Hindawi

Advances in Materials Science and Engineering

Volume 2022, Article ID 1863296, 15 pages
https://doi.org/10.1155/2022/1863296

Research Article
Effect of Low-Cost Perovskite Based Lean NOx Trap to Reduce the
NOx Emission for CRDI Engine

Chinnadurai Kombiah , Senthil Kumar Pachamuthu, and Bharathan Rajappa

Department of Automobile Engineering, MIT Campus, Anna University, Chennai, India

Correspondence should be addressed to Chinnadurai Kombiah; chinnak2406@gmail.com

Received 11 February 2022; Accepted 15 March 2022; Published 13 April 2022

Academic Editor: Qian Chen

Copyright © 2022 Chinnadurai Kombiah et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Lean NOx trap (LNT) is an after-treatment technology that targets NOx emissions from an IC Engine and reduces it by using fuel
as the reductant. The main mechanisms involved in this process are (i) oxidation of NO into NO2, (ii) Entrapment of NO2 over an
alkali Earth metal in the form of nitrates, and (iii) Regeneration and reduction of NO2. The oxidation and reduction reactions are
usually performed over rare Platinum Group Metals (PGM). The alternate cycling of operation modes helps in the regeneration
and reduction stage due to the presence of unburnt hydrocarbons and carbon monoxides. These HC and CO generated during
profitable operation can reduce agents similar to ammonia in SCR (Selective Catalytic Reduction). In the case of LNT, the fuel-
injected for reduction imposes a penalty on the system. Cobalt-based perovskites with the chemical formula “ACoO3” (where “A”
denotes Ba, Sr, etc.) show excellent absorption and reduction tendencies towards NOx and are highly suited for LNT as they have
higher NO oxidation capacity. This research experimented with two different coating procedures (with and without binder) with
three other perovskite catalysts–BaCoO3, SrCoO3, and Ba0.5Sr0.5CoO3. Both the coatings were tested for their adherence, and it
revealed that the binder-less layer had only 4% coating loss after 2 hrs of sonication. BaCoO3 showed higher NOx absorption due
to the higher electropositive nature of barium. The peak conversion efficiency attained in this experiment was 84% at a catalyst
temperature of 350°C and 40% load condition.

1. Introduction are the two significant pollutants targeted in the present

scenario. Impactful NOx emission control is only possible if
Motorization has led to a significant increase in exhaust the formation processes are better understood. The oxidation
emissions worldwide in recent years. As a result, developed of atmospheric Nitrogen is indeed the primary source of NO.
countries that abide by pollution standards, including such The diatomic nitrogen molecule (N2), which accounts for
Tier 3 (US), LEV III (California), as well as Euro 6 (EU), are approximately 78 percent of atmospheric air, has a partially
trying to design testing methods that determine emission static nature due to the presence of a triple bond between the
levels more accurately in real-world conditions rather than in two atoms (N�N). On the other hand, Nitrogen-based
the research group. The Euro 6d emission standard contains compounds are formed if enough energy is supplied to break
both an RDE (Real Driving Emission) test and a WLTC this bond. A high-energy environment like this exists inside
(Worldwide Harmonized Light Vehicle Test Cycle) dyna- the cylinder of a vehicle’s internal combustion engine. Peak
mometer [1]. Developing countries like China, India, and combustion temperature changes (2500K to 3000K) result in a
Pakistan have officially fulfilled Euro 6 standards. However, variation of nitrogen oxides. Exposure to NOx emissions leads
the Indian government has opted to relocate from Bharat Stage to respiratory illness and asthma in human beings [3]. Fur-
IV to Bharat Stage VI standards, commencing in 2020, to meet thermore, NOx tends to react with unburned hydrocarbons
the global emission standards [2]. Because of their hazardous to form smog, which also causes emphysema and eye irrita-
impact, nitrogen oxides (NOx) and particulate matter (PM) tion. It also contributes to particulate matter formation [4].
2 Advances in Materials Science and Engineering

Particulate matter with a diameter of fewer than 10 microns indicated that the NOx storage capacity increased with the
has been linked to lung illnesses like bronchitis. [5]. The seven basicity of the storage component used [11]. The NOx is
nitrogen-based oxides formed depending on the valence states generally stored as nitrates during lean air/fuel conditions.
of atomic nitrogen (N2O5) are nitrous oxide (N2O), nitric However, experiments have shown the presence of nitrites
oxide (NO), di-nitrogen dioxide (N2O2), dinitrogen trioxide during this oxidation process [12]. Figure 1 shows the LNT
(N2O3), nitrogen dioxide (NO2), dinitrogen tetroxide (N2O4), in acts [11].
and dinitrogen pentoxide. The NOX family is composed of The succeeding components are found in the 1st Gen-
these seven compounds. Due to their dominance in the ex- eration LNT: the roles of each element in an LNT have been
haust, the primary two forms of nitrogen oxides concentrated compiled [13], with both the following features having
on automotive emission control are NO and NO2. Pollutant played significant roles:
emissions are caused by the fuel constituent elements, com- Platinum (as a catalyser): In general, the NOx compo-
bustion characteristics, and combustion chemistry. For ex- sition consists mainly of NO (around 95 percent) and NO2
ample, nitric oxide appears throughout the burned gases, both (5 percent). The ability of an LNT to trap NO2 has been
in the flame and behind the flame front, as a direct conse- affected by the concentration of NO2, as NO2 has a sig-
quence of chemical reactions involving nitrogen and oxygen nificantly larger tendency to form nitrates or nitrites on the
atoms that do not reach chemical equilibrium. The fuel/air trapping component. As just a result, the catalyst’s oxidation
equivalence ratio is significant in determining emissions in is a vital step. It also gives atomic oxygen to the nitrite
spark-ignition engines. In the case of diesel engines, the species, making the production of nitrite layers more likely
primary determinant is fuel distribution and how that dis- [14]. Besides that, NOx reduction always occurs over the
tribution varies with time due to mixing [6]. Due to non- platinum catalyst.
uniform mixing in diesel engine combustion, NOx is formed Barium Oxide (Trapping Component): the major
in regions where the mixture is close to stoichiometric. These component of an LNT participates actively in the trapping
various NOx control techniques have developed into three process. According to the literature, barium oxide tends to
categories: trap NO2 than NO [13]. In Figure 2, they demonstrate two
different types of reactions.
(1) Fuel precombustion treatment
Alumina (support): alumina’s primary role in an LNT is
(2) Exhaust after-treatment postcombustion. (a) SCR to support the catalyst and trap component. A larger surface
(using ammonia, urea, or HC as a reducing agent). area leads to more significant catalytic activity in some
(b) LNT changes to the combustion process instances. Some studies [15] show that alumina forms
(3) (a) Oxygen reduction (absorption of less air); (b) aluminum nitrate and, in a manner, traps NOx by doing so.
temperature reduction (water injection and exhaust However, due to Al (NO3)’s low thermal stability, the
gas recirculation); (c) reducing residence time percentage of NOx trapped by alumina is usually ignored.
(burner configuration and chamber duration) Several mechanisms constrain the chemistry of the NOx
trap. The widely accepted NOx trap mechanisms, on either
Each method has distinct advantages and disadvantages.
hand, follow the nitrate or nitrite route description of the
However, EGR, water injection (or electronic steam injection
research work [16]. In the early days, NOx chemistry and
(ESI), retarding injection timing (or spark timing for petrol
NO2 disproportionation reactions were common. However,
engines), engine downsizing, and divided injection [7] are
two parallel pathways have recently been found to occur in a
simple and widely used NOx reduction techniques. On the
lean NOx trap. First, before proceeding to a reduction stage,
other hand, EGR has a limited impact on engine efficiency
the NO2 trapped in the alkali component must regenerate.
[8, 9], water injection poses corrosion risks [10], and split
The nitrate decomposition from the trapping component
injection necessitates expensive and time-consuming elec-
could’ve been caused by the temperature rise during rich fuel
tronic control systems. Furthermore, the presence of any after-
induction and the decrease in oxygen concentration in the
treatment device in the tailpipe diminishes engine efficiency,
decadent phase, minimizing nitrate equilibrium stability
which can be reduced by optimizing the backpressure [10].
[17].
The Pt/Ba/Al2O3 catalyst used in this history became the
Regeneration:
first generation Lean NOx Trap. LNT functionality is divided Δ
Ba(NO3 )2 ⟶ BaO + 2NO2 + (1/2)O2 NO2 ⟶
into five stages:
NO + (1/2)O2 .
(1) NOx (NO & NO2) oxidation to NO2—lean A part of NO2 also reduces into NO.
Reduction:
(2) NO2 Adsorption as nitrates or nitrites on alkali site
2NO (or NO2) + 2 CO ↔ 2 CO2 + N2.
surfaces—lean
NO (or NO2) + HC ↔ CO2 + H2O + N2.
(3) Fuel-rich exhaust gas reducing agents (HC, CO, or (x + y/2)NO (or NO2 ) + Cx Hy ⟶ (y + x/2)N2 + y
H2)—lean CO2 + (n/2)H2 O (x + y).
(4) Regeneration of NO2 from nitrate sites—rich 2H2 + 2NO (or NO2) ⟶ 2H2O + N2.
H2 O + H2 ⟶ NOx − ⟶ NH3 +N2 .
(5) NO2 breaks down into Nitrogen and oxygen—rich
NH3 + NOx ⟶ N2 + H2O.
Because of the stability of the nitrates formed during the Ammonia formed during the intermediate reaction have
reaction of NO2 with barium, the storage mechanisms been further used to reduce NOx into N2 and H2O.
Advances in Materials Science and Engineering 3

OH
O O HC H

O N N O O N CC O N N
O O C
O
Nitrate Nitrate
Pt Pt
BaO BaO
Al2O3 Al2O3
Substrate Substrate
Lean Period Rich Period
Figure 1: Working of a lean NOx trap.

Pt + BaO
Ba(NO2)2
NO + O2 Nitride

Pt / O2
Pt
NO2

NO2 BaO Ba(NO3)2

Nitrate
NO
Figure 2: Nitrate & Nitrite formation in LNT.

H2 O + CO ⟶ NOx − ⟶ H2 + CO2
N2, CO2, and H2O were the most significant byproducts
generated. The reduction and regeneration procedures can
differ depending on the reductant used. The reductant is the fuel
because the fuel type determines the reduction and regeneration
in most cases. Studies would install a secondary injector on the ‘A’ cation
tailpipe to spray a secondary fuel to reduce the NOx stored. This
setup would be a less expensive and more effective reductant. ‘B’ cation
This method is similar to SCR because it uses secondary fuel
tanks and secondary injection control systems.
‘O’ anion

2. Catalysts of the Perovskite-Type and Figure 3: Unit cell of Perovskite.

Their Properties
Gustav Rose discovered perovskites in 1839 and named the 2.1. NOx Trapping Ability of Perovskites. The NOx trapping
others after the Russian mineralogist, Count L. A. Perovski. ability of perovskites has received considerable attention.
The first Perovskite realized was Calcium Titanate, CaTiO3, According to [19], chemisorption of NO2 occurs due to the
which had the general formula ABO3, where “A” and “B” are formation of nitrate or nitrite, which is quite similar to that
cations and “O” is the anion that bonds both, as shown in of traditional LNT. Moreover, [20] proposed that the ox-
Figure 3. Perovskite generally has a cubic crystal structure, as ygen vacancies of perovskites also serve as NOx storage
do most transition metal oxides. sites in their research. The research [21] also disclosed a
Perovskites have several desirable properties, including chemical reaction in which NO2 was stored as metal ni-
electrical, magnetic, optical, and catalytic properties, given trates. Metal nitrates would’ve been reverted to perovskite
that nearly 90 percent of all elements in the periodic table form during the rich phase due to calcination caused by
can be stable in the perovskite structure and the possibility of higher temperatures in the exhaust. The reaction of NO2
partial substitution of A and B sites [13]. The following two trapped in barium cobalt oxide perovskite is represented
conditions must be satisfied to form a stable perovskite: below.
3 BaCoO3 + 6 NO2(g) + 2 (1 + 3y)O2(g) ⟶ 3Ba(NO3)
(1) Cation “A” must have a larger ionic radius than 2(s) + Co3O4(g).
cation “B” to be accommodated properly in the unit
cell. In general, rA> 0.090 nm and rB > 0.051 nm,
with a tolerance factor ranging from 0.8 to 1.0 [18]. 2.2. Perovskite Synthesis. Perovskites are mainly synthesised
(2) The perovskite compound must be electrically using the four methods mentioned below:
neutral, which is only possible if the total positive ion 1. Acetic the Sol-gel Method, 2. Method of co-precipi-
charge of “A” and “B” constitutes the oxygen ion tation, 3. Synthesis of combustion, 4. Reactive grinding
charge. method.
4 Advances in Materials Science and Engineering

Each method has a distinct effect on the Perovskite Perovskites tend to form mullites and aluminates
formed in surface area, structure, and morphological pa- [24]. Hence a layer of Ceria is added over it.
rameters. The sol-gel citric acid method of synthesis to (2) A cerium oxide slurry was formed by dissolving
generate the following catalysts: BaCoO3, SrCoO3, and polyvinyl alcohol in water at 85°C, followed by
Ba0.5Sr0.5CoO3. glycerol and cerium oxide. This slurry was then dip-
This process starts by dissolving the nitrate salts of the coated onto the alumina layer after being ball milled
desired oxides in the stoichiometric amount and then for 20 hours.
proceeding as explained elsewhere [22]. For example, to
(3) This layer was impregnated with the perovskite
begin producing BaCoO3, stoichiometric concentrations of
catalyst by immersing it in a 1 : 3 slurry of nitric acid
barium nitrate (Ba (NO3)2) and cobalt nitrate hexahydrate
and deionized water. The amount of perovskite
(Co (NO3)2.6H2O) are dissolved in deionized water and
catalyst powder added was proportional to the acid
stirred continuously. In just this solution, a concentration of
used. This was dried for 30 minutes at 200°C before
citric acid equal to the metal ion concentration has been
calcining for 5 hours at 700°C.
dissolved. This mixture was stirred for 4-5 hours in a
magnetic stirrer until a transparent purple gel formed. The COAT 2 (Active phase growth on the substrate in situ):
release of gas from the combination illustrates this process.
(1) This type of coating was installed following the
Once the gel has been initiated, the drying and calcination
patent procedures [18]. As previously described, the
processes are conducted to obtain the catalyst.
first step in this type of coating is preparing a pe-
The drying occurred at 120°C for 24 hours before being
rovskite material using the sol-gel technique.
calcined at 850°C for 5 hours. The hot air oven’s airflow rate
had to be kept constant at 100 sccm (standard cubic (2) However, rather than proceeding with the drying
centimeters), and the calcination was done in a muffle stage, this gel is used as a dipping medium. By adding
furnace. Figure 4 depicts the formation of sol, gel, and a suitable reagent, the gel viscosity can be varied as
Perovskite. For phase identification, XRD of the syn- desired (deionized water in this case).
thesised samples was used. A Bruker axstometer (30 kV, The substrate sample is dipped into the gel wash coat and
12 mA) was used to perform X-ray Diffraction with held for one minute. This allows the gel material to fill the
wavelength radiation � 1.5418 A. Each sample’s data was pores of the cordierite. This gel-coated sample is then dried
collected for a scan range of 10 to 50 degrees and a step at 120°C. for one hour before calcining at 700°C. for five
width of 0.02 degrees. The measured XRD data corresponds hours. The gel enters the pore of the substrate and is cal-
to the standard data from the Crystallography Open. cined, allowing high surface perovskite to grow onto the
Database (COD). The contaminants BaCoO3 (COD Id: substrate.
1520993), SrCoO3 (COD Id: 1551939), Ba0.5Sr0.5CoO3 SEM images of coated samples are shown in Figure 6.
(COD Id: 7221487), and Co3O4 (COD Id: 1526734) match Uncoated cordierite is highly porous, with a mean wall
the standard data from the COD database. In some cases, the thickness of 0.8 mm and pores with a mean diameter of
XRD analysis shown in Figure 5 detects the presence of 35 um.
metal carbonates and Co3O4 (spinel). At 850°C, XRD Only when these pores are filled with the coating ma-
confirms that the sample does have a crystalline perovskite terial will the adhesion of the applied coat be satisfactory.
(hexagonal) structure. Because of the multilayered coating, COAT 1 has a smaller
channel volume than the other two types of layers. This may
result in a more significant pressure drop across the exhaust
2.3. Coating Procedure. Substrate activation techniques are
line. The COAT 2 depicts the perovskite material’s in-situ
the most commonly used methods for coating a substrate.
growth. Compared to COAT 1, this has better adhesion and
The different kinds of substrate activation are discussed
a more prominent channel volume.
extensively in [23]. Structure at 850°C.
COAT 1 appears to wither from the substrate in Figure 7
COAT 1 (support deposition followed by active phase
due to higher mass deposition (3 layers). Each time the
deposition):
substrate sample was dipped into the slurry, the substrate’s
(1) Alumina was chosen as a wash coat because it has a mass increased. COAT 2 can be seen in Figure 7 to have a
large surface area. Alumina (Qualigen) with a par- flaky layer formed over the substrate, demonstrating better
ticle size of 67 um was milled for 20 hours in a high- bonding than that of the three-layered coating. Calcination
energy planetary ball mill. Following the procedure of Perovskite has occurred within the pores of cordierite and
outlined in the research study [24], an alumina slurry results in a flaky layer visible in the SEM images. Figure 8
was formed. Formalised paraphrase deionised water depicts the EDAX report for uncoated cordierite, as well as
was mixed with alumina (30% wt), PEG 1000 (2%), COAT 1 and COAT 2.
and alumina sol (4% wt), pH � 4. In some cases, The morphology of each coating is depicted in both SEM
glacial acetic acid (20 mol l-1) was added to improve images in Figure 9. The COAT 2 layer is more refined than
slurry stability [25]. The substrate was dipped for a the COAT 1 layer. In addition, the addition of Ceria to LNT
few seconds in this slurry before being dried and resulted in the formation of bulk sulphate species [26]. The
calcined to increase the necessary alumina weight. nanoparticles in COAT 2 are formed as a result of the sol-gel
Advances in Materials Science and Engineering 5

(a) (b) (c)

Figure 4: (a) Sol formation, (b) gel formation, and (c) perovskite.
Intensisty (counts)

BaCoO3

0 10 20 30 40 50
2theta
Measured profile

Ba0.5Sr0.5CoO3
Intensisty (counts)

0 10 20 30 40 50
2theta
Measured profile
Intensisty (counts)

SrCoO3

10 20 30 40 50
2theta
Measured profile

Figure 5: XRD Peaks of synthesised perovskites.

6 Advances in Materials Science and Engineering

(a) (b)

(c)

Figure 6: SEM images cordierite. (a) Uncoated cordierite. (b) First coating. (c) Second coating.

(a) (b) (c)

Figure 7: Structures of (a) COAT 1 and (b) COAT 2.

Perovskite being calcined. When this nano size comes into to the substrate. An ultra sonicator apparatus was used to
contact with exhaust gases, it can increase catalytic activity. test the adherence of both sample coatings. For 30 minutes,
This smaller particle size also contributes to good adhesion both sample coated cordierites were immersed in a sonicator
Advances in Materials Science and Engineering 7

Selected Area 1 ‑ Det 1

612 Element Weight % Atomic % Error %

544 OK 48.79 61.72 7.58
476 MgK 3.97 3.31 11.29
AI AIK 31.42 23.57 5.16
408
SiK 15.82 11.40 8.14
340

272
Si
204
136
Mg
68 Si
0
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Lsec: 30.0 0 Cnts 0.000 keV Det. Element‑C2B

Selected Area 1‑Det 1

44.0
39.6 Element Weight % Atomic % Error %
35.2 BaL 55.99 35.31 43.09
30.8 CoK 44.01 64.69 34.59

26.4

22.0 Co
17.6
13.2
8.8 Ba Ba Ba
4.4 Ba Co
0.0
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Lsec: 30.0 0 Cnts 8.670 keV Det. Element‑C2B

Selected Area 1‑Det 1

48.6 Element Weight % Atomic % Error %

AIK 9.15 23.78 19.66
43.2
SrL 30.60 24.48 15.82
37.8 BaL 7.72 3.94 56.29
32.4 CeL 21.32 10.66 37.62
27.0 CoK 31.22 37.13 21.08

21.6
16.2 Ba
Co
10.8
Co
5.4
0.0
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Lsec: 30.0 1 Cnts 8.670 keV Det. Element‑C2B
Figure 8: EDAX details of cordierites. (a) Uncoated. (b) COAT 1. (c) COAT 2.

water bath. The morphology of each coating is represented in When this nano size comes into contact with exhaust
both SEM images in Figure 9. The COAT 2 layer is more gases, it can increase catalytic activity. This smaller particle
refined than the COAT 1 layer. In addition, the addition of size also helps with adhesion to the substrate. An Ultra
Ceria to LNT leads to the formation of bulk sulphate species Sonicator apparatus was used to test the adherence of both
[26]. The nanoparticles in COAT 2 are formed due to the sol- sample coatings. For 30 minutes, both sample coated cor-
gel Perovskite’s calcination temperature. dierites were immersed in a sonicator water bath. On the
8 Advances in Materials Science and Engineering

(a) (b)

Figure 9: Structure of (a) COAT 1 and (b) COAT 2.

other hand, a shaker table can be used to vibrate the coated

substrate at a specific frequency to measure the weight loss
due to vibration. Before and after sonicator use, the weights
of both coated samples were measured. Samples were
measured. Micro Crack
After sonication, COAT 1 lost 22 percent of its weight,
while COAT 2 lost only 4 percent. This gravimetric analysis
was carried out using a highly accurate analytical balance
(accuracy of 0.001 gms). This is due to the numerous
microcracks in COAT 1, as shown in Figure 10. The main
result of the adherence tests revealed that layer COAT 1 had
a few disadvantages compared to COAT 2. However, due to
its excellent adhesion quality, COAT 2 was chosen for the
next stage of emission tests. Three cordierite substrates
(cylindrical 400 cpsi) with dimensions of 90 mm dia. and Figure 10: Microcracks in COAT 1.
150 mm length were determined based on the engine to
perform emission tests in real-time. These substrates were
coated using the COAT 2 procedure. The coated substrates control the Lean-Rich injection cycle. The engine-out NOx
were coated with perovskite materials synthesised and emission from this modified engine was measured under
tested, which included BaCoO3, Ba0.5Sr0.5CoO3, and various load conditions. Figure 12 depicts the experimental
SrCoO3. schematic diagram of engine setup and Table 1 shows the
specification of the engine used for the experiment.
The first set of measurements was taken without elec-
3. Materials and Methods tronic injection, i.e., with the engine as it was. Various
3.1. Fabrication of the LNT, Experimental Setup, And Baseline conditions were used to record the emission and perfor-
Reading. Following the appropriate procedures, the coated mance outcomes. Table 2 shows the specifications of the IC
cylindrical substrates were canned. The fabricated shell engine and the electronic fuel injector used in this
volume was kept constant at the same level as the dis- experiment.
placement volume of the engine used. To prevent cordierite
damage due to thermal expansion, an Alumina Fibre mat 4. Results and Discussion
was used at the contact area between the cordierite and steel,
as shown in Figure 11. An emission analyser was used to The engine was first run under various load conditions to
measure engine-out emissions. This experiment modified obtain baseline readings, and then the process was repeated
the Kirloskar AV1, a nonautomotive stationary engine, to to ensure no errors were present. Finally, the fabricated LNT
Advances in Materials Science and Engineering 9

Alumina

Fiber Mat

Substrate

Figure 11: Fabrication images with Alumina support mat.

FUEL
PUMP

AVL FIVE GAS

FUEL INJECTOR ANALYZER
TANK
Lean NO2 Trap
ENGINE
EXHAUST
ECU

ENGINE
LOAD
CONTROLLER
DYNAMOMETER
CURRENT
EDDY

DATA ACQUISITION &

INJECTION CONTROL

Figure 12: Experimental schematic of Lean NOx Trap.

Table 1: IC engine specifications.

Type 4S diesel (1 Cyl.)
Bore 80 mm
Stroke 110 mm
Compression ratio 17.5 : 1
Maximum power 3.7 kW
Speed 1500 rpm
Loading device Eddy current dyno
Mode of starting Manually cranking
Mechanical injector pressure 220 bar
Electronic injector pressure 1000 bar
Injection timing 23° BTDC
10 Advances in Materials Science and Engineering

Table 2: Electronic fuel injector specifications.

Injection pressure 1000 bar
No. of nozzle holes 3
Nozzle diameter 0.518 mm
Start of injection 23° BTDC
Injection duration 650–3000 μsec

70 80

60
60
50

HRR (J/ CA)

40
Load (N‑m)

40

30 20

20
0
10

0 –20

–400 –300 –200 –100 0 100 200 300 400 –400 –300 –200 –100 0 100 200 300 400
Crank angle (theta) Crank angle (theta)

0 N‑m 15 N‑m 0 N‑m 15 N‑m

5 N‑m 20 N‑m 5 N‑m 20 N‑m
10 N‑m 10 N‑m
(a) (b)

Figure 13: Curves for different load conditions with mechanical injection. (a) Load and crank angle curve. (b) HRR and crank angle curve.

device was attached to the exhaust end, and the emission These released oxygen molecules undergo oxidation with
reading for neat Diesel without electronic injection was hydrocarbons and carbon monoxide. However, due to the
recorded. The following observations were made using the poor atomization of the fuel, PJ 15 biodiesel emits an overall
AVL DIGAS 444N Gas Analyser. higher level of HC.
Only NO is recorded because the equipment detects NO Nitrogen oxide reduction mechanisms are slower in low
using an electrochemical sensor. The following are the load conditions than in high load conditions. This leads to
emission results with a BaCoO3 catalyst LNT on a me- lower oxygen release in the exhaust, which has little or no
chanical injection engine. The different loading values and impact on HC emissions under common load conditions.
corresponding emission values were plotted. Figure 13 de- CO emission levels are depicted in Figure 16. CO emissions
picts the P-Theta and HRR curves. The experiment was can be reduced slightly during high load conditions. The
carried out with Neat Diesel and Prosopis-juliflora bio-oil [a oxygen molecules released during NOx decomposition
blend of 85 percent diesel and 15 percent juliflora oil], undergo oxidation of carbon monoxides to carbon dioxide,
denoted as PJ 15, to learn about the impact of LNT on a resulting in a slight increase in CO2 emissions proportional
biofuel. NOx emissions plummet under high load condi- to the decrease in CO emissions.
tions. Because of incomplete combustion, NOx levels in Because of the different fuels used (diesel & PJ15), the
biodiesel PJ15 are even lower [13]. Figure 14 depicts the NOx exhaust gas temperature varies in Figures 14–16, and as a
level at load variations. result, emissions cannot be measured using temperature as a
The main engine modification was the electronic in- reference. As a consequence, the load is being used as the X-
jection control, which was used to control the alternating axis parameter. The impact of LNT causes a rise in CO
lean-rich cycling required for the lean NOx trap. Various emissions in the case of PJ15 biofuel, but the levels are not as
injection cycles were tested, and the best injection strategy high as they are in neat diesel. However, when compared to
was a lean-rich cycle of 120 seconds (lean) to 20 seconds neat diesel, the CO levels are higher due to the poor at-
(rich). As shown in Figure 15, BaCoO3-LNT significantly omization of the biofuel. Collected on the basis of all of these
reduced HC emission levels (with LNT) at high load con- readings, the engine was modified, and the effect of this
ditions while having no effect at low load conditions. The modification is depicted in Figure 17.
main reason for this could be the lower levels of nitrogen This injection cycle was used to collect baseline readings.
oxides. The rich mixture converts NOx to Nitrogen and The engine was run at 40% load (average out NOx of
Oxygen in high load conditions. 740 ppm and a catalyst temperature of 240°C), and the NOx
Advances in Materials Science and Engineering 11

1000

Oxides of Nitrogen (ppm)

800

600

400

200

0
0 5 10 15 20 25
Load (N‑M)

(without LNT) Neat diesel with LNT ( T = 350 C )

with LNT ( T < 240 C ) PJ 15 with LNT ( T = 350 C )
Figure 14: NOx emission curve for Diesel and PJ15 with LNT.

50

45

40
HC (ppm)

35

30

25

20
0 5 10 15 20 25
Load (N‑m)

without LNT Neat diesel with LNT T = 350 C

with LNT T < 240 C PJ 15 with LNT T = 350 C
Figure 15: HC emission for Diesel and PJ15 with LNT.

reading was recorded as a function of time. As a result of the Figure 19 depicts the average NOx reduction achieved in the
rich phase, the injection cycle resulted in periodic NOx same engine with and without a lean NOx trap. The LNT
reduction. reduced engine-out nitrogen oxides with a conversion ef-
Figure 18 depicts NOx absorption and regeneration. All ficiency of 63.5 percent at 240°C. As shown in Figure 20, at
three catalysts exhibit comparable absorption and reduction, 40% engine load, the engine-out NOx is 740 ppm. On the
but BaCoO3 absorbs that much more NOx than SrCoO3. other hand, the LNT-Out NOx at 240 C is 270 ppm. This
This is due to barium’s more electropositive behaviour in the NOx reduction accounts for 63.5 percent of the conversion
past when compared to strontium. In addition, due to its efficiency. If the temperature is raised to 350°C, the LNT-Out
larger size when compared to strontium, barium (Atm no. NOx drops to 118.4 ppm, resulting in a peak conversion
56) quickly loses its outermost electron (Atm no. 38). Fi- efficiency of 84 percent. LNT’s performance in this range is
nally, as previously stated, the metal nitrate or nitrite species optimal, and it can be considered excellent at low tem-
formed during absorption significantly impact trapping peratures. However, after 170°C, there is a noticeable change
capacity. caused by the attainment of the light-off temperature. As a
The release of trapped NOx and incoming engine-out result, the conversion efficiency increases with temperature.
nitrogen oxides cause a sudden spike in NOx emission in the The percentage of conversion is calculated as follows:
exhaust at the start of the regeneration phase. As stated Stored NOx (ppm) � total NOx in−total NOx out.
previously, the metal nitrate or nitrite species formed during NOx conversion (percent) � 100 × (total NOx in−total
absorption significantly impact absorption capacity. NOx out)/total NOx.
12 Advances in Materials Science and Engineering

0.20

0.15

Carbon Monoxide (%)

0.10

0.05

0.00

0 5 10 15 20 25
Load (N‑m)

without LNT Neat diesel with LNT T = 350 C

with LNT T < 240 C PJ 15 with LNT T = 350 C
Figure 16: CO emission for Diesel and PJ 15 with LNT.

Fuel Used: Neat Diesel

750

600
NOx (ppm)

450

300

150

0
30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time (sec)
Engine Out NOx

750

600
NOx (ppm)

450

300
Rich

Rich

Rich

150
Lean Lean Lean

0
30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time (sec)

Engine Out NOx with A/F eyeling

Figure 17: Engine-Out NOx with cyclic injection.

Advances in Materials Science and Engineering 13

Fuel Used: PJ 15

Lean Rich Lean Rich Lean Rich

Engine Out
750 NOx
740 ppm

600
NOx (ppm)

450

300

150

0
30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time (sec)

NOx stored by
SrCoO3
NOx stored by
Ba0.5Sr0.5CoO3
NOx stored by
baCoO3

Figure 18: Average NOx reduction with LNT at 240°C.

Fuel Used: PJ 15
100

84 LNT based on
83 80 the catalyst
80 78 75
72 temperature
NOx conversion (%)

63.5
60 5462

40

18
20 1417
1012
9
0
0 125 170 240 295 350
Temperature (C)

SrCoO3
Ba0.5Sr0.5CoO3
BaCoO3

Figure 19: NOx conversion efficiency.

14 Advances in Materials Science and Engineering

Engine Lead‑40% Catalyst Used : BaCoO3

Fuel Used : PJ 15

750

600
NOx (ppm)

450

300

150

0
30 60 90 120 150 180 210 240 270 300 330 360 390 420
Time (sec)
Figure 20: NOx stored in Lean phase and reduced in the rich phase by three perovskite catalyst.

Conversion efficiencies of up to 84 percent were achieved developed if subjected to the artificial thermal aging test, and
at higher temperatures (350°C). the long-term durability can be analysed.
Figure 19 depicts the variation of LNT efficiency con-
cerning temperature.
Data Availability
5. Conclusions The data used to support the findings of this study are in-
cluded in the article.
The most important part of this research work is to find a
suitable replacement for Pt in a Lean NOx trap. Three
catalysts were selected based on a literature survey and Conflicts of Interest
synthesised by proper procedures. The final output using the
BaCoO3 catalyst in an LNT reactor shows high NOx con- The authors declare that there are no conflicts of interest
version efficiency but not as high as platinum-based LNT. regarding the publication of this study.
Therefore, the low-cost catalyst will serve as a suitable re-
placement but has certain limitations apart from conversion Authors’ Contributions
efficiency.
The selected catalyst was the cheapest among the three, Chinnadurai Kombiah (CK), Bharathan Rajappa (BR), and
and the coating, COAT 2, showed higher adherence with a Senthil Kumar Pachamuthu (SP) have analytically drafted this
mass loss of only 4% during sonication. work for important intellectual content. CK analysed and
Regarding the emission tests, at low load conditions, the gathered technical information about the lean NOx trap (LNT),
NOx decomposition percentage is minimum and erratic. developed methods and process flows to achieve the research
There occurs no gradual rise in NOx decomposition with a goal, acquired the resources necessary, performed coating
stage-wise increase in loads. However, at high load condi- processes and related tests such as XRD, SEM, and Edax, and
tions, the increase in the NOx decomposition percentage is fabricated the LNT reactor according to design calculations
very high. NOx conversion obtained was 63.5% at 240°C and based on the engine specification. BR assisted CK in the design
increased up to 84% at 350°C in the experiment. and fabrication of the LNT reactor and also in the synthesis of
The impact of LNTon other emissions was also recorded. the LNTcatalyst and the collection of NOx reduction reactions.
HC and CO values dropped due to the oxidation of the SP supported CK in collecting emissions testing results and was
oxygen molecule released during nitrogen oxide break up. also instrumental in the development of electronic injection in
No considerable change in smoke emission was observed. the engine setup. SP carefully analysed the data of engine tests
Long-time usage during the experiment showed a drop in to evaluate engine performance and operation for various
performance of the LNT, which could be due to Sulfur poi- injection schemes (with and without the LNT). The final
soning of the catalyst material. The project has a wide range of manuscript has been read and approved by all the authors [27].
future scope. The Coatings studied were selected only based on
their adherence. Since COAT 1 is an alumina-perovskite LNT,
References
it showed the formation of aluminates in some instances. This
hinders the reduction reactions. This could be researched, and [1] A. Clenci, V. Sǎlan, R. Niculescu, V. Iorga-Simǎn, and
methods to prevent this can be developed. The LNT reactor C. Zaharia, “Assessment of real driving emissions via portable
Advances in Materials Science and Engineering 15

emission measurement system,” IOP Conference Series: Ma- using LaCoO3 perovskite as active phase,” Journal of Physical
terials Science and Engineering, vol. 252, no. 1, 2017. Chemistry C, vol. 118, no. 44, Article ID 25403, 2014.
[2] H. Dinodia, N. Mahajan, V. Dwivedi, and V. Khanna, “A [20] V. G. Milt, M. A. Ulla, and E. E. Miró, “NOx trapping and soot
study on real driving emissions for India - an experimental combustion on BaCoO3−y perovskite: LRS and FTIR char-
approach,” SAE Technical Papers, vol. 2018, pp. 1–9, 2018. acterization,” Applied Catalysis B: Environmental, vol. 57,
[3] United states environmental protection agency, “Nitrogen no. 1, pp. 13–21, 2005.
dioxide pollution,” 2019, https://www.epa.gov/no2-pollution. [21] L. Predoana, B. Malic, M. Kosec, M. Carata, M. Caldararu, and
[4] S. K. Hoekman and C. Robbins, “Review of the effects of M. Zaharescu, “Characterization of LaCoO3 powders ob-
biodiesel on NOx emissions,” Fuel Processing Technology, tained by water-based sol-gel method with citric acid,” Journal
vol. 96, pp. 237–249, 2012. of the European Ceramic Society, vol. 27, no. 13–15,
[5] S. Mohankumar and P. Senthilkumar, “Particulate matter pp. 4407–4411, 2007.
formation and its control methodologies for diesel engine: a [22] A. Montebelli, C. G. Visconti, G. Groppi et al., “Methods for
comprehensive review,” Renewable and Sustainable Energy the catalytic activation of metallic structured substrates,”
Reviews, vol. 80, pp. 1227–1238, 2017. Catalysis Science and Technology, vol. 4, no. 9, pp. 2846–2870,
[6] J. B. Heywood, Internal Combustion Engine Fundamentals, 2014.
McGraw-Hill, New York, NY, USA, 1988. [23] N. K. Labhsetwar, A. Watanabe, and T. Mitsuhashi, “New
[7] A. Jain, A. P. Singh, and A. K. Agarwal, “Effect of split fuel improved syntheses of LaRuO3 perovskites and their appli-
injection and EGR on NOx and PM emission reduction in a cations in environmental catalysis,” Applied Catalysis B:
low temperature combustion (LTC) mode diesel engine,” Environmental, vol. 40, no. 1, pp. 21–30, 2003.
2017. [24] S. Zhu, W. Zhu, W. Qian, Y. He, and J. Huang, “A China -
[8] M. T. Chaichan, “Performance and emission characteristics of vietnam collaboration for public health care: a preliminary
CIE using hydrogen, biodiesel, and massive EGR,” Interna- study,” Global Health Research and Policy, vol. 4, no. 1,
tional Journal of Hydrogen Energy, vol. 43, no. 10, pp. 5415– pp. 23–27, 2019.
5435, 2018. [25] L.-H. Tsui, V. Stetsovych, B. Pereda-ayo, and J. R. González-
[9] A. Parlak, V. Ayhan, Y. Üst et al., “New method to reduce velasco, “Experimental control of Ce3+ concentration in ceria
NOx emissions of diesel engines: electronically controlled based model catalysts,” Diesel Engine: Combustion, Emissions
steam injection system,” Journal of the Energy Institute, and Condition Monitoring, vol. x, p. 266, 2015.
vol. 85, no. 3, pp. 135–139, 2012. [26] E. Megiddo, “United States C12 ) Patent Application Publi-
[10] M. Dalla Nora, T. D. M. Lanzanova, and H. Zhao, “Effects of cation Correspondence Address: (Issue 19),” 2008.
valve timing, valve lift and exhaust backpressure on perfor- [27] D. Kannan, S. Pachamuthu, M. Nurun Nabi, J. E. Hustad, and
mance and gas exchanging of a two-stroke GDI engine with T. Løvås, “Theoretical and experimental investigation of diesel
overhead valves,” Energy Conversion and Management, engine performance, combustion and emissions analysis
vol. 123, pp. 71–83, 2016. fuelled with the blends of ethanol, diesel and jatropha methyl
[11] N. Takahashia, H. Shinjoh, T. Iijima et al., “The new concept ester,” Energy Conversion and Management, vol. 53, no. 1,
3-way catalyst for automotive lean-burn engine: NOx storage pp. 322–331, 2012.
and reduction catalyst,” Catalysis Today, vol. 27, no. 1–2,
pp. 63–69, 1996.
[12] H. Mahzoul, J. F. Brilhac, and P. Gilot, “Experimental and
mechanistic study of NOx adsorption over NOx trap cata-
lysts,” Applied Catalysis B: Environmental, vol. 20, no. 1,
pp. 47–55, 1999.
[13] M. A. Peña and J. L. Fierro, “Chemical structures and per-
formance of perovskite oxides,” Chemical Reviews, vol. 101,
no. 7, pp. 1981–2017, 2001.
[14] P. Fanson, M. R. Horton, W. N. Delgass, and J. Lauterbach,
“FTIR analysis of storage behavior and sulfur tolerance in
barium-based NOx storage and reduction (NSR) catalysts,”
Applied Catalysis B: Environmental, vol. 46, no. 2, pp. 393–
413, 2003.
[15] B. Westerberg and E. Fridell, “A transient FTIR study of
species formed during NOx storage in the Pt/BaO/Al2O3
system,” Journal of Molecular Catalysis A: Chemical, vol. 165,
no. 1–2, pp. 249–263, 2001.
[16] P. Forzatti, L. Castoldi, I. Nova, L. Lietti, and E. Tronconi,
“NOx removal catalysis under lean conditions,” Catalysis
Today, vol. 117, no. 1–3, pp. 316–320, 2006.
[17] J. Zhao, Z. Li, S. Li, S. Shuai, S. Liu, and L. Cao, “The effects of
exhaust gas recirculation on NOx storage pathway,” E3S Web
of Conferences, vol. 268, no. 01017, pp. 1–23, 2021.
[18] J. E. De Abreu Goes, A. Kristoffersson, and L. Olsson, “Sulfur
poisoning effects on modern lean nox trap catalysts com-
ponents,” Catalysts, vol. 9, no. 6, pp. 492–498, 2019.
[19] R. You, Y. Zhang, D. Liu et al., “YCeZrO ternary oxide solid
solution supported nonplatinic lean-burn NOx trap catalysts