Fuel 276 (2020) 118120

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Effect of hydrogen and multiwall carbon nanotubes blends on combustion T

performance and emission of diesel engine using Taguchi approach
⁎ ⁎
S. Manigandana,b, , A.E. Atabanic, Vinoth Kumar Ponnusamyb,d,e, , Arivalagan Pugazhendhif,
P. Gunasekara, S. Prakasha
a
Department of Aeronautical Engineering, Faculty of Engineering,Sathyabama Institute of Science and Technology, Chennai City, Tamil Nadu, India
b
Department of Medicinal and Applied Chemistry, Kaohsiung Medical University, Kaohsiung City-807, Taiwan
c
Alternative Fuels Research Laboratory (AFRL), Energy Division, Department of Mechanical Engineering, Faculty of Engineering, Erciyes University, Kayseri 38039, Turkey
d
Research Center for Environmental Medicine, Kaohsiung Medical University, Kaohsiung City-807, Taiwan
e
Department of Medical Research, Kaohsiung Medical University Hospital, Kaohsiung City-807, Taiwan
f
Innovative Green Product Synthesis and Renewable Environment Development Research Group, Faculty of Environment and Labour Safety, Ton Duc Thang University, Ho
Chi Minh City, Viet Nam

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO A B S T R A C T

Keywords: The analysis of optimum engine conditions operating under hydrogen and multiwall carbon nanotubes
Hydrogen fuel (MWCNTs) blends using the Taguchi L16 approach were examined. The effect of hydrogen (10%, 20%, and 30%)
Diesel and MWCNTs (30 ppm, 50 ppm, and 80 ppm) with different fuel blend proportions were evaluated at different
Multiwall carbon nanotubes engine loads of 25%, 50%, 75%, and 100%. Further, the influence of ignition pressure was evaluated at four
Combustion
intervals of 180 bar, 200 bar, 220 bar, and 240 bar. In addition, the ignition timing varied at different intervals
Engine emission
of 210bTDC, 230bTDC, 270bTDC and 310bTDC, respectively. TaguchiL16 approach was designed according to
the combinations of the array, including load, hydrogen, MWCNTs, ignition pressure, and timing with four
optimal conditions. Results revealed that the L16 orthogonal array was a suitable method to find the optimized
working conditions of the diesel engine for higher performance and less emission of NOx, HC, CO, and CO2.
Compared to neat diesel, the addition of hydrogen and MWCNTs reduces the emission with improved brake

⁎
Corresponding authors at: Department of Medicinal and Applied Chemistry, Kaohsiung Medical University, Kaohsiung City-807, Taiwan.
E-mail addresses: [email protected] (S. Manigandan), [email protected] (V.K. Ponnusamy), [email protected] (A. Pugazhendhi).

https://doi.org/10.1016/j.fuel.2020.118120
Received 16 March 2020; Received in revised form 13 May 2020; Accepted 15 May 2020
Available online 23 May 2020
0016-2361/ © 2020 Elsevier Ltd. All rights reserved.
S. Manigandan, et al. Fuel 276 (2020) 118120

Nomenclature NOx Nitrogen oxide

CO2 Carbon dioxide
CI Compression ignition IP Injection pressure
DI Direct injection H2 Hydrogen
bTDC before Top dead centre DC Direct current
CA Crank angle DAS Data acquisition system
BP Brake power SNR Signal to noise ratio
BTE Brake thermal efficiency MM Means of means
BSFC Brake specific fuel consumption LSN Lesser signal noise ratio
HC Hydrocarbons NSN Normal signal noise ratio
CO Carbon monoxide LSN Lesser signal noise ratio

thermal efficiency. The effects of the optimal factors were further analyzed using ANOVA. From the ANOVA
results it is evident that, the addition of hydrogen and MWCNT advances the performance and emission char-
acteristics. Moreover, results showed improved brake power by 13% and reduced brake specific fuel con-
sumption by 8% at full engine load conditions. 22%, 20%, 8%, and 47% emission reductions of HC, NOx, CO,
and CO2 were observed due to the influence of hydrogen and MWCNTs.

1. Introduction additive microalgae to enhance the performance of the engine. From

results, the small percentage of the algal content enhances the engine
Fossil fuels were one of the main resources powering the world to performance by stimulating the carbon atom formation and poly un-
satisfy the demands of energy. The primary energy consumption of the saturation[11]. Vivek et al. performed a review on nanofluid in diesel
world increases by around 18% in the past decade. Although the de- engine powered by biodiesel blends. From the crisp review, they
mand for energy is severe worldwide, populated countries like India eventually found that the addition of nanoparticles improves the caloric
and China are undergoing serious threats compared to other developed value and cetane number of the biodiesel blends which is superfluous
countries [1]. Climate change, air pollution, and the price of fossil fuel behaviour. Despite the improvement in caloric value, there is a slight
have also stimulated the exploitation of renewable energy. Despite increase in the viscosity, density and flashpoint. Having said that, na-
several studies on renewable energy like solar, wind, and hydropower, noparticles also enhances the heat transfer and catalytic activity rate. It
the biodiesel gained profound attention among the scientific commu- is fascinating to see, most of the nano-additives record less NOx, CO, HC
nity due to its power to weight ratio and thermal efficiency. In recent and CO2 production. In regard to smoke, a positive effect is observed
days, nanoparticles and catalytic converter were used to reduce the which is owing to the higher evaporation rate and catalytic oxidation.
pollution from the diesel engine. However, nanoparticles gained lime- Another key factor to remember is concentration. From the mentioned
light of the audience in recent year’s especially metal-based nano- review it is quite obvious; a higher concentration of nanoparticles in-
particles owing to the better property [2]. The addition of metallic creases the CO emission. In addition, employing NP in diesel engines
based nanoparticles increases combustion and reduces the release of improves cylinder pressure, heat release rate and evaporation [12]. Ooi
harmful gases by acting as a catalyst. The main reason for nanoparticles et al added at 0.1% and 0.01% of graphite nanoparticles, aluminium
to reduce emission is high surface area to volume ratio. As the surface oxide and cerium oxide to enhance combustion characteristics of the
area to volume ratio is increased, the combustion is improved by en- diesel engine. Adding nanoparticles improves the preheating, ignition
hancing the contact between fuel and oxidizer. In addition, the eva- and unsteady burning. Further, the parameters such as burn rate, micro
poration rate and ignition delay also shortened [3]. Nanoparticles can explosion and burnout times are also improved with reduced CO, NOx
act as the secondary carrier in addition to the blends to stimulate the and HC. From the series of experiments, they proved graphene oxide
combustion and performance of the engine [4]. Nanoparticles have (GO) was a promising additive than Al2O3 and Ce2O3 [13]. El-seesy
some unique properties to improve the physical properties of the fuel to et al. studied the effect of MWCNT on jojoba methyl ester. The blends
enhance combustion and reduce emissions. Finding the optimized one are added at the ratio of 10, 20, 30, 40, and 50 mg/L, tested at different
is the key thing [5]. Recently, Al2O3, CuO, MgO, Fe, Ce2O3, CuO and engine levels. The addition of MWCNT decreases the ignition delay and
CNT are examined as the effective particles to enhance the combustion advances the peak heat release rate occurrence. Further, the BTE is
and reduce emission. The addition to the above nanoparticles increases increased 10% with 15% reduction in BSFC. Meanwhile, the CO, NOx
the energy density of the liquid fuel which improves the combustion
[6]. However, the efficiency of the nanoparticles is affected due to the
stability and preparation procedure of the samples. Usually, two pro- Table 1
cedures (stirring and utilization of fine nanoparticles) are commonly Specification of the test engine.
used to achieve high stability for nanoparticles to disperse in diesel Parameters Specifications
uniformly [7]. The utilization of fine nanoparticles can be dispersed
very fast and effective, and simple method than the stirring process. The General details Single cylinder, 4 S, constant speed compression ignition
engine, Dual combustion
stirring is usually done by ultrasonic, manual and magnetic stirring.
Cool Air cooled
Nithya et al. proved that larger stirring time leads to higher stability of Bore 86 mm
nanoparticles [8]. Although stability is improved but as time runs out, Stroke 106 mm
the stability is decreased, this requires an extensive improvement [9]. Swept volume 616.56 cc
Despite the physical method, the chemical methods also employed to Compression ratio 16.5:1
Injection Pressure 240 bar
stabilize the fuel. pH is the variable that is responsible for the stability Fuel injection Piston pump
of the nanoparticles [10]. Several experiments investigated the effect of Injector Solenoid controlled
nanoparticles added to the diesel fuel. Rahman et al. used oxygenated Rated speed 1800 rpm

2
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 1. Schematic experimental setup [6].

Table 2 Table 4
Properties of fuel. Factors and their levels.
Specification Diesel Hydrogen Symbol Factors Stage 1 Stage 2 Stage 3 Stage 4

Density at ambient (kg/m3) 824 0.081 A Engine load (%) 25 50 75 100

Lower heat value (MJ/kg) 42.5 120 B Hydrogen (%) 0 10 20 30
Higher heat value (MJ/kg) 44.8 141.9 C Nanoparticles (ppm) 0 30 50 80
Specific gravity 0.83 0.091 D Ignition pressure (bar) 180 200 220 240
Cetane number 44–55 0 E Ignition timing (0bTDC) 21 23 27 31
Carbon content (wt%) 86 0
Hydrogen content (wt%) 13 100
Mole weight (g) 148.3 1.92
experiment period [15]. Najafi et al. performed an analysis of CNT and
Auto-ignition (K) 524 854
Octane number 30 130
Ag nanoparticles at different 40–120 ppm using ultrasonication.
Moreover, the heat release rate for the engine increased by the addition
of nanoparticles. This was mainly due to the addition of extra oxygen to
and UHC also reduced significantly. From the test, it is proven 50 mg/L the fuel by means of nanoparticles. Further, the addition of NP im-
dosage is the optimum for high performance and less emission rate proves the combustion phase and drop the combustion duration. In
[14]. Followed by the above study, El-seesy et al. tested the impact of addition to above, 15% increases in peak pressure also observed and 8%
jatropha and graphene oxide blends on the diesel engine. From the reduced ignition delay. Another key fact is CO and HC is reduced. On
series of tests, it is persuaded, the addition of GO to jatropha blends the contrary, no evidence of NOx and CO2 is reported than diesel fuel
increases both peak cylinder and pressure rate. Further, the addition of [16]. Örs et al. tested the TiO2 on waste cooking oil to measure the
GO enhances the combustion process by shortening the ignition delay. effect in a diesel engine at different blends. The addition of TiO2 affects
The BTE improves by 20% with reduced BSFC of 17% due to the in- the physicochemical property of the fuel. On another end, n-butanol has
fluence of GO nanoparticles. Meanwhile, 60%, 15% and 50% reduction a powerful effect on density, kinematic viscosity, pour point, cloud
in CO, NOx and UHC were reported. Another key thing to remember is point than TiO2. Thus TiO2 considered an effective secondary energy
the deposition of carbon on the diesel engine is negligible during the carrier. However, butanol carries decent cold flow properties than

Table 3
Accuracy of measurements and uncertainties.
Device Used Parameter Measuring Range Resolution Accuracy Maximum Uncertainty

Fuel measuring burette Fuel flow [g/h] 0–100 ml – ± 1 ml ± % 0.5

Electronic precision balance Fuel consumption [g] 0–99 kg/hr – ± 0.02 kg/hr ± % 0.1
Pressure Sensor In cylinder pressure [bar] 0–250 bar – ± 1 bar ± % 0.6
Temperature indicator (K-Type) Temperature [°C] 0–1200 °C – ± 1°C ±% 1
AVL DI Smoke meter 0–100% – ≤ 0.01 ± % 0.1
Testo 350 exhaust gas analyzer CO 0–50 vol% 0.01 vol% ± 0.3%
CO2 0–50 vol% 0.01 vol% ± 0.3
HC 0–40000 ppm 1 ppm ± 10% of reading ± 0.1
NOx 0–3000 ppm 1 ppm ± 5% reading < 2000 ppm ± 0.5%
Data acquisition system National instruments (6218 DAS) – Equipped with one touch

3
S. Manigandan, et al. Fuel 276 (2020) 118120

Table 5 Although the nanoparticles enhance the combustion and reduce the
L16 Orthogonal array for experiments. emission, but the reduction in NOx is not very convincing when they
Expt.Runs Engine load Blends Nanoparticles I.P (bar) I.T(0bTDC) treated with biodiesel blends due to high viscosity. The main factor of
high NOx emission is high cylinder temperature and surplus oxygen in
1 1 1 1 1 1 the combustion chamber. For a diesel engine, the addition of hydrogen
2 1 2 2 2 2
increases the mixing time of A/F. Thus the improved combustion is
3 1 3 3 3 3
4 1 4 4 4 4
possible with reduced formation of soot [19,20]. The main drawback of
5 2 1 2 3 4 adding hydrogen as dual fuel to the existing diesel engine cylinder
6 2 2 1 4 3 pressure rate and combustion knock. This can be avoided by adding
7 2 3 4 1 2 MWCNT in addition to hydrogen [21]. Dimitriou et al. conducted a
8 2 4 3 2 1
comprehensive review of hydrogen as the energy carrier in the diesel
9 3 1 3 4 2
10 3 2 4 3 1 engine. Hydrogen supply tothe CI engines reduces the emission of HC,
11 3 3 1 2 4 CO and CO2 significantly. Hydrogen is a carbonless gas thus it can be
12 3 4 2 1 3 used with the diesel engine without any limitations. However, in-
13 4 1 4 2 3
creasing the ratio of hydrogen on the diesel engine increases the heat
14 4 2 3 1 4
15 4 3 2 4 1
release rate and cylinder pressure, which increases the formation of
16 4 4 1 3 2 NOx. Thus usage of EGR and nanoparticles are suggested assist to avoid
the formation of NOx with knock free operation [22]. Further, in-
creasing the hydrogen mixture at higher loads causes the advancement
Table 6 of the combustion phase leads to an increase in thermal efficiency and
L16 Orthogonal array outcome. unburnt hydrogen emission. Thereby the emission of CO2 is possible
Expt.Runs BTE BSFC HC NOx CO CO2 [23]. In the other end emission of NOx increases with an increase in
hydrogen fraction, thus the use of nanoparticles and EGR must be ob-
1 32.65 755 8.65 120 0.09 2.61 ligatory [24]. Although hydrogen and nanoparticles reduce the emis-
2 33.88 735 8.5 112 0.08 2.55
sion with good performance data, find the optimum conditions are
3 37.3 708 8 108 0.05 2.1
4 35.25 715 8.25 105 0.06 2.32 obligatory for a wide range of operations. With this in mind, Wu hw
5 33.95 662 10.8 210 0.128 4.05 et al. predicted the combustion and emission by the Taguchi method. It
6 32.15 625 10.2 235 0.125 3.95 is observed, Taguchi is the optimal method with confidence interval
7 34.35 539 9.4 210 0.12 3.8 95% with 65% less time consumption. The efficiency of the engine had
8 36.98 468 9.2 198 0.1 3.52
9 33.55 490 13.05 265 0.135 5.52
been improved for the fuel combination of 30% hydrogen, 40% EGR
10 35.12 452 12.19 265 0.149 4.32 than other parameters. Furthermore, the above combinations are more
11 33.84 485 11.95 280 0.14 4.25 promising for obtaining various parameters, including heat release rate
12 34.5 435 11.25 242 0.132 4.15 and BTE. The calculated S/N ratio configured with ANOVA to predict
13 33.56 375 14.68 365 0.158 7.25
the optimized configurations [25]. Sathiskumar et al predicted the
14 34.1 355 13.72 315 0.155 6.75
15 35.95 348 12.68 298 0.145 4.45 production of oil from Manilkara zapotal using Taguchi optimization
16 35.05 368 15.66 338 0.151 6.2 using L9 orthogonal array. From the experimental prediction, 6:1 me-
thanol to oil molar ratio considered to be the optimum concentration to
predict biodiesel production from the transesterification process. Three
metallic nanoparticles. Moreover, the addition of TiO2 improves brake trails are used and the respective results are 94.5%, 93.8% and 96.20%
power by 10% than butanol. Similarly, BSFC also showed a good re- [26]. Ayhan et al. predicted performance and emission using L16 or-
duction of 28% compared to diesel. Also, 27% reduction was produced thogonal array. The factors such as blends, EGR, load and speed con-
by butanol. Since the surface area of TiO2 is higher than combustion is sidered as the initial parameters to predict the variation of engine ef-
very appreciable. Although TiO2 had a capacity to reduce emission, due ficiency and emission [27]. Four different blends 0%, 10%, 20% and
to the waste cooking oil viscosity and oxygen concentration, the NOx 50% are dispersed with the 0%, 10%, 15% and 20% EGR. Further loads
can’t be reduced significantly like CO and HC. Similar to TiO2, and also varied from 40%, 60%, 80% and 100% at two different speeds
butanol also reduces the gaseous emission [17]. Hosseinzaden et al. 1600 rpm to 2400 rpm. From the above spelled out on the Taguchi
conducted a study on aqueous nanoparticles on engine performance method, it is obvious it is an effective tool proposed for optimizing the
and emission. The aqueous solution is mixed with blends at various experimental procedures. Employing Taguchi decreases the number of
fractions of 38 µm, 75 µm, and 150 µm. The inclusion of water resulted experimental runs and effort. Although the Taguchi is applied to IC
in an improvement in emission values. Further, due to the high latent engines to predict the combustion and emission, they barely studied
heat of vaporization, NOx generation is reduced appreciably. The with hydrogen as dual combustion with 5 factors and 4 levels. As a
parameters, such as BP and BTE also increased with reduced BSFC [18]. result, we aimed to study the effect of hydrogen fraction and MWCNTs
Among other blends, 38 µm reported reduced NOx, HC and CO. in four different ratios using the Taguchi approach. Further, the study

Table 7
ANOVA (analysis of variance) for the performance parameters.
Parameters Source DF Seq SS Adj MS F P S R

BTE Hydrogen (%) 3 13.0201 4.34 17.97 0.02 0.49 97.53%

Nanoparticles (ppm) 3 8.5424 2.847 11.79 0.036
Ignition pressure (bar) 3 4.68 1.56 6.46 0.08
Ignition timing (bTDC) 3 2.4 0.802 3.32 0.175
BSFC Hydrogen (%) 3 11,926 3975 0.04 0.987 313.5 8.30%
Nanoparticles (ppm) 3 6846 2282 0.02 0.994
Ignition pressure (bar) 3 3126 1042 0.01 0.998
Ignition timing (bTDC) 3 4726 1599 0.01 0.997

4
S. Manigandan, et al. Fuel 276 (2020) 118120

Table 8
ANOVA (analysis of variance) for the emission parameters.
Parameters Source DF Seq SS Adj MS F P S R

HC Hydrogen (%) 3 3.33 1.11 0.04 0.986 5.103 9.75%

Nanoparticles (ppm) 3 1.43 0.48 0.02 0.996
Ignition pressure (bar) 3 1.73 0.58 0.02 0.995
Ignition timing (bTDC) 3 1.95 0.65 0.02 0.994
NOx Hydrogen (%) 3 886 295 0.01 0.999 183.55 3.92%
Nanoparticles (ppm) 3 1976 659 0.02 0.996
Ignition pressure (bar) 3 639 213 0.01 0.999
Ignition timing (bTDC) 3 624 208 0.01 0.999
CO Hydrogen (%) 3 0.000949 0.000316 0.06 0.977 0.072 10.01%
Nanoparticles (ppm) 3 0.000587 0.000196 0.04 0.988
Ignition pressure (bar) 3 0.00013 0.000043 0.01 0.999
Ignition timing (bTDC) 3 0.00071 0.000024 0 0.999
CO2 Hydrogen (%) 3 3.159 1.053 0.11 0.95 3.13 16.98%
Nanoparticles (ppm) 3 1.125 0.375 0.04 0.988
Ignition pressure (bar) 3 0.274 0.091 0.01 0.999
Ignition timing (bTDC) 3 1.471 0.49 0.05 0.983

Fig. 2. S/N values of factor levels for BTE.

also conducted to identify the role of ignition timing and ignition The MWCNT is dispersed with the diesel by ultrasonication at 25 kHz
pressure using L16 orthogonal array. [29]. Further, the ignition pressure effect discussed at four intervals of
180 bar, 200 bar, 220 bar and 240 bar. In addition to above, the ig-
nition timing varied at 210bTDC, 230bTDC, 270bTDC and 310bTDC
2. Materials and methods
respectively. All proportions were examined at different engine loads of
25%, 50%, 75% and 100%. A high accuracy controller is connected in
2.1. Experimental setup
the hydrogen pipelines to measure the mass flow of hydrogen to the
prescribed value. Hydrogen passed through the flame arrestor for safety
The experiments were performed on a single cylinder, dual direct
and suppressed the fire hazards. Before being of each test, the engine
injection, water-cooled diesel engine whose technical feature is listed in
allowed running at ideal conditions up to 10 min to ensure the relia-
Table 1. Fig. 1 indicates an idea about the schematic layout diagram of
bility of obtained data. The test was repeated at least five cycles and
the test setup. A detailed explanation of the dual DI engine can be found
averaged for the reproducibility data. The documented signals are
in our previous work [6,19]. The test specimen is given in Table 2. A
converted to data by USB-6353-DAS (data acquisition system, National
conventional ignition system and air box was used to build up sa-
instrument) [30].
tisfactory air to attain smooth combustion [28]. The engine setup
The ratio of hydrogen and diesel is monitored to observe the effect
coupled with a dynamometer and DC generator to provide brake load.
of hydrogen on combustion and performance characteristics.
Air-Fuel (A/F) ratio is determined using a flow meter. The AVL DiGas
emission meter linked to the necessary sensors and instruments to Ehydrogen
measure the emission of CO, HC, NOx and CO2. A series of tests were Hydrogen blend energy ratio =
Ehydrogen − Ediesel (1)
conducted by varying the factors such as engine load, hydrogen ratio,
MWCNT concentration, ignition pressure and ignition timing. The hy-
drogen of fraction 10%, 20% and 30% is tested with MWCNT nano-
particles at different concentrations of 30 ppm, 50 ppm and 80 ppm.

5
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 3. Mean values of factor levels for BTE.

were determined to ensure the uncertainties in the documented data.

2.3. Design of experiments

The experimental design was performed using the Taguchi method.

Taguchi is the statistical method to enhance engine performance by
optimizing the working conditions using an orthogonal array. The S/N
ratio and the means of means measured for different experimental runs
based on the factors designed [31]. Table 4 shows the factors and stages
of each run. The Taguchi is performed based on the signal to noise ratio
feature. Taguchi works based on the two key processes parameters.
First, S/N ratio is applied to identify the control factor, which decreases
the variability. Secondly, the elimination of the control factor without
affecting the SN ratio. There are three types of S/N ratio, higher is
better (HSN), normal is better (NSN) and lower is better (LSN) [32]. The
S/N ratio formula was determined for the parameters such as BTE,
BSFC, CO, HC, NOx and CO2. Despite three S/N ratios, for higher ac-
Fig. 4. Interaction ANOVA plot for BTE. curacy of the data, HSN is applied. On the other hand, when the
magnitude and the difference are smaller LSN is used. For instance,
2.2. Uncertainty performance parameters are measured using HSN and the emissions are
using LSN. To avoid the perplexity, all data measured under “smaller
Uncertainties and errors usually take place during measurements the better” expression for the S/N ratio used to determine BTE, BSFC,
due to the calibration, manufacturing defects, environment conditions, NOx, HC, CO and CO2 using MINITAB 18 software [33]. The formula to
instrument selection and optimization. Among several errors observa- calculate the signal to noise ratio [47] is given below
tion error and improper selection of instruments is the primary. A series For S/N higher is better;
of literature found the relation to find the error through the variable R n
1 1
[28], where R is the resultant of the uncertainty of measurements. Signal − noiseratio = −10log ∑
n i=1
yi2 (4)
R = R(X1, X2⋯⋯Xn) (2)

The partial derivative of the sensitivity [30] Lower is better;

2 2 n
UR U 1
= ⎛ Xn∂R ⎞ Xn
⎜ ⎟
Signal − noiseratio = −10log ∑ yi2
R R∂X 2 n (5)
⎝ n ⎠ Xn (3) i=1

R- Given functions of the independent variables Where, n, yi and i shows the repetitions in the experiments. To be
∂R
- The measured result sensitivity for a single variable. concise, y is the response factor and n is the number of observations
∂x1
The obtained uncertainties calibrated from Table 3. By using the made during the trial.
relation, the percentage of the overall uncertainty found to be ± 2.95% In addition to Taguchi, ANOVA is performed from the total sum of
which was in the satisfactory limit. Before performing the test proce- squares, sum of squares, and sum of squares due to parameters [31].
dure, the initial values of the performance and emission equipment The essential formula listed below;

6
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 5. S/N values of factor levels for BSFC.

Fig. 6. Mean values of factor levels for BSFC.

N
S ⎤ T2 Sumofsquarewithin
⎡ MSw =
Totalsumofsquares = ⎢∑ i 2⎥ − DOF (10)
N N (6)
⎣ i=1 ⎦
F-Values are usually from 90% to 99% confidence intervals. During
N
⎡ A 2⎤ T2 the design, the model should have the confidence intervals between the
Sumofsquaresforfactorn = ⎢∑ i ⎥ −
n N (7) above-mentioned ranges [34].
⎣ i = 1 Ai ⎦
Vp
SN12 SN22 SN32 F − value =
Sumofsquaresduetoparameters = + + Ve (11)
N1 N2 N3 (8)
Vp − Variance due to the parameter P; Ve − Error variance
Degree of freedom (DOF),
d fw = k − 1 Sumofsquaresduetoparameters ⎞
(9) Percentcontribution(PC) = 100 ⎜⎛ ⎟

⎝ Totalsumofsquares ⎠ (12)
Mean of squares (MS),

7
S. Manigandan, et al. Fuel 276 (2020) 118120

3.1. Performance characteristics

The performance of the brake thermal efficiency (BTE) at the dif-

ferent load conditions, injection timing, and injection pressure are
shown for different blends. Figs. 2 and 3 show the mean of SN ratios
and the mean of means for the different stages 1 to 4 with the criteria of
LSN. From the results, it is evident that 1A-4B-3C-3D-1E is the best
combination for producing higher brake thermal efficiency. Further,
from the S/N ratios mean it is observed that, as the load increased, the
brake thermal efficiency is improved. Meanwhile, the addition of 30%
hydrogen to the fuel generates more power than other blends. In regard
to the ignition timing, 210bTDC recorded higher thrust during the dual
combustion than 270bTDC. Simultaneously, 80 ppm of MWCNTs im-
proves the combustion of the engine due to the surface area to volume
ratio [15,35]. Furthermore, the higher thermal conductivity [35] and
the ignition property of the nanoparticles are the other crucial reasons
for higher BTE upon the addition of MWCNTs. Although the addition of
Fig. 7. Interaction ANOVA plot for BSFC. nanoparticles enhances the combustion as the ratio is increased, the
stability of the combustion is decreased. Thus, increasing the MWCNTs
In this study, the factors such as load, hydrogen, MWCNT, ignition above 60 ppm is not recommended. Meanwhile, the evaporation rate
pressure and ignition timing are studied with 4 levels. According to the and reduced ignition timing help the model 4A-4B-3C-3D-1E to attain
factors and levels, the orthogonal array is expressed as L16 as shown in higher BTE. Further, increasing the ratio of hydrogen shortens the
Table 5. The ANOVA performed to find the best of five factors which combustion duration and increases the flame velocity [36]. From the
increases the performance with reduced emission. The data is analyzed above regard, it is appealing to state; the BTE can be improved by
based on the degree of freedom (DOF), sequential sum of squares (Seq adding hydrogen to the diesel engine by dual combustion with opti-
SS), adjusted mean squares (Adj SS), responses (S) and variation factor mized ignition timing and ignition pressure. The most compelling ar-
(R). gument is presented by Nag et al [37]. The second significant perfor-
mance is obtained from the 4A-3B-2C-2D-3E combination. Compared to
the first combination, the second combination is cheaper due to the
3. Results and discussion
usage of low concentration of MWCNTs. To give an illustration, the
difference between the top two combinations is 2.5% based on the
The performance and emission characteristics of the engine are
Taguchi optimization and SNR. To identify the interaction between the
optimized using the authors developed L16 orthogonal array with five
factors such as hydrogen, nanoparticles, ignition timing and ignition
factors such as engine load (A), hydrogen (B), MWCNTs (C), ignition
pressure, the interaction plot is compiled using ANOVA as shown in
pressure (D), ignition timing (E). The array is used to find the para-
Fig. 4. Although the BTE is depended on the five factors, the role of
meters like BTE, BSFC, HC, NOx, CO, and CO2. In addition to Taguchi,
hydrogen and the nanoparticles are significant in improving the BTE. In
general linear ANOVA was applied to predict the regression and para-
particular, 30% hydrogen produces more brake power than other
meter errors. Table 6 represents the outcome of each experimental run.
combinations.
Further, Table 7 and Table 8 show the ANOVA for the performance and
In general, the BSFC increases as the blends added to the neat diesel
emission characteristics.

Fig. 8. S/N values of factor levels for NOx.

8
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 9. Mean values of factor levels for NOx emission.

possibility of an increase in combustion efficiency is evident. In addi-

tion to the above, the molecular weight of the fuel is also responsible for
lower BSFC [38]. Upon the addition of hydrogen, the molecular weight
of the substance is reduced which results in the effective degradation
process. Nonetheless, the carbon content in the MWCNT additives in-
creases the carbon oxidation engine and consumes less fuel than con-
ventional fuel. The second best combination for reduced BSFC is 3A-3B-
2C-1D-2E, the difference between the approximately 10%. Although the
usage of the nanoparticles volume and the ratio of hydrogen is less, the
first combination is very effective based on the performance perspec-
tive. From the above regard, it is clear the BSFC depends on all five
factors. However, the role of hydrogen is significant than nanoparticles
[6]. Based on the effects of choose variables on the BSFC the sequential
order is hydrogen, nanoparticles, ignition timing and ignition pressure.
Fig. 7 represents the interaction plot of BSFC for different hydrogen
ratio. From the Fig. 7, it is clear that 30% of hydrogen is an effective
combination for lower BSFC.

3.2. Emission characteristics

Fig. 10. Interaction ANOVA plot for NOx emission.
The inventions to reduce the formation of NOx emission are in-
owing to the calorific value and increased mass flow rate. The BSFC was creased due to the stringent emission rules. NOx emission is higher in
recorded for different combinations of engine load, hydrogen, nano- the conventional diesel engine owing to higher in-cylinder temperature.
particles, ignition pressure and ignition timing. Figs. 5 and 6 shows the Although the diesel does not contain any nitrogen content, during the
SNR and MM for the brake specific fuel consumption by applying Ta- burning, it emits poisonous nitrogen of oxides due to the interaction of
guchi techniques. From the analysis of SNR and MM, the factors which air and fuel during combustion. In general, the emission of NOx in-
are affecting the BSFC were determined. The least fuel consumption is creases when the engine load is increased. A few major findings found
recorded for the combination of 4A-4B-3C-2D-1E. From the Taguchi the addition of biodiesel blends increases the emission of NOx due to
results, as the load is increased the BSFC is dropped significantly. the oxygen content [37]. During the combustion of biodiesel the cy-
Meanwhile, hydrogen and MWCNTs also other key reasons for the de- linder temperature increases due to the presence of oxygen content,
creased BSFC irrespective of engine loads, ignition pressure and timing. which results in the production of higher NOx. However, it can be
The key reason for the above phenomenon is 30% of hydrogen and avoided by bounding to the rich combustion. Based on the designed
50 ppm of nanoparticles. The addition of hydrogen to the diesel engine experimental runs, S/N ratio and MM is represented in Fig. 8 and Fig. 9.
enhances the total calorific value thus BSFC is reduced [24]. Further, From the Taguchi design, the optimized method for less emission of
dispersing nanoparticles increases the oxygen content in the fuel which NOx is 1A-4B-4C-1D-1E. The stated design combination produces less
facilitates the combustion at higher levels. The combination of both amount of NOx due to the role of nanoparticles and the hydrogen ratio
hydrogen and MWCNTs increases the fuel atomization which led to a in the diesel fuel. By intruding the hydrogen with the diesel fuel, the
reduced amount of fuel burnt in the diffusion mode. Thereby there is a cylinder temperature is reduced during the combustion thus leads to

9
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 11. S/N values of factor levels for HC emission.

Fig. 12. Mean values of factor levels for HC emission.

less production of NOx [39]. Furthermore, the addition of nanoparticles reduces the NOx formation even at higher loads. Although all five
enhances the combustion thereby less production of unburnt gases is- factors reduces the NOx but the role of hydrogen is very significant.
sues from the engine during combustion. From Fig. 10, it is evident as Fig. 13 represents the interaction plot of hydrogen fraction on the en-
the load increase the NOx emission increased due to the increase in gine load. The combination with the 30% hydrogen produces the less
temperatures. The lowest amount of NOx obtained at lower cylinder amount of NOx up to 80% of engine load. As the engine load increased,
pressure and lowest engine load. In light of the evidence from Benajes the 20% hydrogen showed an effective decrease in the emission of NOx.
et al [40] it is general to witness, as the engine load is increased NOx From the S/N ratio, the next efficient combination to produce the less
emission increases irrespective of hydrogen ratio, nanoparticles, igni- NOx is 2A-3B-2C-2D-3E. This is a cost-effective due to the utilization of
tion timing and ignition pressure owing to the reaction of oxygen and lower concentration of both hydrogen and MWCNT.
nitrogen. By increasing the concentration of hydrogen and MWCNTs, Hydrocarbons emitted from the diesel engine due to the incomplete
the presence of lower active radicals is decreased [35]. This behaviour combustion process in the combustion chamber. In general higher

10
S. Manigandan, et al. Fuel 276 (2020) 118120

content in diesel fuel. By lowering the ignition pressure and ignition

timing, the combustion rates can be enhanced with reduced quenching
area inside the cylinder. The next best combination from the analysis is
2A-2B-2C-4D-3E. This is the second efficient combination of reduced
HC emission. On the other hand, it is cost-effective due to the reduced
usage of hydrogen and nanoparticles. Besides the low concentration of
nanoparticles, this process is very efficient due to sensitive injection
timing of 270bTDC and higher ignition pressure of 220 bar. Fig. 12
shows the interaction plot for the crucial parameters such as hydrogen
to engine load. This clearly emphasise, the addition of hydrogen re-
duces the formation of hydrocarbons. Further, up to 75% of engine
load, the 30% hydrogen is very effective. As the load increased above
75%, the emission of hydrocarbons is higher due to the cylinder tem-
perature and shorter burning duration [45]. The 20% hydrogen ob-
served to be the most persuasive than other ratios of hydrogen in the
diesel engine. This trend is consistent with our previous work results.
CO emission depends on the carbon content in the fuel. Since diesel
fuel contains high carbon content emission of CO is always a major
worry for the engines running in conventional fuels. When the fuel is
Fig. 13. Interaction ANOVA plot for HC emission. oxidized with the oxygen, the CO and CO2 are formed based on the
operating condition [43]. Figs. 14 and 15, shows the S/N ratio and
compression ratio reduces the HC emission significantly at low engine means of means for the HC for various factors. The best combination
speed. But due to the power demand, all engines are operated at higher design for generating the least CO emission is 1A-4B-3C-3D-3E. From
speeds. Thus, the emission of HC increases as the engine load increase. the optimal design, it is evident that the contribution of hydrogen to
Although the lower compression ratio reduces the HC but it is not en- reduce the CO emission is huge. Higher hydrogen fraction to the diesel
ough to be safe to the ecosystem [12]. Mostly the reason for the in- engine reduces the total carbon content in the fuel which obliviously
complete combustion is wall wetting and flame quenching [41]. Thus reduces the CO formations. In the other hand, the role of nanoparticles
addition of nanoparticles and the hydrogen to the diesel engine can up to 60 ppm is very impressive. Once the ratio of nanoparticles in-
reduce the flame quenching leads to reduced HC emission. Figs. 11-12 creases the oxygen content in the fuel is increased which leads to lean
states the S/N ratio and MM based on the variance factors. From the combustion [18]. Further, factors such as injection timing and pressure
Taguchi design, the best combination for the lowest HC emission is 1A- also contribute to reducing CO emission. From the several experimental
3B-2C-1D-1E. Among five factors, the role of hydrogen on reduction of runs, the optimized condition to run the engine are 200 bar and
hydrocarbons and nanoparticles is profound compared to timing and 270bTDC with 30% of hydrogen fraction ratio. Consequently, the
pressure. By adding MWCNT to the diesel increases the calorific value second-best operation condition from the Taguchi is 2A-3B-2C-4D-4E.
of the fuel and also improves the combustion by acting as the catalytic This is a very promising design due to the utilization of the least con-
reactor [42]. Meanwhile addition of hydrogen reduces the latent heat of centration of MWCNT. The main reason for the least emission of CO is
vaporization which resulted in rich fuel and air mixing. Besides in- high injection pressure of 220 bar and advanced ignition timing of
creasing the ratio of hydrogen reduces the paraffin and cycloparaffin 310bTDC owing to the least dwell time for oxidation at lower engine

Fig. 14. S/N values of factor levels for CO emission.

11
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 15. Mean values of factor levels for CO emission.

in the hydrogen and MWCNT is another reason for the reduced emission
of CO2. More to the point, the stoichiometric air to fuel ratio for the
hydrogen and MWCNT blends is lower than neat diesel owing to their
higher oxygen concentration. Figs. 18 and 19 represent the S/N ratio
and MM. From the Taguchi design, the effective combination for the
least emission of CO2 is 1A-3B-2C-3D-1E. From the analysis, it is proved
the addition of hydrogen and the nanoparticles to diesel reduces the
emission of CO2. However, increasing the concentration of hydrogen
and MWCNTs increases the emission of CO2 at higher engine load due
to the inherent property of fuel blends to contain oxygen [44]. To get
the clear representation the Fig. 19 drawn the interaction plot to ob-
serve the effect of CO2 on engine load and hydrogen using ANOVA. The
results depicted that the addition of 30% H2 to the engine reduces the
CO2 up to the 75% engine load. As the load elevated to higher levels,
the increase in CO2 is reported. On the other end, the positive effect is
witnessed by 20% H2. The next best combination is 2A-4B-4C-4D-4E.
Increasing the ignition timing and pressure reduces the emission of CO2
by obtaining a high combustion rate. As the load increased the emission
Fig. 16. Interaction ANOVA plot for CO emission. of CO2 is increased owing to late burning caused by delay ignition [24].
Considering above all, the addition of hydrogen reduces the CO2 by
loads. From the ANOVA it is found, the role of hydrogen is profound. improving combustion and reducing the carbon content in the fuel.
However, the addition of MWCNT in the diesel converts the CO to CO2
by its convincing redox-active property [46]. Moreover, addition cat- 4. Conclusion
alytic MWCNT increases the oxygen present in the diesel fuel which
enhances the combustion and atomization. Fig. 16 shows the interac- Being sustainable renewable energy, hydrogen plays a prevailing
tion plot for hydrogen and engine load. Although the addition of role in the future of the transportation sector. This investigation was set
MWCNT and hydrogen enhances the combustion, however, the emis- to observe the effects of hydrogen and nanoparticles on the combustion
sion of CO is uncontrollable at high engine loads which are due to the and emission of the diesel engine. The experiments are conducted by
high in-cylinder temperature. Compared to other hydrogen fraction, varying the engine load, hydrogen fraction, nanoparticles ratio, ignition
30% H2 showed a profound decrease in CO. On the contrary, at higher timing, and ignition pressure using full factorial design with 16 runs.
loads, the 20% H2 showed a notable contribution by generating less CO The obtained results are analyzed using the Taguchi method. By ex-
than 30% H2 (Fig. 17). amining the results, it is found that the addition of hydrogen and
The reduction of CO2 emission is internationally most interesting MWCNT on diesel engine successfully enhances the engine performance
since the huge quality of gases emitted at high concentration to the and emission characteristics. The Taguchi method involved to predict
atmosphere. By achieving the higher combustion the emission of CO2 the optimum factors for the maximization of BTE, reduction of BSFC
can be degraded [39]. Thus increasing the combustion by the addition and exhaust. The optimum condition for generating higher BTE ob-
of MWCNT is possible by oxidation of carbon into CO2. An almost si- tained at 30% H2 and 50 ppm of MWCNT at one-quarter of the engine
milar argument was put forward by Benajeset al. Further, lower carbon load. However, at higher loads, the BTE is very profound due to the

12
S. Manigandan, et al. Fuel 276 (2020) 118120

Fig. 17. S/N values of factor levels for CO2 emission.

Fig. 18. Mean values of factor levels for CO2 emission.

increase the ignition pressure. In regard to BSFC, the optimum condi- CRediT authorship contribution statement
tion is observed at higher engine load and hydrogen fraction. Further,
the addition of 50 ppm of MWCNT reduces the BSFC extremely irre- S. Manigandan: Conceptualization, Methodology, Writing - ori-
spective of engine load. The preeminent result for NOx obtained at the ginal draft preparation. A. E. Atabani: Writing - review & editing.
30% H2 and higher concentration of MWCNT without advancing the Vinoth Kumar Ponnusamy: Supervision, Project administration.
ignition timing. Meanwhile, the best value for HC, CO, and CO2 ob- Arivalagan Pugazhendhi: Writing - original draft preparation. P.
tained at 20% H2 and 50 ppm of MWCNT. From the ANOVA analysis, it Gunasekar: Investigation, Writing - original draft preparation. S.
is typical that the impact of hydrogen on multiple characteristics in the Prakash: Writing - review & editing.
engine output is higher than nanoparticles, ignition pressure, and
timing. All things considered, the application of Taguchi to identify the
Declaration of Competing Interest
optimal operating condition is viable by saving ample of time and cost.
Further, the role of hydrogen and MWCN are significant.
The authors declare that they have no known competing financial

13
S. Manigandan, et al. Fuel 276 (2020) 118120

[14] EL-Seesy AI, Hassan H, Ookawara SJ. Performance, combustion, and emission
characteristics of a diesel engine fueled with Jatropha methyl ester and graphene
oxide additives. Energy Conversion Manage 2018;15(166):674–86.
[15] El-Seesy AI, Abdel-Rahman AK, Bady M, Ookawara SJ. Performance, combustion,
and emission characteristics of a diesel engine fueled by biodiesel-diesel mixtures
with multi-walled carbon nanotubes additives. Energy Convers Manage
2017;1(135):373–93.
[16] Najafi G. Diesel engine combustion characteristics using nano-particles in biodiesel-
diesel blends. Fuel 2018;15(212):668–78.
[17] Örs I, Sarıkoç S, Atabani AE, Ünalan S, Akansu SO. The effects on performance,
combustion and emission characteristics of DICI engine fuelled with TiO2 nano-
particles addition in diesel/biodiesel/n-butanol blends. Fuel 2018;15(234):177–88.
[18] Hosseinzadeh-Bandbafha H, Khalife E, Tabatabaei M, Aghbashlo M, Khanali M,
Mohammadi P, et al. Effects of aqueous carbon nanoparticles as a novel na-
noadditive in water-emulsified diesel/biodiesel blends on performance and emis-
sions parameters of a diesel engine. Energy Convers Manage
2019;15(196):1153–66.
[19] Manigandan S, Ponnusamy VK, Devi PB, Oke SA, Sohret Y, Venkatesh S, et al. Effect
of nanoparticles and hydrogen on combustion performance and exhaust emission of
corn blended biodiesel in compression ignition engine with advanced timing. Int J
Hydrogen Energy 2019.
[20] Gunasekar P, Manigandan S, Venkatesh S, Gokulnath R, Vimal R, Boomadevi P.
Fig. 19. Interaction ANOVA plot for CO2 Emission.
Effect of hydrogen addition on exergetic performance of gas turbine engine. Aircraft
Eng Aerospace Technol 2019.
[21] Verma S, Suman A, Das LM, Kaushik SC, Tyagi SK. A renewable pathway towards
interests or personal relationships that could have appeared to influ- increased utilization of hydrogen in diesel engines. Int J Hydrogen Energy 2019.
ence the work reported in this paper. [22] Dimitriou P, Tsujimura T. A review of hydrogen as a compression ignition engine
fuel. Int J Hydrogen Energy 2017;42(38):24470–86.
[23] Tsujimura T, Suzuki Y. The utilization of hydrogen in hydrogen/diesel dual fuel
Acknowledgment engine. Int J Hydrogen Energy 2017;42(19):14019–29.
[24] Manigandan S, Gunasekar P, Poorchilamban S, Nithya S, Devipriya J,
The authors are also thankful to Sathyabama Institute of Science Vasanthkumar G. Effect of addition of hydrogen and TiO2 in gasoline engine in
various exhaust gas recirculation ratio. Int J Hydrogen Energy
and Technology, Chennai City, Tamil Nadu, India, for supporting the
2019;44(21):11205–18.
work. The authors are also grateful to the Ministry of Science and [25] Wu HW, Wu ZY. Using Taguchi method on combustion performance of a diesel
Technology-Taiwan Research Grant (107-2113-M-037-007-MY2), and engine with diesel/biodiesel blend and port-inducting H2. Appl Energy
also supported by the Research Center for Environmental Medicine, 2013;1(104):362–70.
[26] Kumar RS, Sureshkumar K, Velraj R. Optimization of biodiesel production from
Kaohsiung Medical University, Kaohsiung City, Taiwan from “The Manilkarazapota (L.) seed oil using Taguchi method. Fuel 2015;15(140):90–6.
Featured Areas Research Center Program within the framework of the [27] Ayhan V, Çangal Ç, Cesur İ, Çoban A, Ergen G, Çay Y, et al. Optimization of the
Higher Education Sprout Project” by the Ministry of Education (MOE) factors affecting performance and emissions in a diesel engine using biodiesel and
EGR with Taguchi method. Fuel 2020;1(261):116371.
in Taiwan, and NSYSU-KMU collaboration research project (NSYSU- [28] Manigandan S, Gunasekar P, Devipriya J, Nithya S. Emission and injection char-
KMU 107-I004)-Taiwan for research grant support. acteristics of corn biodiesel blends in diesel engine. Fuel 2019;1(235):723–35.
[29] Naddaf A, Heris SZ, Pouladi B. An experimental study on heat transfer performance
and pressure drop of nanofluids using graphene and multi-walled carbon nanotubes
References based on diesel oil. Powder Technol 2019;15(352):369–80.
[30] Dalla Nora M, Lanzanova TD, Zhao H. Investigation of performance and combustion
[1] BP statistical review of world energy 2019, London. characteristics of a four-valve supercharged two-stroke DI engine fuelled with ga-
[2] Elwardany AE, Marei MN, Eldrainy Y, Ali RM, Ismail M, El-kassaby MM. Improving soline and ethanol. Fuel 2018;1(227):401–11.
performance and emissions characteristics of compression ignition engine: Effect of [31] Ansari NA, Sharma A, Singh Y. Performance and emission analysis of a diesel engine
ferrocene nanoparticles to diesel-biodiesel blend. Fuel 2020;15(270):117574. implementing polanga biodiesel and optimization using Taguchi method. Process
[3] Tyagi H, Phelan PE, Prasher R, Peck R, Lee T, Pacheco JR, et al. Increased hot-plate Saf Environ Prot 2018;1(120):146–54.
ignition probability for nanoparticle-laden diesel fuel. Nano Lett 2008;8(5):1410–6. [32] Sharma A, Singh Y, Singh NK, Singla A. Sustainability of jojoba biodiesel/diesel
[4] Wang X, Zhang J, Yin MA, Wang G, JinPeng HA, Dai M, et al. A comprehensive blends for DI diesel engine applications-taguchi and response surface methodology
review on the properties of nanofluid fuel and its additive effects to compression concept. Ind Crops Prod 2019;1(139):111587.
ignition engines. Appl Surf Sci 2019;5:144581. [33] Manigandan S, Gunasekar P, Nithya S, Devipriya J, Raja S, Venkatesh S. Influence
[5] Heidari-Maleni A, Gundoshmian TM, Jahanbakhshi A, Ghobadian B. Performance of injection parameters on NOx emission from biodiesel powered diesel engine by
improvement and exhaust emissions reduction in diesel engine through the use of Taguchi technique. Int J Ambient Energy 2018;18:1–5.
graphene quantum dot (GQD) nanoparticles and ethanol-biodiesel blends. Fuel [34] Wu ZY, Wu HW, Hung CH. Applying Taguchi method to combustion characteristics
2020;1(267):117116. and optimal factors determination in diesel/biodiesel engines with port-injecting
[6] Manigandan S, Sarweswaran R, Devi PB, Sohret Y, Kondratiev A, Venkatesh S, et al. LPG. Fuel 2014;30(117):8–14.
Comparative study of nanoadditives TiO2, CNT, Al2O3, CuO and CeO2 on reduction [35] Zhang Z, Lu Y, Wang Y, Yu X, Smallbone A, Dong C, et al. Comparative study of
of diesel engine emission operating on hydrogen fuel blends. Fuel using multi-wall carbon nanotube and two different sizes of cerium oxide nano-
2020;15(262):116336. powders as fuel additives under various diesel engine conditions. Fuel
[7] Saxena V, Kumar N, Saxena VK. A comprehensive review on combustion and sta- 2019;15(256):115904.
bility aspects of metal nanoparticles and its additive effect on diesel and biodiesel [36] Gong C, Li Z, Chen Y, Liu J, Liu F, Han Y. Influence of ignition timing on combustion
fuelled CI engine. Renew Sustain Energy Rev 2017;1(70):563–88. and emissions of a spark-ignition methanol engine with added hydrogen under lean-
[8] Nithya S, Manigandan S, Gunasekar P, Devipriya J, Saravanan WS. The effect of burn conditions. Fuel 2019;1(235):227–38.
engine emission on canola biodiesel blends with TiO2. Int J Ambient Energy [37] Nag S, Sharma P, Gupta A, Dhar A. Combustion, vibration and noise analysis of
2019;40(8):838–41. hydrogen-diesel dual fuelled engine. Fuel 2019;1(241):488–94.
[9] Ghamari M, Ratner A. Combustion characteristics of colloidal droplets of jet fuel [38] Tayari S, Abedi R. Effect of Chlorella vulgaris methyl ester enriched with hydrogen
and carbon based nanoparticles. Fuel 2017;15(188):182–9. on performance and emission characteristics of CI engine. Fuel
[10] Riani P, Garbarino G, Cavattoni T, Canepa F, Busca G. Unsupported cobalt nano- 2019;15(256):115906.
particles as catalysts: Effect of preparation method on catalytic activity in CO2 [39] Jhang SR, Chen KS, Lin SL, Lin YC, Cheng WL. Reducing pollutant emissions from a
methanation and ethanol steam reforming. Int J Hydrogen Energy heavy-duty diesel engine by using hydrogen additions. Fuel 2016;15(172):89–95.
2019;44(50):27319–28. [40] Benajes J, Pastor JV, García A, Monsalve-Serrano J. The potential of RCCI concept
[11] Rahman MM, Stevanovic S, Islam MA, Heimann K, Nabi MN, Thomas G, et al. to meet EURO VI NOx limitation and ultra-low soot emissions in a heavy-duty
Particle emissions from microalgae biodiesel combustion and their relative oxida- engine over the whole engine map. Fuel 2015;1(159):952–61.
tive potential. Environ Sci Processes Impacts 2015;17(9):1601–10. [41] Lee Z, Park S. Particulate and gaseous emissions from a direct-injection spark ig-
[12] Khond VW, Kriplani VM. Effect of nanofluid additives on performances and emis- nition engine fueled with bioethanol and gasoline blends at ultra-high injection
sions of emulsified diesel and biodiesel fueled stationary CI engine: A comprehen- pressure. Renewable Energy 2019.
sive review. Renew Sustain Energy Rev 2016;1(59):1338–48. [42] Gad MS, Yehia KA, Abdelhakeem AA. Effect of multi carbon nanosheet on diesel
[13] Ooi JB, Ismail HM, Swamy V, Wang X, Swain AK, Rajanren JR. Graphite oxide engine performance and emissions. Full Nanotubes Carbon Nanostruct
nanoparticle as a diesel fuel additive for cleaner emissions and lower fuel con- 2018;26(11):722–8.
sumption. Energy Fuels 2016;30(2):1341–53. [43] Jamrozik A, Tutak W, Pyrc M, Gruca M, Kočiško M. Study on co-combustion of

14
S. Manigandan, et al. Fuel 276 (2020) 118120

diesel fuel with oxygenated alcohols in a compression ignition dual-fuel engine. [46] Karavalakis G, Gysel N, Schmitz DA, Cho AK, Sioutas C, Schauer JJ, et al. Impact of
Fuel 2018;1(221):329–45. biodiesel on regulated and unregulated emissions, and redox and proinflammatory
[44] Parthasarathy M, Lalvani JI, Dhinesh B, Annamalai K. Effect of hydrogen on etha- properties of PM emitted from heavy-duty vehicles. Sci Total Environ
nol–biodiesel blend on performance and emission characteristics of a direct injec- 2017;15(584):1230–8.
tion diesel engine. Ecotoxicol Environ Saf 2016;1(134):433–9. [47] Teo SH, Islam A, Chan ES, Choong ST, Alharthi NH, Taufiq-Yap YH, et al. Efficient
[45] Zhang B, Ji C, Wang S. Combustion analysis and emissions characteristics of a biodiesel production from Jatropha curcus using CaSO4/Fe2O3-SiO2 core-shell
hydrogen-blended methanol engine at various spark timings. Int J Hydrogen Energy magnetic nano particles. J Cleaner Prod 2019;20(208):816–26.
2015;40(13):4707–16.

15