i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Effect of hydrogen enrichment on combustion

characteristics, emissions and performance of
a diesel engine

Yasin Karago _
€z a,**, Ilker Güler b, Tarkan Sandalcı a, Levent Yüksek a,
c,*
Ahmet Selim Dalkılıç
a
Automotive Division, Department of Mechanical Engineering, Mechanical Engineering Faculty, Yıldız Technical
University, Yildiz, Besiktas, Istanbul, 34349, Turkey
b
Türk Trakt€
or ve Ziraat Makineleri A.S‚., Sakarya, Turkey
c
Heat and Thermodynamics Division, Department of Mechanical Engineering, Mechanical Engineering Faculty,
Yildiz Technical University, Yildiz, Besiktas, Istanbul, 34349, Turkey

article info abstract

Article history: In this study, hydrogen fuel was injected into intake manifold using an LPG-CNG injector
Received 5 July 2015 that is controlled by a self-developed ECU, whereas diesel fuel was directly injected into
Received in revised form cylinder using diesel injector. Different hydrogen energy fractions are used in a diesel-
18 September 2015 fueled CI engine at 1100 rpm constant engine speed and full load. The effect of 0% (pure
Accepted 21 September 2015 diesel), 22%, and 53% hydrogen addition of total fuel energy (hydrogen þ diesel fuel) on CO,
Available online 14 November 2015 THC, smoke, and NOx emissions, engine performance (BSFC and brake thermal efficiency),
and combustion characteristics (in-cylinder pressure, heat release rate etc.) were experi-
Keywords: mentally investigated. According to obtained results, a great improvement was provided
Hydrogen with increasing percentage of hydrogen on CO (67.3% and 69.2%, for 22% and 53% hydrogen
Diesel engine enrichment, respectively) and smoke emissions (43.6% and 58.6%, for 22% and 53%
NOx hydrogen enrichment, respectively). Even though a slight raise was observed on THC
THC emissions, it is below emission regulations and can be ignored. On the other hand,
CO although a slight increase (almost the same value) was observed with 22% hydrogen
Smoke addition, a dramatic increase could not be prevented with 53% hydrogen addition in NOx
emissions compared with pure diesel fuel (0% hydrogen). Also, peak-in-cylinder pressure
values increased by 7.81% and 36.2% with 22% and 53% hydrogen addition, respectively, in
comparison to pure diesel fuel. Furthermore, a 25.77% increase in peak heat release rate
was obtained with 22% hydrogen addition and a great increase of 110.94% was acquired
with 53% hydrogen enrichment.
Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.

* Corresponding author. Tel.: þ90 212 383 2821; fax: þ90 212 383 2765.
** Corresponding author. Tel.: þ90 212 383 2901; fax: þ90 212 383 2765.
€ z), [email protected] (A.S. Dalkılıç).
E-mail addresses: [email protected] (Y. Karago
http://dx.doi.org/10.1016/j.ijhydene.2015.09.064
0360-3199/Copyright © 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5 657

an increase in thermal efficiency between 28% and 43%, and a

Introduction 20% decrease in NOx emissions were acquired.
Using hydrogen as additional fuel in compression ignition
The fast depletion of fuels and increase in prices of conven- engines is possible. According to this notion, hydrogen is used
tional fossil fuels in recent years force researchers to find a as additional fuel, and ignition can be provided by pilot diesel
new eco-friendly, renewable, and alternative fuel [1,2]. On the injection. In this dual fuel mode, three basic methods are
other hand, it is mandatory to find a new, clean, and alter- applied in the literature: timing hydrogen manifold injection,
native fuel to reduce negative effects on environment and on continuous hydrogen injection, and direct hydrogen injection
human beings, which often are generated by combustion of [7]. Continuous hydrogen injection causes some combustion
fossil fuels [3e5]. Most of the energy demand in transportation problems such as backfiring, and engine operating range is
is still provided by the petroleum based fuels. Even though the limited. Also, both continuous hydrogen injection and timing
measures have followed the Kyoto Protocol, the increase in hydrogen manifold injection methods lead to low volumetric
CO2 emissions could not be prevented, and a 27% increase in efficiency and poor power output since hydrogen displaces
total CO2 emissions and 37% increase in transport related to intake air [7]. By using direct hydrogen injection, these
CO2 emissions between 1990 and 2004 were obtained [6]. Also, handicaps can be overcome. However, direct hydrogen injec-
stringent emission regulations in regulated emissions (NOx, tion method requires a specially designed high-pressure
CO, THC, PM) force researchers to investigate alternative fuels injector that resists high in-cylinder temperature, pressure,
[7e10]. For this reason, some researchers [1,11e13] have been and modification on cylinder head. Consequently, timing
studying with alternative fuels such as ethanol, LPG, CNG, hydrogen manifold injection is the most appropriate method
LNG, producer gas, biogas, hydrogen, etc. Especially, hydrogen for hydrogen þ diesel dual fuel [8,16,31].
is an outstanding alternative fuel among these fuels, since it Hydrogen has unique combustion properties. For instance,
emits neither greenhouse gases (GHG) nor toxic pollutants the flammability range of hydrogen is wider than other fuels
[14,15]. Furthermore, hydrogen is a long-period renewable fuel and the flame speed of hydrogen is higher than other fuels
[7,8,16]. Hydrogen can be acquired by several methods like [18,32,33]. On the other hand, the quenching distance of
water electrolysis, biomass, chemical wastes, etc. [17,18]. hydrogen is shorter than other fuels. Thus, using hydrogen in
Hydrogen can be an alternative for electric energy because it diesel fuels as additional fuel improves CO, CO2, HC, and PM
can be assumed as an energy carrier [1]. emissions since it does not contain any carbon atoms [22].
Using hydrogen in spark ignition engines is more conve- Bose and Maji [1] sent 0.15 kg/h hydrogen into intake
nient than in compression ignition engines since it improves manifold on a single cylinder, and water compression ignition
lean operation region, combustion stability, brake thermal engine and partial load conditions were tested. Also, 10% and
efficiency, carbon monoxide, and unburned hydrocarbons 20% EGR were applied to prevent an increase in soot emis-
[19e22]. Although some researchers have assessed hydrogen sions. According to their results, CO, CO2, THC, and soot
in homogeneous charge compression ignition (HCCI) engines emissions were reduced with hydrogen addition, and NOx
[10,23,24], it is difficult to commercialize them due to their emissions were dropped by means of EGR. Liew et al. [22]
high level of cyclic variations, narrow engine operation investigated the effect of hydrogen on the combustion char-
ranges, and difficulties in ignition period. acteristics of a four-stroke, six-cylinder diesel engine experi-
Hydrocarbons, carbon monoxide, nitrogen oxides, and mentally. According to their results, the combustion duration
particulate matter are the basic regulated emissions that are was shortened and the peak rate of heat release value
emitted from petroleum-based fuels [8]. IC engines will be increased. Bari and Esmaeil [3] tested the effect of
used in transport in the near future [25]. Diesel engines are hydrogen þ oxygen gas blend as supplementary fuel that is
widely used in the transport sector, and the percentage of produced by an alkaline electrolyzer in a four-cylinder, direct-
diesel engines have been increasing. Diesel engines have injection diesel engine at different engine loads. They found
some advantages such as high thermal efficiency, low CO2, that HC, CO, and CO2 emissions were improved, but NOx
and THC emissions compared with spark ignition engines. emissions increased with hydrogen þ oxygen gas mixture
However, diesel engines produce high levels of NOx and addition. Ko € se and Ciniviz [7] studied on a four-cylinder,
smoke emissions [4,26e28]. Hydrogen could not be directly water-cooled, direct-injection diesel engine to investigate ef-
used as diesel fuel due to its high self-ignition temperature of fect of different percentage of (2.5%, 5%, and 7.5% on volume
576  C [1]. Therefore, an ignition source is required for using basis) hydrogen on the engine performance. They obtained
hydrogen in CI engines [7]. To use hydrogen directly as diesel that the increase in NOx emissions could not be prevented and
fuel, a glow plug or heating intake air is required. The high that HC, CO, and O2 emissions reduced with hydrogen
self-ignition temperature and high ignition delay are the most enrichment. Saravanan et al. [8] studied a single-cylinder,
important barriers beyond using hydrogen directly as diesel water-cooled, direct-injection compression-ignition engine,
fuel [29]. and 10 lpm constant hydrogen flow rate was supplied. In
Ikegami et al. [30] directly used hydrogen in a CI engine as addition, diesel fuel was injected into cylinder at 23 before
diesel fuel thanks to a glow plug. Antunes et al. [29] directly top dead center. Three hydrogen-injection duration (30 CA,
used hydrogen in a single-cylinder, naturally aspirated, direct- 60 CA and 90 CA) and five hydrogen-injection advance
injection compression-ignition engine instead of diesel fuel (5 BTDC to 15 ATDC) were experimentally studied. 5 ATDC
heating intake air to 80  C and a self-developed high-pressure hydrogen injection advance and 90 CA hydrogen injection
gas injector was used to inject hydrogen into cylinder. Ac- duration show the best emission and engine performance.
cording to their obtained results, a 14% power increase, Miyamoto et al. [26] worked on hydrogen enrichment on a
658 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5

single-cylinder, common-rail system in CI engine that has EGR dyno are presented in Table 1. A TLF 23 NewFlow thermal
system. They controlled engine noise and NOx emission with mass flow meter and a VZ 0.2 AL-S Sika miniature turbine-
two different strategies. These strategies are low-temperature type positive-displacement flow meter were used to mea-
combustion in which diesel fuel injection advance was sure hydrogen flow rate and diesel flow rate, respectively. An
delayed until 2 after top dead center and EGR method. inclined tube manometer was used to measure pressure dif-
Stringent emission regulations and fast depletion of fossil ference in air where air is introduced into an air tank and
fuels force researchers to study on a new, clean, and renew- passes into an orifice. Air flow rate was calculated with ob-
able alternative fuel to use in ICEs. Also, using after-treatment tained pressure difference in orifice. Engine coolant temper-
systems on road vehicles becomes obligatory to meet strin- ature was measured with K-type thermocouples. Schemetic
gent emission regulations. However, the cost of after- view of the experimental setup is illustrated in Fig. 1. The
treatment systems is close to engine cost because of high diesel fuel consumption, hydrogen fuel consumption, the
expense of catalyst materials. For this reason, hydrogen, load-cell, AVL Dicom 4000 exhaust gas analyzer, and AVL 415S
which has unique combustion properties, is an attractive exhaust analyzer were connected via a USB-type data acqui-
alternative fuel. It is an eco-friendly fuel, since neither carbon- sition card (NI 6215) and it was connected to a personal
based pollutants nor GHG emittes are formed with combus- computer. Using Labwiev software, a program interface was
tion of hydrogen. In a previous study, Sandalcı et al. [34] developed to collect data.
investigated the effect of 0%, 16%, 36%, and 46% hydrogen
enrichment (energy basis) on performance and emission H2 fuel line
characteristics of a compression ignition engine at 100% en-
gine load and 1300 rpm constant engine speed, which is equal A schematic diagram of hydrogen fuel line was given in Fig. 2.
to C engine speed of European Stationary Cycle (ESC). In this Hydrogen fuel was stored in a high-pressure tank, and the
study, the effect of 0%, 22%, and 53% hydrogen addition on pressure of hydrogen was dropped using a double-stage
energy basis is investigated at 1100 rpm constant engine pressure regulator (SS 316). A shut-off valve was installed
speed, which is equal to A engine speed of ESC. Our aim is to online to prevent a possible backfiring. A flame arrester and a
investigate the effect of a different level of hydrogen addition check valve were installed before the hydrogen was sent into
at low engine speed since compression ignition engines usu- the intake manifold to prevent an explosion. A relief valve was
ally have emissions (especially smoke emission). However, set into the fuel line to avoid overpressure, and a discharge
increasing amounts of hydrogen level can overcome this line was set, which purged overpressure outside the labora-
problem; the formation of oxides of nitrogen can be affected tory. A line regulator was used to fix line pressure to 4 bar. A
negatively. For this reason, investigation of effect of different rotameter and a thermal mass flowmeter were used, which
levels of hydrogen addition (0%, 22% and 53%) on performance were calibrated to hydrogen. A Keihin LPG-CNG gas injector
and emissions (especially on smoke and NOx emissions) at full was used to inject hydrogen into an intake manifold. A self-
load and 1100 rpm engine speed has been performed in the developed electronic control unit was used to control gas
current study. Because of the combustion problems after 53% injector. The material of hydrogen line is stainless steel, and
hydrogen enrichment (especially backfire and pre-ignition), the line was durable up to 250 bar.
this level could not be exceeded. Furthermore, the effect of
hydrogen addition on combustion characteristics is investi- Tail-pipe emission measurement
gated elaborately.
CO, THC, and NOx emissions were measured by AVL Dicom
4000 exhaust gas analyzer, and the smoke emission was
Experimental setup measured by AVL 415S smoke analyzer. However, AVL Dicom
4000 measures CO and THC emissions as %vol.; NOx emissions
An experimental study was performed on the emissions, as ppm, respectively, and AVL 415S measures as FSN or mg/
performance, and combustion characteristics of a hydrogen- m3 all were converted into g/kWh according to VDMA Exhaust
enriched diesel engine at 1100 rpm constant engine speed. Emission Legislation to obtain more general results.
All of the tests were done at the IC Engines Laboratory at Yildiz
Technical University. The test cell and the compression igni-
tion engine were adapted to work with hydrogen in the Table 1 e The test engine and dyno specifications.
laboratory. Engine manufacturer Ferryman 4-stroke
CFR engine
Test engine and dyno Aspiration Natural
Number of cylinders 1
Bore  stroke [mm] 90  120
A single-cylinder, four-stroke, naturally aspirated, water-
Cylinder volume [cm3] 799
cooled, CFR (cooperative fuel research) engine was operated.
Compression ratio 19 (adjusted for this study)
It can be operated as both SI (spark ignition) and CI Scavenge volume [cc] 765
(compression ignition) engine by means of its variable Speed range minemax [rpm] 600e2000
compression ratio. The CFR engine was operated on CI mode, Number of intake & exhaust 1&1
and the compression ratio was set to 19:1. The engine was valves
coupled with a DC dynamometer, and the engine torque was Cooling Water cooled
Dyno type & power [kW] DC & 7.5
measured by a load cell. Details about the test engine and DC
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5 659

Fig. 1 e Schematic view of the experimental setup.

reached regime temperature (coolant inlet and coolant outlet

Experimental procedure temperature), and after steady-state operating conditions
were provided, the tests were conducted. This procedure was
A single-cylinder, 4-stroke, water-cooled CFR engine was repeated for all hydrogen energy fractions (0%, 22%, and 53%).
coupled with a DC dynamometer, and all the experiments The engine brake power value was fixed at the same value
were carried out with this test bench. The hydrogen fuel that (3.07 kW). Details about experimental procedure are shown in
was used in the tests has 99.99% purity. Diesel fuel met EN 590 Table 2. Any combustion problems such as backfiring, pre-
standard, and it was injected into cylinder 22oBTDC (before ignition, or knock were not seen during experimental tests.
top dead center) during compression stroke. In addition, The accuracies and calculated uncertainties according to
hydrogen fuel was injected into intake manifold at TDC (top Kline and McClintock method [35] are given in Table 3.
dead center) during intake stroke. All of the tests were per-
formed at the ESC (European stationary cycle) at 1100 rpm
engine speed for test engine and 100% engine load (full load) Results and discussion
condition. Three different hydrogen energy fractions, which
are 0%, 22%, and 53% of total fuel, were applied and compared Brake thermal efficiency
with each other. Firstly, the test engine was warmed up until it
The variation of brake thermal efficiency versus hydrogen
energy fraction is presented in Fig. 3. The brake thermal effi-
ciency decreased with increasing hydrogen amount. The
brake thermal efficiency is obtained as 29.43 at 0% hydrogen
energy fraction (pure diesel), as 27.06 at 22% hydrogen energy
fraction, and as 23.04 at 53% hydrogen energy fraction. The
brake thermal efficiency is reduced to 8.02% and 21.69% with
22% and 53% hydrogen addition on energy basis compared
with pure diesel (0% hydrogen), respectively. Normally,
hydrogen flame speed is much higher than diesel and other

Table 2 e Experimental conditions.

Engine Engine Start of H2 energy Diesel
brake speed diesel fraction energy
power injection fraction
3.07 kW 1100 rpm 22 BTDC 0% 100%
3.07 kW 1100 rpm 22 BTDC 22% 78%
3.07 kW 1100 rpm 22 BTDC 53% 47%
Fig. 2 e Schematic diagram of the hydrogen fuel line.
660 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5

400
Table 3 e Accuracies and the calculated uncertainties.
Measured Measurement Accuracy
parameter device
300
Engine torque Load cell ±0.05 Nm
Engine speed Incremental encoder ±5 rpm

Bsfc [g/kWh]
Diesel flow rate Sika VZ 0.2 ±1% (of reading)
H2 flow rate New-Flow TLF 23 ±1% (F.S.) 200
CO AVL DiCom 4000 0.01% Vol.
CO2 AVL DiCom 4000 0.1% Vol.
HC AVL DiCom 4000 1 ppm
100
NOx AVL DiCom 4000 1 ppm
Smoke AVL 415S 0.4% Vol.
Calculated results Uncertainty
Power ±0.43% 0
BSFC ±1.08% (0% H2) 0 22 53
±1.59% (22% H2) Hydrogen energy fraction [%]
±1.38% (53% H2)
Fig. 4 e Effect of hydrogen enrichment (0%, 22% and 53% on
energy basis) on the brake specific fuel consumption at
petroleum-based fuels [36,37]. For this reason, hydrogen- 1100 rpm constant engine speed and 3.07 kW constant
enriched diesel fuel burns more completely and peak-in- engine power.
cylinder pressure increases with hydrogen addition [3].
Nevertheless, Varde and Frame [38] found that hydrogen
addition can decrease brake thermal efficiency. This is
because of the position of the piston during peak-in-cylinder heating values) of fuels. Then, the equivalent diesel value of
pressure. The diesel injection advance should be modified to consumed hydrogen and consumed diesel fuel was added.
prevent this inefficiency. Moreover, Owston et al. [39] showed The BSFC value was increased with hydrogen addition as
that heat flux of hydrogen is higher than other conventional shown in Fig. 5. With 0%, 22%, and 53% hydrogen addition of
fuels. The thermal loss increases with hydrogen addition. The total fuel on energy basis, the BSFC values were obtained as
results of this study are consistent with the results of Varde 287.8, 312.8, and 367.5, respectively. Increases of 8.72% and
and Frame [38] and Zhou et al. [40] in terms of brake thermal 27.71% were calculated for 22% and 53% hydrogen addition
efficiency. according to pure diesel (0% hydrogen). In this study, the BSFC
is negatively affected from increasing percentage of hydrogen,
Brake-specific fuel consumption since injection advance of diesel injector was not modified
according to hydrogen enrichment, and combustion period is
The variation of brake-specific fuel consumption (BSFC) not appropriate for the piston top dead center as acknowl-
versus hydrogen energy fraction is depicted in Fig. 4. The BSFC edged in the work of Varde and Frame [38]. Furthermore, with
is found according to diesel fuel, and consumed hydrogen fuel an increasing amount of heat flux with hydrogen addition, the
value was converted into diesel fuel according to LHVs (lower thermal efficiency is affected negatively [40]. Similar results
regarding BSFC were found in a study by Zhou et al. [40] and
Varde and Frame [38].
40
3,0
Brake thermal effi ci ency [%]

30 2,5

2,0
CO [g/kWh]

20
1,5

10 1,0

0,5
0
0 22 53 0,0
Hydrogen energy fraction [%] 0 22 53
Hydrogen energy fraction [%]
Fig. 3 e Effect of hydrogen enrichment (0%, 22% and 53% on
energy basis) on the brake thermal efficiency at 1100 rpm Fig. 5 e Effect of hydrogen enrichment (0%, 22% and 53% on
constant engine speed and 3.07 kW constant engine energy basis) on CO emission at 1100 rpm constant engine
power. speed and 3.07 kW constant engine power.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5 661

Carbon monoxide enrichment for high-engine loads on THC emissions can be

ignored, according to their study. The results of Zhou et al. [40]
Carbon monoxide emissions versus hydrogen energy content are consistent with results of this study.
is illustrated in Fig. 5. CO emission was measured as 2.52 g/
kWh under 0% hydrogen, 0.82 g/kWh under 22% hydrogen, Smoke
and 0.77 g/kWh under 53% hydrogen energy fraction. The
improvement in CO emissions may be higher, but AVL Dicom The variation of smoke emissions versus hydrogen energy
4000 exhaust analyzer could not measure values lower than fraction is given in Fig. 7. A great improvement was achieved
0.01% by volume. Therefore, the improvement in CO emis- with hydrogen enrichment. The smoke emission level
sions is more than obtained results. However, 67.2% and 69.2% decreased to 1.29 g/kWh from 2.91 g/kWh with 22% hydrogen
improvements were obtained with 22% and 53% hydrogen enrichment on energy basis. The smoke value was decreased
enrichment compared with neat diesel, respectively. The to 0.94 g/kWh from 2.91 g/kWh with 53% hydrogen enrich-
diffusion coefficient of hydrogen is higher than other con- ment. A reduction of 43.6% and 58.6% was obtained with 22%
ventional fuels, and it causes a decrease in heterogeneity of and 53% hydrogen addition, respectively, in comparison to
combustible mixture. The availability of oxygen increases neat diesel (0% hydrogen). Only water is emitted from
with hydrogen enrichment [41]. According to the results, only hydrogen fuel combustion, and it does not reveal any partic-
a small amount of carbon monoxide is emitted from the en- ulate matter [1]. Moreover, the improvement in smoke emis-
gine because of pilot diesel fuel and lube oil [1]. Bari and sions was achieved because hydrogen fuel does not contain
Esmaeil [3], Bose and Maji [1], and Zhou et al. [40] obtained any carbon element [8]. Also, higher diffusion coefficient of
similar results with this study in terms of carbon monoxide. hydrogen in air provides a decrease in the heterogeneity of the
mixture, the oxygen availability is improved, and conse-
Total hydrocarbon quently combustion efficiency is increased with hydrogen
addition [24,45]. For this reason, smoke emissions decreased
Hydrocarbons are formed because of the incomplete com- with an increase in hydrogen percentage. Zhou et al. [40] and
bustion of fuel, and they are organic-based compounds. The Bose and Maji [1] found similar results with this study in terms
unburned or partially burned hydrocarbons (HC) in tail pipes of smoke emissions.
are usually known as total hydrocarbon (THC) [4]. Fig. 6 shows
the variation of total hydrocarbons versus hydrogen energy Oxides of nitrogen
fraction. The THC emissions are measured as 0.025 g/kWh,
0.032 g/kWh, and 0.046 g/kWh with 0%, 22%, and 53% hydrogen Effect of hydrogen addition (0%, 22%, and 53% on energy basis)
enrichment, respectively. The accuracy of AVL Dicom 4000 on the NOx emission at 1100 rpm constant engine speed and
exhaust analyzer is 1 ppm, and it could not measure differ- 3.07 kW constant engine power is depicted in Fig. 8. With
ences below 1 ppm. For this reason, although THC emissions increasing percentage of hydrogen, specific NOx emissions
were increased with increasing percentage of hydrogen increase. However, only a slight increase was measured with
enrichment, the differences of THC values were very small and 22% hydrogen addition on energy basis, a dramatic rise
could be ignored. In addition to this, the measured THC values observed with 53% hydrogen addition compared with neat
are below the regulation limits. Zhou et al. [40] found that diesel (0% hydrogen). Specific oxides of nitrogen emissions
hydrocarbon emissions increased at some partial engine loads (NOx) increased to 3.72 g/kWh from 3.33 g/kWh with 22%
with hydrogen addition. Moreover, the effect of hydrogen hydrogen addition according to neat diesel. Specific NOx

0,05 2,5

0,04 2,0
Smoke [g/kWh]
THC [g/kW h]

0,03 1,5

0,02 1,0

0,01 0,5

0,00 0,0
0 22 53 0 22 53
Hydrogen energy fraction [%] Hydrogen energy fraction [%]

Fig. 6 e Effect of hydrogen enrichment (0%, 22% and 53% on Fig. 7 e Effect of hydrogen enrichment (0%, 22% and 53% on
energy basis) on THC emission at 1100 rpm constant energy basis) on the smoke emission at 1100 rpm constant
engine speed and 3.07 kW constant engine power. engine speed and 3.07 kW constant engine power.
662 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5

12 Cylinder gas pressure

10 The effect of a different percentage of hydrogen enrichment

on cylinder gas pressure related to crank angle at 1100 rpm
8 engine speed and full engine load is depicted in Fig. 9. It is
NOx [g/kWh]

shown that peak-in-cylinder pressure value increases pro-

portionally with an increasing amount of hydrogen. Results
6
show that peak in cylinder rises to 59.82 bar from 55.48 bar
with 22% hydrogen addition and peak in cylinder increases to
4
75.57 bar from 55.48 bar with 53% hydrogen enrichment on
energy basis at 3.07 kW engine power and 1100 rpm constant
2 engine speed. According to acquired results, 7.81% and 36.20%
increases were obtained with 22% and 53% hydrogen addition,
0 respectively. High flame speed of hydrogen provides rapid
0 22 53
combustion of combustible mixture, and peak-in-cylinder
Hydrogen energy fraction [%]
temperature rises with an increasing amount of hydrogen,
Fig. 8 e Effect of hydrogen addition (0%, 22% and 53% on especially with 53% hydrogen enrichment. The results of this
energy basis) on the NOx emission at 1100 rpm constant study are similar to those of the study by Christodoulou and
engine speed and 3.07 kW constant engine power. Megaritis [48].

Heat release rate

emissions increased to 11.23 g/kWh from 3.33 g/kWh with 53% Fig. 10 shows the effect of different percentage of hydrogen
hydrogen addition. According to obtained results, an 11.73% enrichment on heat-release rate related to crank angle at
and 237.47% increase was calculated in NOx values with 22% 1100 rpm engine speed and full engine load. With an
and 53% hydrogen addition compared with neat diesel, increasing amount of hydrogen, peak heat-release-rate value
respectively. increases proportionally. With hydrogen addition, explosive
NOx emissions are basically dependent on reaction dura- type combustion is observed instead of classical diesel com-
tion, gas temperature, and oxygen availability [7]. Normally, bustion [8]. High flame speed of hydrogen increases the causes
hydrogen addition causes an increase in cylinder gas tem- to premixed combustion-phase-dominated heat-release rate.
perature; therefore, it is assumed that NOx formation in- The released energy value in premixed combustion phase is
creases with hydrogen addition [3]. Despite this, White et al. increased and released energy value in mixing controlled
[42] found that NOx formation is related to excess air ratio and combustion phase is reduced with increasing hydrogen level.
that it may decrease if there is not appropriate excess air ratio However, combustion problems such as knock, backfiring,
value. Also, Rortveit et al. [43] showed that N2, He, and CO2 and pre-ignition may happen with this type of explosive
gases dilute in-cylinder combustible gas mixture, and they combustion. For this reason, the hydrogen energy fraction did
found that NOx emissions may decrease significantly if the not exceed 53% since knock problem was seen time to time
level of gases exceeds critical value. Frassoldati et al. [44] after this hydrogen level. The peak heat release rate increases
found that NOx emissions can be limited with dilution of to 35.20 J/o from 27.99 J/o with 22% hydrogen enrichment.
combustible mixture. Moreover, it is known that there are lots
of studies that use N2 to simulate EGR in internal combustion 80
engines (ICEs) [45,46]. 0% hydrogen
On the other part, the adiabatic flame speed of hydrogen is 22% hydrogen
60 53% hydrogen
higher than other petroleum fuels [36,37]. The ideal thermo-
In-cylinder pressure [bar]

dynamic cycle is more reachable with hydrogen enrichment

thanks to unique combustion properties of hydrogen.
40
Furthermore, the LHV of hydrogen is about three times as
diesel fuel. Therefore, peak gas temperature value and gas
pressure value increase with increasing percentage of 20
hydrogen [36,37]. NOx emission could be increased with
hydrogen addition because increasing cylinder gas tempera-
ture with hydrogen addition is one of the most important 0
reasons for NOx formation. However, if the effects of hydrogen
(both dilution effect and peak cylinder gas temperature) are
considered together, a slight increase was seen in NOx emis- 250 300 350 400 450
sions with 22% hydrogen addition and a dramatic increase Crank angle [degree]
was observed that with 53% hydrogen addition. According to
obtained results, the percentage of hydrogen has a huge effect Fig. 9 e Effect of different percentage of hydrogen
on NOx emissions. The results of this study are consistent enrichment on cylinder gas pressure related to crank angle
with the results of Pan et al. [47] and Zhou et al. [40]. at 1100 rpm engine speed and full engine load.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5 663

80 cylinder gas pressure values were obtained, then cylinder gas

0% hydrogen pressure was found using PV ¼ mRT equation. The obtained
22% hydrogen maximum cylinder gas values were 1360  C, 1355  C, and
60 53% hydrogen
1453  C for 0%, 22%, and 53% hydrogen enrichment, respec-
Heat release rate [J/oCA]

tively. The maximum cylinder temperature was reduced to

40 0.39% with 22% hydrogen addition compared with pure diesel
situation. Also, the maximum cylinder temperature increased
to 6.85% with 53% hydrogen enrichment according to neat
20
diesel. The radicals of O and OH accelerate the reaction with
hydrogen addition, and the combustion duration may be
0 shortened; thus, local cylinder gas temperature and pressure
values increased [22]. Furthermore, higher flame speed of
hydrogen than petroleum-based fuels shortens combustion
-20
300 320 340 360 380 400 420 440
duration. On the other hand, the exhaust losses and cooling
losses may be improved with hydrogen addition and, thus,
Crank angle [degree]
increased cylinder gas temperature with hydrogen addition is
Fig. 10 e Effect of different percentage of hydrogen not a surprise [6]. However, a slight decrease was obtained
enrichment on heat release rate related to crank angle at with 22% hydrogen addition. This unimportant decrease in
1100 rpm engine speed and full engine load. cylinder gas temperature is related to the position of the pis-
ton in cylinder during the combustion period. Since injection
advance of diesel injector was not modified and optimized
according to dual fuel combustion, a slight decrease occurred.
Also, peak heat release rate increases to 59.04 J/o from 27.99 J/o
with 53% hydrogen addition. The increases in heat release rate Indicator diagram
are 25.77% and 110.94% with 22% and 53% hydrogen enrich-
ment compared with neat diesel fuel, respectively. It is clear Fig. 12 shows the effect of different percentage of hydrogen
that an explosive type combustion period occurred with enrichment on cylinder gas pressure related to cylinder vol-
hydrogen addition due to hydrogen injection into intake air. ume at 1100 rpm engine speed and full engine load. According
The results of Saravanan and Nagarajan [49] are consistent to obtained results, an instantaneous and explosive type
with the results of this study in terms of heat-release rate. combustion period is shown with hydrogen enrichment on
PeV diagram (pressureevolume diagram). On the other hand,
exhaust losses and incomplete combustion losses improved
Cylinder gas temperature
with hydrogen addition thanks to unique combustion prop-
erties of hydrogen. Higher diffusion coefficient of hydrogen
Effect of different percentage of hydrogen enrichment on in-
compared to diesel fuel provides more homogenous
cylinder gas temperature related to crank angle at 1100 rpm
combustible mixture. Also, higher flame speed of hydrogen
engine speed and full engine load is illustrated in Fig. 11. At
compared to fossil fuels causes a more complete combustion.
least 100 cycles were measured in terms of cylinder pressure,
Similar results were obtained in the study by Christodoulou
and average values were used to calculate cylinder gas tem-
and Megaritis [48].
perature. First of all, using a Kistler 6061B pressure sensor

1600
0% hydrogen
1400
22% hydrogen
53% hydrogen
In-cylinder temperature [oC]

1200

1000

800

600

400

200

0 100 200 300 400 500 600 700

Crank angle [degree]

Fig. 11 e Effect of different percentage of hydrogen Fig. 12 e Effect of different percentage of hydrogen
enrichment on cylinder gas temperature related to crank enrichment on cylinder gas pressure related to cylinder
angle at 1100 rpm engine speed and full engine load. volume at 1100 rpm engine speed and full engine load.
664 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5

Conclusion Acknowledgement

According to the study, the engine performance (BSFC and This research was supported by the Yıldız Technical Univer-
brake thermal efficiency), emissions (CO, THC, smoke, and sity Scientific Research Projects Coordination Department.
NOx) were tested, and combustion characteristics (in-cylinder Project Number: 2011-06-01-YULAP01. Also, the authors are
pressure, heat release rate, in-cylinder temperature and in- indebted to Türk Trakto€ r ve Ziraat Makineleri A.S‚. for test
dicator diagram) was analyzed on a single-cylinder CFR en- apparatus and equipment donation.
gine at 1100 rpm constant engine speed, 3.07 kW constant
brake engine power, and different hydrogen levels (0%, 22%, Nomenclature
and 53%).
Obtained test results are listed below: BSFC Brake specific fuel consumption
CFR Cooperative fuel research
a. The brake thermal efficiency is decreased with an CI Compression ignition
increasing amount of hydrogen. The brake thermal effi- CNG Compressed natural gas
ciency is obtained as 29.43 at 0% hydrogen energy fraction CO Carbon monoxide
(neat diesel), as 27.06 at 22% hydrogen energy fraction, and CO2 Carbon dioxide
as 23.04 at 53% hydrogen energy fraction. The brake ther- ECU Electronic control unit
mal efficiency is reduced to 8.02% and 21.69% with 22% and GHG Greenhouse gases
53% hydrogen addition, respectively, on energy basis H2 Hydrogen molecule
compared with neat diesel. HC Hydrocarbons
b. The CO emission was measured as 2.52 g/kWh under 0% HCCI Homogenous charge compression ignition
hydrogen, as 0.82 g/kWh under 22% hydrogen and 0.77 g/ He Helium
kWh under 53% hydrogen energy fraction. The improve- IC Internal combustion
ment in CO emissions may be higher, but AVL Dicom 4000 ICE Internal combustion engine
exhaust analyzer could not measure values lower than LPG Liquefied petroleum gas
0.01% by volume. Therefore, the improvement in CO N2 Nitrogen
emissions is supposed to be much more than obtained NOx Oxides of nitrogen
results. O2 Oxygen molecule
c. THC emissions are measured as 0.025 g/kWh, 0.032 g/kWh, SI Spark ignition
and 0.046 g/kWh with 0%, 22%, and 53% hydrogen enrich- THC Total unburned hydrocarbons
ment, respectively. The accuracy of AVL Dicom 4000
exhaust analyzer is 1 ppm, and it could not measure dif-
ferences below 1 ppm. However, THC emissions were references
increased with an increasing percentage of hydrogen
enrichment. The differences of THC values are very small,
and they could be ignored. Also, the measured THC values [1] Bose PK, Maji D. An experimental investigation on engine
performance and emissions of a single cylinder diesel engine
are below the regulation limits.
using hydrogen as inducted fuel and diesel as injected fuel
d. A great improvement was achieved with hydrogen with exhaust gas recirculation. Int J Hydrogen Energy
enrichment on smoke emissions. The smoke emission 2009;34:4847e54.
level dropped to 1.29 g/kWh from 2.91 g/kWh with 22% [2] Yoon SH, Han SC. Effects of high EGR rate on dimethyl ether
hydrogen enrichment on energy basis. The smoke value (DME) combustion and pollutant emission characteristics in
decreased to 0.94 g/kWh from 2.91 g/kWh with 53% a direct injection diesel engine. Energies 2013;6:5157e67.
[3] Bari S, Esmaeil MM. Effect of H2/O2 addition in increasing the
hydrogen enrichment. A reduction of 43.6% and 58.6% was
thermal efficiency of a diesel engine. Fuel 2010;89:378e83.
obtained with 22% and 53% hydrogen addition, respec-
[4] Heywood JB. Internal combustion engine fundamentals. New
tively, according to neat diesel. York, NY, USA: McGraw Hill, Inc; 1988.
e. A remarkable result was obtained on oxides of nitrogen [5] Tesfa B, Gu F, Mishra R, Ball A. Emission characteristics of a
formation with hydrogen enrichment. According to ob- CI engine running with a range of biodiesel feedstocks.
tained results, an 11.73% increase on NOx emissions was Energies 2014;7:334e50.
found with 22% hydrogen addition, and a dramatic in- [6] Sopena C, Die guez PM, Sa inz D, Urroz JC, Guelbenzu E,
Gandı́a LM. Conversion of a commercial spark ignition
crease (237.47%) was obtained with 53% hydrogen enrich-
engine to run on hydrogen: performance comparison using
ment compared with neat diesel fuel.
hydrogen and gasoline. Int J Hydrogen Energy
f. Maximum cylinder gas pressure values increased by 2010;35:1420e9.
7.81% and 36.20% with 22% and 53% hydrogen addition, [7] Ko€ se H, Ciniviz M. An experimental investigation of effect on
according to pure diesel fuel, respectively. Peak heat diesel engine performance and exhaust emissions of
release rate values with hydrogen addition increased by addition at dual fuel mode of hydrogen. Fuel Process Technol
25.77% and 110.94% with 22% and 53% hydrogen enrich- 2013;114:26e34.
[8] Saravanan N, Nagarajan G, Dhanasekaran C, Kalaiselvan KM.
ment in comparison to those with neat diesel,
Experimental investigation of hydrogen port fuel injection in
respectively. DI diesel engine. Int J Hydrogen Energy 2007;32:4071e80.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 1 ( 2 0 1 6 ) 6 5 6 e6 6 5 665

[9] Nieminen J, D'Souza N, Dincer I. Comparative combustion [30] Ikegami M, Miwa M, Shioji M. A study on hydrogen fuelled
characteristics of gasoline and hydrogen fuelled ICEs. Int J compression ignition engines. Int J Hydrogen Energy
Hydrogen Energy 2010;35:5114e23. 1982;7:341e53.
[10] Liu Y, Yang J, Sun J, Zhu A, Zhou QA. Phenomenological [31] Roy MM, Tomita E, Kawahara N, Harada Y, Sakane A. An
model for prediction auto-ignition and soot formation of experimental investigation on engine performance and
turbulent diffusion combustion in a high pressure common emissions of a supercharged H2-diesel dual-fuel engine. Int J
rail diesel engine. Int J Hydrogen Energy 2011;4:894e912. Hydrogen Energy 2010;35:844e53.
[11] Ryu K. Effects of pilot injection pressure on the combustion [32] Sierens R, Verhelst S. Hydrogen fuelled V-8 engine for city-
and emissions characteristics in a diesel engine using bus application. Int J Hydrogen Energy 2001;2:39e45.
biodiesel-CNG dual fuel. Energy Convers Manag [33] Mohammadi A, Shioji M, Yasuyuki N, Ishikura W, Eizo T.
2013;76:506e16. Performance and combustion characteristics of a direct
[12] Cacua K, Amell A, Cadavid F. Effects of oxygen enriched air injection SI hydrogen engine. Int J Hydrogen Energy
on the operation and performance of a diesel-biogas dual 2007;32:296e304.
fuel engine. Biomass Bioenergy 2012;45:159e67. [34] Sandalcı T, Karago€ z Y. Experimental investigation of the
[13] Huang J, Wang Y, Li S, Roskilly AP, Yu H, Li H. Experimental combustion characteristics, emissions and performance of
investigation on the performance and emissions of a diesel hydrogen port fuel injection in a diesel engine. Int J Hydrogen
engine fuelled with ethanol-diesel blends. Appl Therm Eng Energy 2014;39:18480e9.
2009;29:2484e90. [35] Kline SJ, McClintock FA. Describing uncertainties in single-
[14] Al-Baghdadi MARS. Effect of compression ratio, equivalence sample experiments. Mech Eng 1953;75:3e8.
ratio and engine speed on the performance and emission [36] Verhelst S, Woolley R, Lawes M, Sierens R. Laminar and
characteristics of a spark ignition engine using hydrogen as a unstable burning velocities and Markstein lengths of
fuel. Renew Energy 2004;29:2245e60. hydrogen-air mixtures at engine-like conditions. Int J
[15] Sastri MVC. Hydrogen energy research-and-development in Hydrogen Energy 2005;30:209e16.
India e an overview. Int J Hydrogen Energy 1987;12:137e45. [37] Kumar MS, Ramesh A, Nagalingam B. Use of hydrogen to
[16] Saravanan N, Nagarajan G. An experimental investigation of enhance the performance of a vegetable oil fuelled
hydrogen-enriched air induction in a diesel engine system. compression ignition engine. Int J Hydrogen Energy
Int J Hydrogen Energy 2008;33:1769e75. 2003;10:1143e54.
[17] McLellan B, Shoko E, Dicks AL, Diniz da Costa JC. Hydrogen [38] Varde KS, Frame GA. Hydrogen aspiration in direct injection
production and utilisation opportunities for Australia. Int J type diesel engine-its effect on smoke and other engine
Hydrogen Energy 2005;30:669e79. performance parameters. Int J Hydrogen Energy
[18] Verhelst S, Sierens R. Hydrogen engine-specific properties. 1983;8:549e55.
Int J Hydrogen Energy 2001;26:987e90. [39] Owston R, Magi V, Abraham J. Wall interactions of hydrogen
[19] Bauer CG, Forest TW. Effect of hydrogen addition on the flames compared with hydrocarbon flames. SAE Tech Paper
performance of methane-fueled vehicles, part I: effect on 2007-01-1466 2007.
S.I. engine performance. Int J Hydrogen Energy [40] Zhou JH, Cheung CS, Leung CW. Combustion, performance,
2001;26:55e70. regulated and unregulated emissions of a diesel engine with
[20] Das LM. Hydrogen engines: a review of the past and a look hydrogen addition. Appl Energy 2014;126:1e12.
into the future. Int J Hydrogen Energy 1990;15:425e43. [41] Ghazal OH. Performance and combustion characteristic of CI
[21] Ma F, Wang Y, Liu H, Li Y, Wang J, Zhao S. Experimental engine fueled with hydrogen enriched diesel. Int J Hydrogen
study on thermal efficiency and emission characteristics of a Energy 2013;38:15469e76.
lean burn hydrogen enriched natural gas engine. Int J [42] White CM, Steeper RR, Lutz AE. The hydrogen-fueled internal
Hydrogen Energy 2007;32:5067e75. combustion engine: a technical review. Int J Hydrogen
[22] Liew C, Li H, Nuszkowski J, Liu S, Gatts T, Atkinson R, et al. Energy 2006;31:1292e305.
An experimental investigation of the combustion process of [43] Rortveit GJ, Hustad JE, Li SC, Williams FA. Effects of diluents
a heavy-duty diesel engine enriched with H2. Int J Hydrogen on NOx formation in hydrogen counterflow flames. Combust
Energy 2010;35:11357e65. Flame 2002;130:48e61.
[23] Gomes-Antunes JM, Mikalsen R, Roskilly AP. An investigation [44] Frassoldati A, Faravelli T, Ranzi E. A wide range modeling
of hydrogen fuelled HCCI engine performance and operation. study of NOx formation and nitrogen chemistry in hydrogen
Int J Hydrogen Energy 2008;33:5823e8. combustion. Int J Hydrogen Energy 2006;31:2310e28.
[24] Szwaja S, Grab-Rogalinski K. Hydrogen combustion in a [45] Christodoulou F, Megaritis A. Experimental investigation of
compression ignition diesel engine. Int J Hydrogen Energy the effects of simultaneous hydrogen and nitrogen addition
2009;34:4413e21. on the emissions and combustion of a diesel engine. Int J
[25] Clenci A, Bı̂zı̂iac A, Podevin P, Descombes G, Deligant M, Hydrogen Energy 2014;39:2692e702.
Niculescu R. Idle operation with low intake valve lift in a port [46] Wang Z, Zuo H, Liu Z, Li W, Dou H. Impact of N2 dilution on
fuel injected engine. Energies 2013;6:2874e91. combustion and emissions in a spark ignition CNG engine.
[26] Miyamoto T, Hasegawa H, Mikami M, Kojima N, Energy Convers Manag 2014;85:354e60.
Kabashima H, Urata Y. Effect of hydrogen addition to intake [47] Pan H, Pournazeri S, Princevac M, Miller JW, Mahalingam S,
gas on combustion and exhaust emission characteristics of a Khan MY, et al. Effect of hydrogen addition on criteria and
diesel engine. Int J Hydrogen Energy 2011;36:13138e49. greenhouse gas emissions for a marine diesel engine. Int J
[27] Jeong K, Lee D, Park S, Lee CS. Effect of two-stage fuel Hydrogen Energy 2014;39:11336e45.
injection parameters on NOx reduction characteristics in a [48] Christodoulou F, Megaritis A. Experimental investigation of
DI diesel engine. Energies 2011;4:2049e60. the effects of separate hydrogen and nitrogen addition on
[28] Meloni R, Naso V. An insight into the effect of advanced the emissions and combustion of a diesel engine. Int J
injection strategies on pollutant emissions of a heavy-duty Hydrogen Energy 2013;38:10126e40.
diesel engine. Energies 2013;6:4331e51. [49] Saravanan N, Nagarajan G. Experimental investigation on
[29] Gomes-Antunes JM, Mikalsen R, Roskilly AP. An performance and emission characteristics of dual fuel DI
experimental study of a direct injection compression ignition diesel engine with hydrogen fuel. SAE Paper No. 2009-26-032
hydrogen engine. Int J Hydrogen Energy 2009;34:6516e22. 2009.