the lean combustion limit ( ϕ ≈ 0.3) increases
rapidly with time. In each fuel spray core, the mixture is
significantly richer than stoichiometric. However, fuel close
to the spray boundary has already mixed beyond the lean
limit of combustion and will not auto-ignite. Within this
region, unburned fuel, fuel decomposition products, and
partial oxidation products will exist; some of these will
escape the cylinder without being burned. Overleaning of fuel
injected during the ignition delay period is a significant
source of hydrocarbon emissions, especially under conditions
where the ignition delay is long.
UHC formation in CI engines
There are two primary paths by which hydrocarbons can escape this diesel combustion process
unburned: Overleaning and Undermixing.

Undermixing: Under transient conditions as the engine goes through an acceleration process,
overfueling can occur. Even though the overall equivalence ratio remains lean, locally over-rich
conditions may persist through the expansion stroke and into the exhaust process. HC emissions are
unaffected by an increasing equivalence ratio until a critical value of about 0.9 is reached, when levels
increase dramatically. This mechanism is not significant under normal operating conditions, but can
contribute HC emissions under heavy acceleration conditions if over-fueling occurs.
Soot formation and control
Overview
Particulate emissions have long been an issue with diesels. While much of the soot burns up inside the
cylinder, a portion of it does not and is exhausted.

• With port fuel injection gasoline engines, the particulate emissions are much lower than in diesels:
by a factor of about 70 on a mass basis.
• However, direct-injection gasoline engine particulate emissions are close to a factor of 10 higher
than those from port-fuel-injection equivalent engines, again by mass.
Soot formation process
Growth of soot precursors: Benzene → Naphthalene (2-ring) → Pyrene (4-ring) → Coronene (7-ring)
Nucleation: Condensation of gaseous PAH molecules to solid soot particles.
Surface growth: Soot surface growth happens through the HACA (H-abstraction C-addition)
mechanism. Through surface growth, the molecular weight of the particle increases, and the ring
structures collapse to create spherical globules. These globules are also called nascent soot in the
literature, having a diameter of 2-10 nm.

Hydrogen abstraction:
Carbon addition:

Coagulation: When two nascent soot globules collide, they may coagulate to create a larger spherical
globule. The growth of these globules through coagulation continues till a critical diameter of 10-50
nm. The diameter of the spherical globules does not grow beyond this size. They are called primary
particles. A single spherule contains 105 to 106 carbon atoms.
Agglomeration: When two primary particles collide, they agglomerate instead of coagulating, giving
rise to the typical chain-like structure of mature soot particles seen under TEM. In a diesel spray, the
mature soot particles may form near the jet tip, just before oxidation in the diffusion flame front. A
cluster may contain 20-100 individual spherules.
Soot formation process: Schematic

Primary Primary
particle particle

100 nm 100 nm

Dastanpour, Ramin, and Steven N. Rogak. Aerosol Science and Technology 48.10 (2014): 1043-1049.
Soot formation in diesel spray

Conceptual sketch of soot processes near the lift-off length in turbulent

lifted jet flame. The schematic shows one-half of an axisymmetric jet.

Reference: Kook, Sanghoon, and Lyle M. Pickett, SAE

International Journal of Fuels and Lubricants 5.2 (2012): 647-664.
Research paper on soot volume fraction (fv) measurement
Slides 44-50 discuss the contents of the following research paper on the quantitative measurement of
soot volume fraction in an optical chamber using various fuels.

𝑉𝑠𝑜𝑜𝑡
𝑆𝑜𝑜𝑡 𝑣𝑜𝑢𝑚𝑒 𝑓𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑓𝑣 =
𝑉𝑔𝑎𝑠

Reference: Kook, Sanghoon, and Lyle M. Pickett, SAE International Journal of Fuels and Lubricants
5.2 (2012): 647-664.
Property of the fuels studied
Ambient and injector conditions

The ambient gas conditions were selected to

mimic operating conditions for the latest
generation of diesel engines that use moderate
EGR and have minimal NOx emissions. The
ambient density was fixed at 22.8 kg/m3 while
temperature and pressure were varied.
Flame lift off length
Quantitative soot volume fraction (fv) for all tested fuels

At 900 K, 𝑓𝑣 is lower than at

1000 K.

𝑓𝑣 is strongly dependent on the

fuel property.

How can we explain the

observations?

It seems the FLOL is inversely

correlated with 𝑓𝑣 . What is the
reason?
Explanation

High cetane number and high ambient temperature reduce ignition delay → Flame lift-off length is
reduced → Air entrainment into the fuel is reduced → Equivalence ratio at the flame lift-off location is
higher → measured soot volume fraction is greater.
Explanation

Differences in the sooting propensity for fuels with the same FLOL
can be explained through their aromaticity. For example, The FLOL
is similar for D2, JC, and JW. But 𝑓𝑣 is highest with D2 as aromatic
volume % is the highest.
NOx-soot trade off in CI engines

The soot formation region (richer than ϕ ≈ 2 with

T between 1600 K and 2200 K) with percentages
of fuel carbon converted to soot, and the NO
formation region, are shown. A typical path for a
fuel parcel in a conventional diesel combustion
process (with air) ignites at a ϕ in the 2 to 4
range, and enters the soot formation region as
rich reacting fuel-air mixture; then in the
Typical path of a fuel diffusion flame ( ϕ ≈ 1, T = 2700 K) the burned
parcel in a diesel engine gases are in the NO formation region.
GDI engine particulates
Unlike port fuel injection engines, GDI sprays fuel directly into the cylinder. The residence time
available for atomization and mixing is less compared to PFI, leading to regions of locally rich mixture
and incomplete atomization. These effects promote soot and particulate matter (PM) formation, even
though the overall mixture is lean or stoichiometric. The particulates are typically in the sub-micron
range (PM₂.₅ and below).

The introduction of GDI led to the regulation of PM emissions in Euro6/BS6. Emission norms before
Euro 6 didn’t have any restriction on PM emissions from SI engines. Aftertreatment devices such as
gasoline particulate filters (GPF) are essential for BS6 compliance with GDI-based engines.
After-treatment devices
Basic operating principle of catalysts
The task of catalytic aftertreatment devices is to oxidize the CO and HC in the exhaust gas to CO2
and H2O at temperature levels that the exhaust gases can readily provide. With NOx, the task is
reduction to N2, at or below the available exhaust gas temperature. The activation energy (or energy
barrier) for breaking these air pollutant molecule bonds is above the thermal energy available at
average exhaust gas temperatures. The catalyst facilitates the desired chemistry by reducing this
energy barrier so reactions can occur at significantly lower temperatures.

For CO and HC oxidation on the catalyst surface,

sufficient oxygen must be available. With a
stoichiometric mixture, the air (oxygen) associated
with the CO and unburned HC is available. For NO
reduction in spark-ignition engines, reducing agents
in the exhaust gas (CO, H2, and HC) are present.
Catalytic oxidation of CO and hydrocarbons and
reduction of NOx in the exhaust can be achieved at
temperatures as low as 250 to 300°C.
Three-way catalytic converter (TWC)

Fuel Engine-out Exhaust to

emissions Three-way atmosphere
SI Engine catalytic
converter
Air

𝑈𝐻𝐶 → 𝐶𝑂2 , 𝐻2 𝑂
Oxidation
𝐶𝑂 → 𝐶𝑂2
𝑁𝑂𝑥 → 𝑁2 Reduction
TWC structure - monolith
Monolith

[Link]
Individual channel cross-
section: 1 mm X 1 mm

Monolithic catalyst substrates are made of ceramics or metal. Ceramic substrates usually have square
cells, while most metallic substrates have sinusoidal channels. Parts that were commercially used for
internal combustion engine applications in the 1990s had cell densities between 200 and 600 cpsi (cells
per square inch), with 400 cpsi being the most common cell density in gasoline car applications. Since
then, increasingly higher cell densities have been introduced in gasoline cars, with parts in excess of
1000 cpsi being available in both the ceramic and metallic designs.
 TWC structure – washcoat and catalysts

Catalyst washcoat Schematic of catalytically active sites dispersed over the washcoat surface

Foils used for metal substrates have no porosity. Since high surface area carriers are required for catalysts,
it is necessary to deposit a high surface area coating onto the channel walls. That coating, called the
washcoat, is composed of porous, high surface area inorganic oxides such as gamma-Al2O3. Noble metal
catalysts, such as platinum, palladium, and rhodium, are deposited on the surface and within the pores
of the washcoat. The thickness of the washcoat layer is on the order of 20-40 µm. Rhodium is the
principal catalyst used to remove NO. Platinum is commonly used for oxidizing paraffinic compounds; it
has good reactivity under stoichiometric and slightly lean conditions. Palladium has excellent reactivity
for olefinic and aromatic HC.
TWC operation: Effect of equivalence ratio
Catalyst conversion efficiency can be defined as:
𝑈𝐻𝐶 → 𝐶𝑂2 , 𝐻2 𝑂
Oxidation 𝑚ሶ 𝑈𝐻𝐶,𝑖𝑛 − 𝑚ሶ 𝑈𝐻𝐶,𝑜𝑢𝑡
𝐶𝑂 → 𝐶𝑂2 𝜂𝐶𝑎𝑡𝑎𝑙𝑦𝑠𝑡,𝑈𝐻𝐶 =
𝑚ሶ 𝑈𝐻𝐶,𝑖𝑛
𝑁𝑂𝑥 → 𝑁2 Reduction

NOx

Conversion efficiency
CO

HC

Lean Rich 𝜙
𝜙=1
TWC are only effective if the exhaust gas is stoichiometric. Precisely at this condition, there is
oxygen available for oxidation and reducing agents (CO, H2, and HC) are available for NO reduction.
TWC operation: Effect of operating temperature
Catalyst conversion efficiency can be defined as:
𝑚ሶ 𝑈𝐻𝐶,𝑖𝑛 − 𝑚ሶ 𝑈𝐻𝐶,𝑜𝑢𝑡
𝜂𝐶𝑎𝑡𝑎𝑙𝑦𝑠𝑡,𝑈𝐻𝐶 =
𝑚ሶ 𝑈𝐻𝐶,𝑖𝑛

Catalyst efficiency increases as temperature increases in an “S-shaped” curve. At the low temperature
end of the curve, there is little catalyst activity. At some point, the catalyst “lights off” and catalytic
chemical activity becomes significant. A commonly used light-off temperature definition is the
temperature ( T50) at which the catalyst efficiency reaches 50%. While 250°-300°C is the typical light-
off temperature, at 400° - 800°C, the steady-state conversion efficiency of a new oxidation catalyst is
typically 98 to 99% for CO and above 95% for HC.
Diesel after-treatment devices
Fuel Engine-out Urea
emissions Diesel Diesel injection Selective
CI Engine oxidation particulate catalytic
catalyst filter reduction
Air
(oxidizes UHC) (traps soot) (reduces NOx)

Diesel oxidation catalysts (DOC) in diesel engines operate based on the same principle as three-way
catalytic converters (TWC) in petrol engines. The key point of difference being that DOC operates
under lean condition, whereas TWC operates close to stoichiometry. CO, UHC, and NO gets
oxidized in the DOC as follows.

𝑈𝐻𝐶 → 𝐶𝑂2 , 𝐻2 𝑂
𝐶𝑂 → 𝐶𝑂2
𝑁𝑂 → 𝑁𝑂2
Diesel particulate filter
An exhaust treatment technology that substantially reduces engine particulate emissions is particulate
filters or traps. A temperature-tolerant filter or trap removes the particulate material from the exhaust
gas: the filter is then “cleaned off” by oxidizing the accumulated particulates—the regeneration
process. Diesel particulate filters (DPF) are in common use.

The technology can also be used on gasoline engines (GPF) and may be needed on direct-injection
gasoline engines since their particulate emissions are significantly higher than equivalent port-fuel-
injected gasoline engine emissions, though lower than diesel engine-out levels.
Challenges with DPF design
The challenges to implementing particulate filter technology are many:

(1) The filter, even when clean, increases the pressure in the engine exhaust manifold.
(2) The pressure drop across the DPF increases steadily rises as the filter collects particulate matter.
(3) Under normal diesel engine operating conditions, the collected particulate matter is not hot enough
to spontaneously ignite and oxidize.
(4) Once ignition of the filtered particulate matter occurs, the burn-up process must be controlled to
prevent excessively high temperatures and trap damage or destruction.
Ceramic wall-flow DPF
The majority of the filters used commercially are ceramic wall-flow based. Such wall-flow filters are
based on monolith catalyst substrates where half the cells are plugged at the inlet end and the other half
at the outlet end. Thus, the exhaust gas is forced to flow through the porous ceramic walls. The
thickness of these walls and their porosity are carefully controlled to provide good particulate filtration
with a low enough pressure drop.
Regeneration
Such filters quickly accumulate considerable volumes of
soot. The pressure drop across the DPF steadily increases as
particulate matter accumulates, and to avoid excessively high
pressures in the engine’s exhaust manifold, the filter must be
regenerated. There are several approaches to filter
regeneration. Thermal regeneration is the primary method:
the solid particulates (soot) in the DPF are oxidized to CO2
(and some H2O). The rate of oxidation of soot is significantly
lower than the oxidation of gaseous emissions such as HC
and CO. The reason that solid particulates are not oxidized in
flow-through oxidation catalysts designed to remove gaseous
pollutants is that they provide relatively short residence
times. Diesel soot oxidation rates are shown as a function of
temperature: temperatures above 500°C are required in the
absence of catalysts. The temperature of diesel exhaust gas is Diesel soot oxidation rate as
normally well below this level. So, the exhaust gas a function of temperature.
temperature needs to be raised.
Diesel engine exhaust temperature

Typical exhaust-gas temperature of a standard application

as a function of engine speed and load (for a diesel engine)
Exhaust temperature is raised through post-injection

TDC

Figure: Comparison of the temperatures at the inlet

and outlet of the DOC with 14+5 mg injection. The
main injection was kept constant at 10 deg bTDC.

Figure: Comparison of the temperatures at the inlet Yoon, Sungjun, et al., Journal of Engineering for
and outlet of the DOC with 14+10 mg injection. Gas Turbines and Power 138.10 (2016): 102810.
Exhaust temperature is raised through post-injection - Explanation
The exhaust gas temperature needs to be raised for DPF regeneration. Fuel injection towards the end
of the expansion stroke is carried out for this purpose. The fuel injected towards the end of the
expansion stroke does not burn within the engine; rather, it flows through the exhaust and burns in the
DOC in the presence of the catalyst.

The figures in the previous slide show that when the late injection is carried out 80° after the TDC,
there is a significant temperature rise across the DOC. This temperature rise indicates that the late-
injected fuel combustion is taking place in the DOC. This is desirable from the perspective of
providing a high temperature to the DPF for DPF regeneration.
Regeneration strategy

Temperature Temperature
after DOC after DPF
Temperature
ahead of DOC

Filter
pressure drop
 Regeneration strategy – Explanation
The figure shows a particulate filter regeneration process lasting about 200 s. The DPF system in this
light-duty diesel vehicle consists of a filter with a DOC in the exhaust system ahead of the DPF. The
engine’s common- rail fuel-injection system can provide two injection pulses: a main injection (primarily
responsible for generating bmep/torque), and a late or after injection well into the expansion process
primarily to raise the exhaust gas temperature to initiate the particulate filter regeneration process. The
figure shows the start of regeneration at 25 seconds, where late fuel injection occurs, and the energy
release from that fuel burning starts to raise the exhaust temperature above about 200°C. The DOC
through additional oxidation of this late-injected fuel pulse, further raises the exhaust gas temperature
downstream of the DOC and upstream of the filter. At 175 seconds, the soot in the filter starts oxidizing
and releasing its chemical energy, causing the filter exit gas temperature to rise sharply. As the soot within
the DPF burns (following the start of regeneration), the filter pressure drop falls.

Regeneration must be carried out approximately every 500 kilometers; this figure may vary from 300 to
800 kilometers depending on the engine operating conditions. The need for regeneration is decided based
on the calculated pressure drop across the DPF.
Lean NOx Trap – Schematic of the operation
The key challenge with NOx emission reduction is that NOx need to go through a reduction reaction
to form N2. However, in the diesel engine exhaust, the surrounding environment is oxidizing (lean
environment with excess O2). To circumvent this challenge, the lean NOx trap operates in two
phases: Storage and Regeneration.

Lean operation (𝜙 < 1) Rich operation (𝜙 > 1)

Lean NOx Trap – Storage
The key challenge with NOx emission reduction is that NOx need to go through a reduction reaction
to form N2. However, in the diesel engine exhaust, the surrounding environment is oxidizing (lean
environment with excess O2). To circumvent this challenge, the lean NOx trap operates in two
phases: Storage and Regeneration.

Storage: The NOx-rich exhaust gases pass through a catalyst bed (Pt catalyst) that converts NO to
NO2.
2𝑁𝑂 + 𝑂2 → 2𝑁𝑂2

NO2 then reacts with a storage material, Barium Oxide (BaO), and gets adsorbed into the material as
barium nitrate. Oxides of Ca, Sr, and Mg are other materials that are also effective.
1
𝐵𝑎𝑂 + 2𝑁𝑂2 + 𝑂2 → 𝐵𝑎 𝑁𝑂3 2
2

NO may get directly adsorbed to the storage material as well. However, this reaction is less efficient.

3
𝐵𝑎𝑂 + 2𝑁𝑂 + 𝑂2 → 𝐵𝑎 𝑁𝑂3 2
2
Lean NOx Trap – Regeneration
The regeneration phase spans a few seconds. In the regeneration phase, the diesel engine is temporarily
operated under a fuel-rich condition. As the engine operates fuel-rich, conditions suitable for NOx
reduction are created in the LNT. When operating under rich mode, the stored barium nitrate thermally
decomposes to produce NO2, which reduces with CO, H2 in the presence of platinum and rhodium
catalysts to form N2.

Equivalence ratio during the rich phase is around 1.05.

1
𝐵𝑎 𝑁𝑂3 2 → 𝐵𝑎𝑂 + 2𝑁𝑂2 + 𝑂2
2
2𝑁𝑂 + 𝐶𝑂 → 𝑁2 + 𝐶𝑂2
2𝑁𝑂 + 2𝐻2 → 𝑁2 + 2𝐻2 𝑂

If the mixture is strongly rich, ammonia formation is also possible.

5
𝑁𝑂 + 𝐻2 → 𝑁𝐻3 + 𝐻2 𝑂
2
Lean NOx Trap
As NO2 is adsorbed into BaO, the adsorption sites become quickly saturated, and the adsorption
efficiency starts to decline. One study has reported that after 200s of lean operation, the adsorption
efficiency fell from 100% to 60%. This allows a significant amount of NOx to escape the adsorber
into the environment. Hence, the trap needs to be regenerated.

𝑁𝑂𝑥 𝑎𝑑𝑠𝑜𝑟𝑏𝑒𝑑 𝑏𝑦 𝑚𝑎𝑠𝑠

𝑆𝑡𝑜𝑟𝑎𝑔𝑒 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 =
𝑁𝑂𝑥 𝑖𝑛 𝑡ℎ𝑒 𝑒𝑥ℎ𝑎𝑢𝑠𝑡 𝑠𝑡𝑟𝑒𝑎𝑚

Schönebaum, Simon, et al. ChemCatChem 13.7 (2021): 1787-1805.

Lean NOx Trap – Temperature dependence
The storage efficiency is also a strong function of the operating temperature. In the previous slide, it
is shown that the storage efficiency is lower at 250°C compared to 350°C. Typically, 300°-400°C is
the range of efficient working temperature.

Schönebaum, Simon, et al. ChemCatChem 13.7 (2021): 1787-1805.

Regeneration efficiency
𝑁𝑂𝑥 𝑎𝑑𝑠𝑜𝑟𝑏𝑒𝑑 𝑏𝑦 𝑚𝑎𝑠𝑠
𝑆𝑡𝑜𝑟𝑎𝑔𝑒 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = These formulas are as per the definition used in
𝑁𝑂𝑥 𝑖𝑛 𝑡ℎ𝑒 𝑒𝑥ℎ𝑎𝑢𝑠𝑡 𝑠𝑡𝑟𝑒𝑎𝑚 this research paper: Schönebaum, Simon, et al.
ChemCatChem 13.7 (2021): 1787-1805.
𝑅𝑒𝑔𝑒𝑛𝑒𝑟𝑎𝑡𝑖𝑜𝑛 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦
𝑆𝑡𝑜𝑟𝑎𝑔𝑒 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 𝑎𝑡 𝑡ℎ𝑒 𝑏𝑒𝑔𝑖𝑛𝑛𝑖𝑛𝑔 𝑜𝑓 𝑡ℎ𝑒 𝑠𝑖𝑥𝑡ℎ 𝑐𝑦𝑐𝑙𝑒
=
𝑆𝑡𝑜𝑟𝑎𝑔𝑒 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 𝑎𝑡 𝑡ℎ𝑒 𝑏𝑒𝑔𝑖𝑛𝑛𝑖𝑛𝑔 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑖𝑟𝑠𝑡 𝑐𝑦𝑐𝑙𝑒 Effect of regeneration time
The regeneration efficiency depends on the duration of 20s 20s
the rich phase. The longer the regeneration phase lasts,

Regeneration efficiency (%)

the higher is the regeneration efficiency. However,
having prolonged rich phases negatively affects the 10s
thermal efficiency and bsfc. Due to this, the use of LNT
10s
is limited to smaller vehicles where the application of
SCR is not feasible.
5s
5s
For example, Tata Nexon (smaller vehicle) uses LNT
technology, whereas Tata Harrier (larger vehicle) uses
the SCR technology.
Schönebaum, Simon, et al. ChemCatChem 13.7 (2021): 1787-1805. Operating temperature (°C)
 Lean NOx Trap cycles
Rich
Equivalence ratio

1 cycle
Lean phase
phase

Time

Storage efficiency,
assuming
regeneration
efficiency = 100%
Time
Storage efficiency,
assuming
regeneration
efficiency < 100%
Time
When the regeneration efficiency is less than 100%, the storage efficiency reduces after a few cycles of
operation.
Issues with the LNT technology
Limited storage capacity: As discussed, the storage efficiency continuously decreases during the
lean phase of operation. Hence, NOx slip through the trap keeps increasing till the storage material is
regenerated again.
Rich regeneration: The LNT suffers from significantly high pollution levels (CO, HC, soot, NOx)
during the rich regeneration phase.
Thermal efficiency: The rich regeneration phase also increases fuel consumption, reducing the
thermal efficiency.
Narrow window of operation: Effective NOx storage occurs only in a narrow temperature range
(roughly 300–400 °C). During urban driving conditions, the exhaust temperature is typically too low
for the LNT to operate within this window.
Selective catalytic reduction system components
Selective Catalytic Reduction System
Here, “selective” means that the reducing agent prefers to oxidize selectively with the oxygen
contained in the nitrogen oxides instead of with the molecular oxygen present in much greater
quantities in the exhaust gas. Ammonia (NH3) has proven to be a highly selective reducing agent in
this case. In a car environment, the quantities of NH3 required would raise safety issues due to the
toxicity of the chemical. However, NH3 can be produced from nontoxic carrier substances, such as
urea. An aqueous urea solution made with 32.5% urea and 67.5% deionized water is used now in
commercial systems. This solution is termed ‘AdBlue’.

If more reducing agent is dispensed than is converted in reduction with NOx, it may result in NH3
leakage. NH3 is a gas and has a very low odor threshold (15 ppm). This may cause a nuisance to the
environment, but it is avoidable. NH3 is removable by placing an additional oxidation catalyst
downstream of the SCR catalytic converter. This blocking catalytic converter oxidizes any ammonia
that may occur to form N2 and H2O.

One key parameter for the application is the feed ratio α, which is defined as the molar ratio of metered
NH3 as a factor of NOx present in the exhaust gas. Under ideal operating conditions, NOx reduction of
100% is achievable at α = 1. In practice, however, a NOx reduction of 90% is achievable.
Selective Catalytic Reduction - Schematic
(reaction 1)
(reaction 2)
(reaction 3)

At low temperatures (< 300°C), conversion mainly takes place using reaction 2𝑁𝐻3 + 𝑁𝑂 + 𝑁𝑂2 →
2𝑁2 + 3𝐻2 𝑂. For this reason, it is necessary to set a NO2 : NO ratio of about 1 : 1 to achieve good
conversion at low temperatures. Under these circumstances, reaction 2 can take place at temperatures
starting at 170 to 200°C.
Selective Catalytic Reduction - Explanation
The established NOx catalytic reduction approach for diesels in passenger cars and for most heavy-
duty and off-road applications is selective catalytic reduction (SCR). Urea is generally the source of
the ammonia (NH3) used as the reducing agent in this catalyst system. SCR catalysts have a higher
efficiency than LNT technology.

The overall system is complex, and a typical layout is shown in the previous figure. The exhaust first
passes through an oxidation catalyst (V), where the NO is largely oxidized to NO2. A high NO2 /NOx
ratio entering the SCR catalyst is important. The HC and CO in the exhaust are also oxidized to CO2
and H2O. Urea (NH2)2CO is injected into and mixes with the exhaust gases, and that mixture then
passes through a hydrolysis catalyst (H)—the first part of the SCR system—where water in the
exhaust reacts with the urea, releasing ammonia (NH3). The figure shows the overall chemical
reactions that occur in each stage of the total process. The ammonia is the reducing agent in the SCR
catalyst (S), which converts the NO2 and NO in the mixture exiting the oxidation catalyst to N2. The
amount of NH3 supplied is about 20% higher than the amount required for the NO2 and NO to be
reduced to N2. A final oxidation catalyst (O) then removes any excess NH3 by oxidizing it to N2.
SCR catalysts
The overall NOx conversion efficiency over a specific operating temperature depends on the catalyst
used. Platinum catalyst technology can only be used at temperatures below about 250°C due to its
poor selectivity for NOx reduction at higher temperatures. A vanadium catalyst with a titanium oxide
carrier shows good efficiency up to above 500°C. Copper zeolite catalysts have been developed to the
point where they show high efficiency (>95%) over a wide temperature range up to about 650°C.
LNT – SCR combination
SCR is more efficient and more reliable than LNT for reducing NOx emission levels to the
environment. The only issue with SCR is that it requires a large sub-system dedicated to urea: fuel
tank, filter, pumps, atomizers. All this increases the cost of the system, making SCR unfeasible for
smaller vehicles.

Some manufacturers (Tata Nexon) use a clever combination of LNT and SCR to get around the cost
issue associated with a full SCR system. Remember that the urea reduces to ammonia, and ammonia
reduces nitrogen oxides to nitrogen. In these LNT – SCR combined systems, instead of externally
supplying urea → ammonia, ammonia is generated in the LNT. This ammonia is used as a reductant
in the SCR.

If LNT operates with a significantly rich air-fuel mixture, the following reaction takes place. Through
this, LNT creates ammonia, which is the reductant used in the SCR. This combined system has its
own limitations. Hence, a full-fledged SCR system with externally supplied urea solution is common
with larger vehicles (buses, trucks, etc.).
5
𝑁𝑂 + 𝐻2 → 𝑁𝐻3 + 𝐻2 𝑂
2