Accepted Manuscript

Experimental Investigation of Autoignition of Hydrogen-Air Charge in a

Compression Ignition Engine under Dual-Fuel mode

V. Chintala, K.A. Subramanian

PII: S0360-5442(17)31244-6

DOI: 10.1016/j.energy.2017.07.068

Reference: EGY 11256

To appear in: Energy

Received Date: 02 November 2015

Revised Date: 26 May 2017

Accepted Date: 11 July 2017

Please cite this article as: V. Chintala, K.A. Subramanian, Experimental Investigation of
Autoignition of Hydrogen-Air Charge in a Compression Ignition Engine under Dual-Fuel mode,
Energy (2017), doi: 10.1016/j.energy.2017.07.068

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form.
Please note that during the production process errors may be discovered which could affect the
content, and all legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Experimental Investigation of Autoignition of Hydrogen-Air Charge in a Compression

2 Ignition Engine under Dual-Fuel mode

4 V. Chintala and K. A. Subramanian*

5 Engines and Unconventional Fuels Laboratory,

6 Centre for Energy Studies, Indian Institute of Technology- Delhi, New Delhi – 110 016, India.

7 *E-mail address: subra@ces.iitd.ernet.in

9 Abstract

10 High amount of hydrogen substitution in a compression ignition (CI) engine under dual fuel

11 mode is limited due to more probability of autoignition of hydrogen-air charge and knocking

12 problem. The study deals with analysis of autoignition of hydrogen-air charge in a 7.4 kW rated

13 power output of CI engine under dual fuel mode (diesel-hydrogen) at 100% load (Case I) and

14 50% load (Case II). Experimental results indicate that the significant increase in in-cylinder

15 temperature is the predominant factor for autoignition of hydrogen-air charge. The in-cylinder

16 temperature increased due to combustion advancement with hydrogen addition into the engine.

17 Computational fluid dynamics (CFD) simulation study also confirms the combustion

18 advancement with hydrogen addition in the engine. Experimental tests were extended further

19 with water injection into the engine under dual fuel mode (Case III). A clear conclusion emerged

20 from the study is that the hydrogen-air charge gets autoignite without any external ignition aid

21 when the reactants temperature is about 953 K ± 8 K. It could also be observed that knock

22 limited hydrogen energy share in the engine at 100% load was increased from 18.8% with

1
 ACCEPTED MANUSCRIPT

23 conventional dual fuel mode to 60.7% with water injection due to decrease in in-cylinder

24 temperature.

25

26 Keywords: Dual-fuel engine; Hydrogen energy share; Autoignition; In-cylinder temperature;

27 Knock limited factor, Computational fluid dynamics.

28

29 1. Introduction

30 Hydrogen (H2) is being considered as a supplementary fuel for internal combustion engines in

31 order to yield the twin benefits of energy efficiency improvement and emissions reduction [1, 2].

32 The Ministry of New and Renewable Energy, Government of India envisaged in its roadmap that

33 one million automotive vehicles are to be fueled with hydrogen by 2020 [3]. The comparative

34 analysis of hydrogen, electric and biofuel transitional pathways to a future sustainable road

35 transport in a renewable-based energy system shows that hydrogen scenario could be

36 advantageous in reducing fuel import and consumer total fuel costs [4]. Hydrogen as a fuel is

37 more suitable for spark ignition (SI) engines due to its high octane number [5]. As hydrogen

38 fueled SI engines face major setbacks of power drop and back firing, a dedicative system needs

39 to be developed for effective utilization of hydrogen in SI engines [6]. In an alternative way,

40 hydrogen could be used in compression ignition (CI) engines under dual fuel mode (hydrogen-

41 diesel) without power drop and back firing problems. In addition, there is no need of major

42 engine hardware modifications for a CI engine to be operated under dual fuel mode. Hydrogen

43 based dual fuel engines offer significant benefits such as high energy efficiency [7], high

44 combustion efficiency, lower specific energy consumption [8], near zero carbon based emissions

45 (hydro carbon (HC), carbon monoxide (CO) and smoke/particulate matter), and less greenhouse

2
 ACCEPTED MANUSCRIPT

46 gas emissions (CO2, CH4 and N2O) [4, 9]. Several investigations were reported on utilization of

47 hydrogen in CI engines under dual fuel mode with exploration of improved performance and

48 emission characteristics. For example, energy efficiency improved about 14% with 20%

49 hydrogen addition in a single cylinder direct injection CI engine (7.4 kW rated power at 1500

50 rpm) under dual fuel mode [10]. Yadav et al. found 11.6% improvement in thermal efficiency of

51 a CI engine (4.4 kW rated power at 1500 rpm) with an addition of 120 g/h hydrogen under dual

52 fuel mode due to better combustion characteristics of hydrogen [11]. The maximum brake

53 thermal efficiency of 39.53% was obtained with hydrogen addition in a CI engine (5.2 kW rated

54 power at 1500 rpm) at 60% load [12]. Wu H-W and Wu Z-Y reported a significant improvement

55 in thermal efficiency of a CI engine under dual-fuel mode with 30% hydrogen energy share at

56 100% load [13]. Edwin et al. also confirmed the same that thermal efficiency of a CI engine

57 increased from 29.9% with base diesel mode to 31.6% with 10.1% hydrogen energy share at

58 100% load [14]. The reasons for this improvement could be due to better mixing characteristics

59 of these gaseous fuel with air resulting in better combustion. Even though there are significant

60 benefits in terms of thermal efficiency improvement and emissions reduction, hydrogen based

61 dual fuel engines have a severe problem of knocking with high amounts of hydrogen substitution

62 [10, 15-17]. The literature information on the maximum amount of hydrogen substituted in CI

63 engines under dual-fuel mode are summarized in Table 1 [1, 7]. It could be observed from the

64 table that the maximum hydrogen energy share achieved is in the range from 6% to 20% at

65 moderate and high loads (BMEP range: 5 bar to 9.2 bar). But, higher amount of hydrogen about

66 30% can be substituted in the engine at low load (BMEP: 2.2 bar). It could be concluded from

67 the information given in the table that the hydrogen energy share in CI engines decreases with

3
 ACCEPTED MANUSCRIPT

68 increase in engine load. Hydrogen energy share in a 7.4 kW rated power output of compression

69 ignition engine with water addition at 100% load can be increased from 18.8% to 48.4% [10].

70

71 1.1. Review on knocking problem in dual fuel engines

72 Generally knocking in a SI engine occurs due to spontaneous ignition of a portion of the end gas

73 mixture in the combustion chamber ahead of the propagating flame. With the use of hydrogen fuel

74 in internal combustion engines, knocking may occur not only at the end stage of combustion

75 process but also at the earlier stage of combustion [5, 15, 26]. In hydrogen dual fuel engines,

76 knocking combustion was observed by some of the researchers during early stage of hydrogen

77 combustion [5, 15, 26]. Knocking during combustion is defined as abnormal combustion

78 phenomenon (abnormal rate of pressure rise) which degrades the engine performance. In dual fuel

79 engines, a gaseous fuel (main fuel: hydrogen) is generally injected into the intake manifold during

80 suction stroke and a liquid fuel (pilot fuel: diesel) is directly injected at the end of compression

81 stroke for initiating the combustion [27-29]. First, the diesel fuel gets self-ignited and act as an

82 ignition source for initiating the combustion of hydrogen-air mixture which is spread around the

83 combustion chamber. In contrast, if combustion of the hydrogen-air mixture is initiated by

84 hydrogen itself prior to diesel fuel injection, the combustion would proceed with severe knock.

85 Karim stated that the primary requirement of any gaseous fuel for satisfactory operation under dual

86 fuel mode is that its mixture with air would not autoignite spontaneously during or following the

87 rapid pilot energy release [30]. Hydrogen addition in a CI engine leads to production of knocking

88 or detonation because of its lower ignition energy, wider flammability range, and shorter

89 quenching distance [31]. With hydrogen fuel, knocking problem could happen not only at the end

90 stage of combustion process as in case of SI engines but also at the earlier stage of combustion

4
 ACCEPTED MANUSCRIPT

91 process [5, 26, 32, 33]. High amounts of hydrogen supply to CI engines under dual-fuel mode

92 create several problems including abnormal rate of pressure rise, high in-cylinder peak pressure,

93 too advanced combustion, high in-cylinder peak temperature, autoignition of premixed hydrogen-

94 air charge, and loss of available work due to advance in start of combustion [16, 34-36]. Severe

95 knocking occurs when hydrogen is self-ignited, unlike hydrogen burning under controlled ignition

96 [15, 26]. If knock occurs, the engine would be in severe damage including breakage of piston

97 rings, piston melting, and cracking of cylinder head. Therefore, a dual fuel CI engine has to be

98 operated with lower hydrogen energy share in order to avoid knock. Edwin Geo et al. reported the

99 maximum possible hydrogen energy share without knock was about 12.7% in a single cylinder CI

100 engine (4.4 kW rated power at 1500 rpm) [14]. Yadav et al. substituted the maximum hydrogen

101 energy share of 16.4% in a CI engine (4.4 kW rated power at 1500 rpm) without knocking problem

102 [37]. Szwaja et al. concluded that addition of a small amount of hydrogen (i.e. 5% hydrogen

103 energy share) has no effect on knocking [15]. However, with increasing hydrogen energy share,

104 high frequency component of in-cylinder pressure increased substantially to 4 MPa and resulted

105 knocking at about 17% hydrogen energy share in the engine at rated load (with base compression

106 ratio of 17:1) [15]. Similarly, Chintala and Subramanian reported the knocking tendency in a

107 hydrogen based dual fuel engine (7.4 kW rated power at 1500 rpm with compression ratio of

108 19.5:1) in terms of ringing intensity and concluded in their investigation that about 19% hydrogen

109 share was the maximum amount that could be substituted in the engine at 100% load for knock

110 free operation (with base compression ratio of 19.5:1) [1]. Saravanan and Nagarajan observed

111 knocking problem at about 50% hydrogen volume share in a CI engine (3.7 kW rated power at

112 1500 rpm) under dual fuel mode at rated load [38]. Varde and Frame in their experimental study

113 on a CI engine (single cylinder direct injection engine with compression ratio of 17.4:1), measured

5
 ACCEPTED MANUSCRIPT

114 acoustic noise levels in the test cell at two different locations, and observed a substantial increment

115 in the acoustic noise level beyond 11% hydrogen energy share [17]. With increasing hydrogen

116 energy share, start of combustion (SOC) advances significantly which subsequently leads to

117 knocking problem due to autoignition of premixed hydrogen-air charge [10]. With increasing

118 amount of hydrogen fuel substitution, rate of pressure rise increases at higher loads whereas it

119 decreases at low loads in a CI engine under dual-fuel mode. For example, Edwin et al. found an

120 increase in the maximum rate of pressure rise from about 5.2 bar/oCA with base diesel mode (0%

121 hydrogen energy share) to about 6.1 bar/oCA with 12.5% hydrogen energy share at 100% load

122 [14]. However, the maximum rate of pressure rise decreased from about 2.9 bar/oCA with base

123 diesel mode to about 2.3 bar/oCA with 28% hydrogen energy share at 25% load [14]. Even though

124 many studies have been reported on some combustion characteristics (in-cylinder pressure, peak

125 pressure, rate of pressure rise, and het release rate), less attention was given to an important

126 phenomenon of knocking and the reasons for knocking in case of hydrogen based dual-fuel

127 engines. Literature details on autoignition of gaseous fuel-air mixture in dual fuel engines are given

128 below;

129

130 1.2. Review on autoignition of gaseous fuel-air mixture in dual fuel engines

131 Autoignition is the term used for a rapid chemical reaction of fuel-air mixture which is not

132 initiated by any external ignition source [39]. The autoignition of a gaseous fuel-air mixture

133 occurs when the energy released by the reaction as heat is larger than the heat lost to the

134 surroundings. Tsujimura et al. studied autoignition of hydrogen-air mixture in a constant volume

135 chamber and concluded that intake air temperature has significant effect on autoignition of the

136 mixture [40]. For the intake air temperatures below 1100 K, the auto-ignition linearly depends on

6
 ACCEPTED MANUSCRIPT

137 the temperature in the Arrhenius coordinates whereas for the temperatures greater than 1100 K,

138 the temperature dependence of the auto-ignition is weak [40]. Similarly, Naber and Siebers

139 obtained a strong correlation between ignition delay and intake air temperature during the

140 investigation on a constant volume combustion chamber under diesel engine condition [33]. Start

141 of combustion advances with hydrogen addition in a CI engine under dual fuel mode. For

142 example, Szwaja et al. reported about 4o CA combustion advancement with increasing hydrogen

143 energy share to 25% in a direct injection CI engine [15]. Lata et al. found a significant

144 advancement (about 10o CA) in combustion of fuel-air mixture with increase in hydrogen

145 volume share from 4% to 16% in a multi-cylinder CI engine [41]. This may be due to hydrogen-

146 air mixture surrounding the pilot diesel fuel promotes faster initial combustion reaction rate and

147 further increase of in-cylinder temperature [41]. The start of combustion advanced by 5o CA with

148 30% hydrogen addition in a diesel engine (3.7 kW at 1500 rpm) under dual fuel mode due to

149 instantaneous combustion of gaseous-air mixture [38]. Research findings of Miyamato et al.

150 indicate that hydrogen in a CI engine was autoignited by itself without ignition assistance by

151 pilot fuel (diesel) when hydrogen fraction is higher than 8% volume in a CI engine [34].

152 Similarly, spontaneous autoignition of air-gaseous fuel (propane) was found in a diesel-propane

153 dual fuel engine due to its high reactivity nature like hydrogen fuel [35]. Wong and Karim stated

154 the reasons for autoignition of hydrogen-air charge are due to high polytropic index of hydrogen,

155 higher in-cylinder temperature, and increasing preignition chemical reactions [36].

156

157 It is clearly understood from the literature information that the start of combustion of diesel-

158 hydrogen-air mixture advances with increase in amount of hydrogen addition along with diesel

159 fuel in a CI engine. If the start of combustion advances drastically, autoignition of hydrogen-air

7
 ACCEPTED MANUSCRIPT

160 may takes place, which further leads to knocking and suppress the amount of hydrogen

161 substitution in the engine. A very few investigations on influence of high amount of hydrogen

162 substitution on autoignition of premixed hydrogen-air charge in dual fuel engines were reported

163 in the literature. However, the reasons for autoignition of the charge in dual fuel engines are not

164 addressed properly. In this context, the present study is aimed at investigation of effect of high

165 hydrogen energy share on autoignition of hydrogen-air charge in a single cylinder hydrogen dual

166 fuel engine (7.4 kW rated power at 1500 rpm). The effects of various parameters such as in-

167 cylinder pressure, in-cylinder temperature, and gaseous fuel concentration on autoignition of the

168 charge, knocking, and maximum achievable hydrogen energy share are analyzed. Experimental

169 tests were also conducted with water injection in the dual fuel engine, to assess the effect of in-

170 cylinder temperature on autoignition of hydrogen-air charge. A computational fluid dynamics

171 (CFD) study was carried out to support the fact of combustion advancement/autoignition with

172 hydrogen addition in a CI engine.

173

174 2. Experimental tests and simulation details

175 2.1. Experimental test setup for dual fuel operation

176 An experimental test setup was developed by modifying a 7.4 kW direct injection CI engine to

177 hydrogen based dual-fuel mode. A photographic view of the experimental setup is shown in

178 Figure 1 and technical specifications of the engine are given in Table 2. The test engine was

179 loaded with an eddy current dynamometer as shown in the figure. In the present study, hydrogen

180 was used as main fuel and diesel as pilot fuel. Hydrogen was injected into the intake manifold

181 using a solenoid gas injector and the liquid fuel (diesel) was directly injected into the combustion

182 chamber using a conventional high pressure liquid injector for initiating the ignition of

8
 ACCEPTED MANUSCRIPT

183 hydrogen-air mixture. The physiochemical properties of the fuels are given in Table 3 [7, 9]. An

184 electronic control unit (ECU) was used to adjust the hydrogen injection timings including start

185 and duration of fuel injection. Hydrogen was injected into the intake manifold after the outlet

186 valve closed (43o CA after TDC) in order to avoid scavenging losses [29]. Start of hydrogen

187 injection was maintained constant as 43o CA after TDC throughout the experimentation whereas

188 the hydrogen injection duration was varied with respect to the engine loading. These injection

189 timings were optimized based on better performance and lower emissions of the engine in the

190 earlier study [29].

191

192 2.2. Brief details of water injection system

193 The water injection system consists of water reservoir, submersible pump, solenoid water

194 injector, pressure indicator, flow control valves, ECU, and electronic weighing balance as shown

195 in Figure 2. The ECU, which was used for hydrogen gas injection, controls the start and end of

196 water injection timings. Timed intake port injection system was used for water injection into the

197 dual fuel engine. Water was injected into the intake port during the suction stroke at the pressure

198 of 2-3 bar. Experimental tests were carried out on the dual-fuel engine for Specific Water

199 Consumption (SWC) of 480 g/kWh at 100% load and a constant speed of 1500 rpm. The start of

200 injection for hydrogen and water kept constant as 43o and 45o CA-after TDC throughout the

201 experiments [8, 10]. The end of injection for both fluids (hydrogen and water) varied

202 independently with respect to hydrogen energy share and water consumption. Based on the

203 maximum available time for reaching the injected hydrogen/water from the point of injection to

204 the engine cylinder, the duration of injection was optimized for hydrogen/water [29].

205

9
 ACCEPTED MANUSCRIPT

206 2.3. Engine combustion diagnosis system

207 The engine combustion diagnosis system consists of an inbuilt charge amplifier, voltage

208 amplifier, and data acquisition system (Figure 3). The system was mainly attached with three

209 kinds of sensors; (i) piezoelectric pressure transducer (ii) optical encoder and (iii) piezoelectric

210 strain gauge. The piezoelectric pressure transducer with nominal sensitivity 45 Pico coulomb/bar

211 was flush mounted on the cylinder head of the engine, for measurement of instantaneous in-

212 cylinder pressure data during the engine operation. The optical encoder (720 pulses/revolution)

213 was mounted on one end of crankshaft of the engine for crank angle (CA) measurement with an

214 accuracy of 0.1o CA. The piezoelectric strain gauge pressure transducer (pressure measurement

215 range: 0 to 2000 bar) was fitted on high-pressure diesel pilot fuel line to measure the pilot fuel

216 injection pressure. The in-cylinder pressure analog signal was amplified by the charge amplifier

217 and then, the analog signal was converted to a digital signal by data acquisition system for

218 further processing of the acquired data. The in-cylinder pressure signal, in-line pilot fuel pressure

219 signal, and TDC encoder signals were processed together by the combustion analyzer. The post

220 processing software was used for processing of pressure-crank angle data that captured for 200

221 consecutive engine operating cycles.

222

223 2.4. Details of experimental tests

224 Experimental tests were carried out on the engine under hydrogen based dual fuel mode at a

225 constant speed of 1500 rpm for 100% load (Case I) and 50% load (Case II). Experiments on the

226 dual fuel engine were further extended with water injection (Specific Water Consumption (SWC)

227 of 410 g/kWh) at 100% load and a constant speed of 1500 rpm (Case III). The SWC could be

228 determined using Eq. (1) [8]. As the study is mainly focused on autoignition of the hydrogen-air

10
 ACCEPTED MANUSCRIPT

229 charge in a dual fuel engine, Case I is analyzed in detail whereas remaining other two cases

230 (Case II and Case III) are discussed briefly for understanding the in-cylinder temperature effect.

231 Details of measurement range and resolution of the instruments used in the study are given in

232 Table 4. The uncertainty associated with the various parameters including hydrogen energy

233 share were computed using Eq. (2) [7]. A computational fluid dynamics (CFD) study was also

234 carried out to understand the phenomenon of combustion advancement/autoignition with

235 increasing amount of hydrogen substitution in the engine (details are given in Section 2.5).

236

mwater
237 Specific water consumption (SWC) = Brake power × 3600 (1)

238

239 [ ∂q
1
∂q
∆q = (∂x ∆x1)2 + (∂x ∆x2)2 + ………… + (∂x ∆xn)2
2
∂q
n
] 0.5
(2)

240 Where,

241 q : f (x1, x2,…..xn) (q is the function of x1, x2,…..xn)

242 x1, x2,…..xn : Measured variables

∂q ∂q ∂q
243 ∂x 1, ∂x 2,……., ∂x n : Partial differential of calculated parameter q which depends on variables

244

245 2.5. Brief details of computational fluid dynamics (CFD) simulation

246 A three dimensional hexahedral mesh for the engine’s combustion chamber geometry (a sector of

247 72 degrees consisting cylinder and piston: 47270 Cells, 52548 nodes, and 136701 quadrilateral

248 interior faces) was created using GAMBIT software (Figure 4) and then numerical simulation was

249 carried out using ANSYS FLUENT software. Dynamic mesh model was used for dynamic motion

250 of piston where the shape of the domain is continuously changed with respect to time. Mesh

11
 ACCEPTED MANUSCRIPT

251 independence was also verified with variation of mesh elements of the given geometry. A time

252 step of 0.1o CA was selected for simulation during combustion period. Various models used in

253 CFD simulation study in the hydrogen based dual fuel engine are given in Table 5. It may be noted

254 that as the study is mainly focused on the experimental investigations, a few details of CFD

255 simulation are presented in this section in support of combustion advancement/autoignition at high

256 amounts of hydrogen energy share in the engine.

257

258 3. Methodology

259 3.1. Determination of autoignition of hydrogen-air mixture

260 In hydrogen based dual-fuel engines, typically diesel is used to initiate the ignition whereas

261 hydrogen is used as main fuel (main energy source). As hydrogen gas is injected into the intake

262 manifold of the engine along with intake air during suction stroke, the gaseous fuel could well mix

263 with the air, and forms high degree of homogeneous premixed hydrogen-air charge. When diesel

264 fuel is directly injected (sprayed) into the combustion chamber at the end of compression process,

265 entrainment of the premixed hydrogen-air charge into the diesel spray takes place and forms diesel-

266 hydrogen-air mixture. In this diesel-hydrogen-air mixture, the amount of diesel fuel is dominant

267 than the amount of hydrogen and hence the mixture is simply considered as diesel-air mixture.

268 Combustion initiates due to autoignition of diesel-air mixture as the in-cylinder temperature at the

269 end of compression (about 700 K) is higher than the autoignition temperature of diesel fuel (about

270 530 K). However, if the amount of hydrogen exceeds beyond a certain limit, autoignition of

271 premixed hydrogen-air charge could takes place without diesel aid, which may result in

272 combustion with knock. Autoignition of hydrogen-air charge could be defined as the initiation of

273 combustion of the premixed hydrogen-air charge without diesel ignition source (i.e., combustion

12
 ACCEPTED MANUSCRIPT

274 starts prior to the diesel injection). With low amount of hydrogen addition in to a CI engine

275 (Operation 1), generally combustion starts (SOC 1) after pilot diesel fuel injection (SOI 1),

276 whereas with high amount of hydrogen addition, combustion (SOC 2) may start earlier than the

277 diesel injection (SOI 2) as shown in Figure 5. Start of diesel fuel injection and start of combustion

278 of fuel-air charge (diesel-air (Operation 1) or hydrogen-air (Operation 2)) are together represented

279 in Figure 5. Later case (Operation 2) indicates autoignition of premixed hydrogen-air charge which

280 subsequently may create a problem of severe knock, whereas Operation 1 indicates a typical

281 combustion process in a dual-fuel engine without knock. Polk et al. expressed the similar tendency

282 of autoignition of propane-air mixture in a 1.9 liter turbo charged CI engine under dual fuel mode

283 [35].

284

285 3.2. Activation energy of hydrogen-air charge under dual fuel mode

286 The minimum ignition energy required for initiating chemical reaction is known as activation

287 energy. As the minimum ignition energy of hydrogen (0.02 mJ) is lower than the conventional

288 diesel fuel (0.6 mJ), the hydrogen promotes combustion reactions in a faster way. It may be noted

289 that pre-ignition reaction rate increases significantly with increasing hydrogen energy share in the

290 engine under dual fuel mode. The activation energy for hydrogen-air charge could be determined

291 using Eqs. (4-7) which were obtained by plotting log ID (Ignition delay) as a function of 1/T (T:

292 In-cylinder temperature) (Figure 6). Ignition delay of any fuel in a CI engine could be defined as

293 the time interval (or crank angle duration) between the start of injection of fuel and start of

294 combustion. The start of injection of fuel is usually taken as the time when the injector needle lifts

295 of its seat. The start of combustion could be defined as the change in slope of the heat release rate

296 from negative to positive. In the present study the start of injection was determined using fuel

13
 ACCEPTED MANUSCRIPT

297 injection pressure measurement i.e., the degree crank angle at which the injector nozzle starts open

298 at 250 bar pressure (nozzle opening pressure : 250 bar). The start of combustion was determined

299 using heat release rate, determined from in-cylinder pressure data with respect to degree crank

300 angle. Similarly in-cylinder temperature with respect to degree crank angle was determined using

301 the ideal gas equation with the input data of measured in-cylinder pressure [16]. The activation

302 energy of the hydrogen for initiating ignition under dual fuel mode is found to be 74992 J/mol at

303 950 K. The data is almost equal to the literature data reported by Chen, et al. that the activation

304 energy for the reaction of hydrogen-air charge at 900 K is about 79000 J/mol [43].

305

1
306 For τi = f (T) (3)

307 Where

308 τi = Ignition delay, ms

309 T = In-cylinder temperature at the time of pilot fuel injection, K

310 From Figure 6 the ignition delay could be expressed as Eq. (4),

311 ln(τi) = 9020 (1/T) - 0.7899 (4)

312 The above equation could be rewritten as Eq. (5).

9020
313 τi = 2.2032 exp ( T ) (5)

314 The general equation for determining ignition delay that could be expressed as Eq. (6).
9020
315 τi = μ exp ( T ) (6)

316 Where µ = Constant

317 T = In-cylinder temperature

318 From Eq. (5) and Eq. (6), the activation energy of hydrogen-air mixture could be obtained as

319 follows,

14
 ACCEPTED MANUSCRIPT

320 For Ea/Ru = 9020 and Ru = 8.314 J/mol-K,

321 The activation energy (Ea) = 74992 J/mol

322

323 It may be noted that the products (air-fuel charge/burned products) in the combustion chamber are

324 assumed as ideal gas which is used for calculating in-cylinder temperature with respect to degree

325 crank angle (Eqs. (7-9)).

326 p v = mRT (7)

327 T = pv/mR (8)

d d 𝑝𝑣
328 dθ(T) = dθ(mR) (9)

329 Assuming no variation in mass of air fuel charge (m) and characteristic gas constant (R) with

330 respect to degree crank angle the Eq.( 9) could be written as given in Eq.( 10).
dv dp
dT 𝑝dθ + v dθ
331 dθ = mair ‒ fuel charge R (10)

332

333 3.3 Knock limited factor for maximum hydrogen energy share

334 Abnormal rate of pressure rise (RPR) during combustion was considered as knocking in the

335 present study. Heywood stated the maximum RPR for knocking combustion is about 10 bar/

336 oCA in case of CI engines [39]. It is well known that the RPR is strong dependence on oxidation

337 reaction rate which is described by Arrhenius Eq. (11) [39].

338
Ea
(R T)
339 RPR (Maximum H2 energy share limit) ∝ exp u
(11)

340

341 Where

15
 ACCEPTED MANUSCRIPT

342 Ea = Activation energy of species

343 Ru = Universal gas constant

344 T = Reaction temperature (or) In-cylinder temperature

345 It may be noted that compression ratio and equivalence ratio are the other major influencing

346 design and operating parameters for knocking problem in the engine. The onset of knock in the

347 engine is inversely proportional to these parameters (Eqs. (12, 13)) [44]. Reduction in

348 compression ratio of the engine leads to decrease in the in-cylinder temperature which allows

349 higher energy share in the engine without knock. Typically low temperature combustion

350 strategies such as water injection, retarded diesel fuel injection, and compression ratio reduction

351 could increase the amount of hydrogen substitution in a CI engine [8].

352

1
353 Maximum H2 energy share limit ∝ compression ratio (CR) (12)

354

1
355 Maximum H2 energy share limit ∝ Equivalence ratio(∅) (13)

356

357 It could be observed from the literature that increase in hydrogen energy share in a CI engine

358 leads to autoignition of the hydrogen-air charge and subsequently to onset of knock [10, 34] i.e.,

359 knocking is directly proportional to hydrogen energy share in the engine (Eq. (14)).

360

361 Knock in the engine ∝ H2 energy share (14)

362

363 Knock limited factor (KLF) could be defined as a limiting value at which maximum hydrogen

364 could be substituted in a CI engine without knock. With the above inferences, the KLF for

16
 ACCEPTED MANUSCRIPT

365 maximum hydrogen energy share could be expressed as given in Eq. (15). The KLF is a

366 dimensionless indicator that could be used for finding maximum achievable hydrogen energy

367 share in a CI engine under dual fuel mode.

368
Ea
(R T)
369 Knock limited factor (KLF) = k (H2energy share)α(∅) ‒ β(CR) ‒ γexp u
(15)

370 Where

371 Constant, k = 1.2 (depends on engine configuration and loading conditions)

372 Hydrogen energy share exponent, α = 3.5

373 Equivalence ratio exponent, β = 0.351

374 CR exponent, γ = 0.3

375

376 4. Results and discussion

377 The concept of autoignition of premixed hydrogen-air charge in the hydrogen dual fuel engine

378 was introduced in the earlier study [1], but its detailed analysis was not provided. In the present

379 study, a detailed analysis with wide range of experimental tests including water injection in the

380 test engine is provided. Effects of high amount of hydrogen addition in the engine on combustion

381 advancement and subsequently on autoignition of hydrogen-air charge are discussed in this

382 section. The main reasons for autoignition of premixed hydrogen-air charge in the dual fuel

383 engine are also addressed.

384

385 4.1. Effect of high amount of hydrogen addition on combustion advancement in the dual

386 fuel engine

387 4.1.1. Experimental test results of combustion advancement

17
 ACCEPTED MANUSCRIPT

388 Figure 7 shows variation of liquid fuel injection pressure and heat release rate with respect to

389 degree crank angle of the engine under dual fuel mode at 100% load (Case I). It could be

390 observed from the figure that start of combustion (SOC) advanced with increase in hydrogen

391 energy share. This advancement in combustion occurred with larger degrees crank angle at high

392 amount of hydrogen substitution as compared to low amount of hydrogen substitution. For

393 example, the SOC was advanced about 1.8o CA duration with 14.5% hydrogen energy share

394 whereas with 18.8% hydrogen energy share the SOC was advanced about 6o CA duration as

395 compared to base diesel mode. The SOC was advanced from 1.9o CA after TDC with 0%

396 hydrogen energy share (diesel mode) to 4.1o CA before TDC with 18.8% hydrogen energy share

397 in the engine under dual fuel mode. The main reason for the combustion advancement could be

398 due to increase in the in-cylinder temperature with hydrogen addition.

399

400 4.1.2. CFD simulation results of combustion advancement

401 It may be noted that the fact of advancement in combustion with increasing hydrogen energy

402 share in the dual fuel engine is also supported by CFD simulation results. Progress of combustion

403 in the engine cylinder is illustrated with in-cylinder temperature contours as shown in Figure 9.

404 The degree crank angle at which a spontaneous increase in the in-cylinder temperature occurs

405 could be indicated as start of combustion in the engine. It is evident from the CFD simulation

406 results (Figure 9) that the combustion in the cylinder advanced with increase in hydrogen energy

407 share. For example, the in-cylinder temperature was spontaneously increased at 2.6o CA after

408 TDC with base diesel mode (0% hydrogen energy share), whereas it advanced to 0.9o and 0.6o

409 CA after TDC with 14.5% and 16.7% hydrogen energy shares (Figure 9). Similarly from the

410 experimental test results, it could be observed that the start of combustion was advanced from

18
 ACCEPTED MANUSCRIPT

411 1.9o ATDC with base diesel mode (0% hydrogen energy share) to 0.3o ATDC (after TDC) and

412 0.6o BTDC (before TDC) and 4.1o BTDC with 14.5%, 16.7% and 18.8% hydrogen energy shares

413 respectively as shown in Figure 7. Simulated in-cylinder pressure and in-cylinder temperature

414 results obtained from the CFD analysis on Mesh type 3 were validated with the experimental

415 results for the same engine operating conditions. Figure 8 shows validation of CFD simulation

416 results of in-cylinder pressure and in-cylinder temperature with the experimental results for base

417 diesel and dual-fuel modes at 100% load. In-cylinder peak pressure was slightly over predicted

418 (particularly in premixed combustion phase) and subsequently this error was amplified in case of

419 in-cylinder temperature. It could also be clearly seen that start of combustion (SOCcfd) was

420 advanced in CFD simulation as compared to experimental results (SOCexp) [45]. For example, at

421 16.7% hydrogen energy share SOCcfd occurred at 2o CA after TDC whereas SOCexp occurred at

422 4o CA after TDC. This over prediction of in-cylinder pressure and in-cylinder temperature could

423 be due to neglecting crevice volumes, blow-by losses, mismatching of compression ratio and

424 uncertainty associated with equivalence ratio. Similar type of trends was reported by Maghbouli

425 et al. in their CFD simulation work in a dual-fuel engine [46]. Liu and Karim also expressed the

426 similar results of higher in-cylinder peak pressure value (about 78 bar) with CFD simulation than

427 the experimental in-cylinder pressure about 74 bar in a dual-fuel engine [47]. The variations in

428 the in-cylinder pressure and in-cylinder temperature are affected by the total kinetic energy

429 variations and a slight pressure difference existing in the main combustion chambers [47].

430

431 Localized maximum in-cylinder temperature increased to about 2100 oC spontaneously (in all

432 cases) at the start of combustion as seen in the figure. Average temperature scale was represented

433 at high degree crank angles (i.e., at 3o, 5o, and 10o etc.). It may be noted that the phenomenon of

19
 ACCEPTED MANUSCRIPT

434 autoignition couldn’t be represented in a single domain using CFD simulation results

435 (temperature/ pressure/ species concentration) due to the software constraints (i.e., simultaneous

436 representation of both start of diesel fuel injection and start of combustion in a single domain

437 can’t be done). However, the in-cylinder temperature contours could be used as indirect

438 representation of autoignition concept in the engine. When ignition of the fuel-air charge occurs

439 inside the combustion chamber, in-cylinder temperature increases spontaneously. Santoso et al.

440 also used in-cylinder temperature contours (obtained from CFD simulation) for representing

441 combustion progress in a hydrogen based dual fuel engine [48].

442

443 It could also be observed from Figure 9 that the temperature contours are widespread towards

444 cylinder walls (wide propagation of flames towards end charge) in the combustion chamber with

445 hydrogen addition in the engine. This could be interpreted as due to rapid combustion of

446 hydrogen-air mixture in the engine under dual fuel mode. For a particular crank angle rotation

447 (for example at 3o CA), it could be observed qualitatively from Figure 9 that flame propagation

448 distance increased significantly towards the cylinder walls with increase in hydrogen energy

449 share. However, a detailed CFD study needs to be carried out in order to assess the

450 characteristics of flames (i.e., multiple ignition centers, flame propagation, and flame quenching

451 etc.) in dual fuel engines.

452

453 4.2. Effect of high amount of hydrogen addition on autoignition of hydrogen-air mixture

454 It is clearly seen in Figure 7 that until 16.7% hydrogen energy share, SOC occurred after the

455 injection of diesel fuel (ignition source), which indicates the autoignition of diesel-air charge

456 (diesel dominant diesel-hydrogen-air mixture) occurred first and later hydrogen-air mixture

20
 ACCEPTED MANUSCRIPT

457 (spread around the diesel spray) combusted in the cylinder. However, with 18.8% (about 19%)

458 hydrogen energy share, combustion started earlier (4.1o CA before TDC) than diesel fuel injection

459 (1.8o CA before TDC) which indicates autoignition of hydrogen-air charge without any external

460 ignition source (pilot diesel fuel). Further enhancement of hydrogen energy share leads to onset of

461 knock resulting in sharp increase in the in-cylinder peak pressure to about 90 bar (maximum

462 allowable peak pressure). Due to the autoignition of hydrogen-air mixture, the amount of hydrogen

463 substitution is restricted up to about 19% in the engine under conventional dual fuel mode at 100%

464 load. However, the maximum hydrogen energy share in the engine increased drastically from 19%

465 with duel fuel mode at 100% load (Case I) to 48.3% and 60.7% with dual fuel mode at 50% load

466 (Case II) and water added dual fuel mode (Case III). At 19% hydrogen energy share, there is no

467 such problem of autoignition in Case II and Case III operations. It could be observed from Figure

468 10 that at 50% load (Case II), there was no autoignition of hydrogen-air mixture for all hydrogen

469 energy shares ranging from 0% to 48.3%. This indicates the engine runs smoothly for all hydrogen

470 energy shares as the combustion (SOC) proceeded after the pilot diesel fuel injection as in typical

471 dual fuel operation. The equivalence ratio in the engine under dual-fuel mode decreased with

472 increase in hydrogen energy share at all loads. At 100% load, the equivalence ratio decreased from

473 0.4 with base diesel mode to 0.18 with diesel hydrogen dual-fuel mode (18.8% hydrogen energy

474 share). It may be noted that the equivalence ratio at higher and lower flammability limits for

475 hydrogen-air mixture are about 1.5 and 0.1. As the equivalence ratio of premixed hydrogen-air

476 mixture is under the flammability limits, there are more chances of autoignition of the charge

477 without diesel ignition source. It could be observed from Figure 11 (Case III) that the occurrence

478 of autoignition of hydrogen-air charge was suppressed until 56.5% hydrogen energy share, but

479 beyond this energy share autoignition of the premixed hydrogen-air charge problem was occurred.

21
 ACCEPTED MANUSCRIPT

480 The primary reason is due to substantial reduction in the in-cylinder temperature by the addition

481 of water which has high specific heat and also it dilutes the charge. The added water does function

482 of slowing down the reaction rate of hydrogen-diesel-air during combustion by reducing the in-

483 cylinder temperature and enhancing the charge dilution. With water addition in the engine (Case

484 III), knock limited hydrogen energy share (without autoignition problem) was increased from

485 18.8% to 56.5% as shown in Figure 11. Until 56.5% hydrogen energy share, start of combustion

486 occurred after the pilot fuel injection (diesel), which indicates the normal combustion process.

487 However, beyond 56.5% hydrogen energy share, start of combustion (5o CA before TDC) was

488 occurred prior to the fuel injection (1o CA before TDC) due to autoignition of hydrogen-air charge.

489

490 Similar findings were reported by Miyamoto et al. that the hydrogen was autoignited without the

491 ignition aid of pilot diesel fuel in a diesel-hydrogen dual fuel engine [34]. They found that with

492 high amount of hydrogen substitution, heat energy released prior to start of diesel injection, and

493 the heat energy release continued even after the diesel fuel cut off [34]. Experimental

494 investigations carried out by Polk et al. supported the fact of spontaneous autoignition of propane

495 in a diesel-propane based dual fuel engine due to its high reactivity [35]. They stated the reasons

496 for propane autoignition are due to high in-cylinder temperature and high equivalence ratio that

497 are conducive to rapid preignition. In case of hydrogen dual fuel mode, the reasons for

498 autoignition of hydrogen-air charge could be due to high in-cylinder temperature [49], rapid

499 preignition reactions [36], production of high concentration of free radicals (O, H, and OH) [50],

500 reduction in flame nucleation period [50]. In the present study, some of these reasons such as the

501 increase in the in-cylinder temperature and free radical concentration for autoignition of

502 hydrogen-air charge in the engine are discussed as given below;

22
 ACCEPTED MANUSCRIPT

503

504 4.2.1. Effect of in-cylinder temperature on autoignition of hydrogen-air charge

505 It may be noted that at 19% hydrogen energy share, autoignition of hydrogen-air charge occurred

506 in Case I operation (conventional dual fuel mode at 100% load), whereas no such problem

507 occurred in Case II (conventional dual fuel mode at 100% load) and Case III (water added dual

508 fuel mode at 100% load) operations. The main reason could be due to lower in-cylinder

509 temperature with water injection and lower load (50%) operation than high load (100%)

510 operation. For example, at 19% hydrogen energy share, the in-cylinder peak temperature

511 decreased from 1876 K with conventional diesel-hydrogen dual fuel mode at 100% load to 1680

512 K with water injected dual-fuel mode at 100% load and 1564 K with conventional diesel-

513 hydrogen dual fuel mode at 50% load (Figure 12). With this reduction in the in-cylinder

514 temperature, the possibilities of autoignition and knocking were reduced, that could lead to

515 significant enhancement of hydrogen energy share under dual fuel mode.

516

517 Autoignition-temperature of a fuel could be defined as the temperature at which the fuel will

518 spontaneously ignite [39]. Leishear measured the relationship between pressure and autoignition

519 temperature of hydrogen-air mixture at stoichiometric condition in a piping system used in

520 nuclear reactor [51]. Typically the stoichiometric hydrogen-air charge gets autoignition at

521 temperature of 580 oC and atmospheric pressure of 1 bar [51]. In case of internal combustion

522 engines, the autoignition-temperature of fuel-air charge depends on various parameters including

523 pressure and equivalence ratio. For example, at an equivalence ratio of 1 (stoichiometric

524 condition), autoignition temperature of hydrogen-air charge decreased from 580 oC to 440 oC due

525 to increase in pressure from 1 bar to 40 bar [51]. Similarly, in the present study, autoignition-

23
 ACCEPTED MANUSCRIPT

526 temperature of hydrogen-air charge in the engine is found about 670 oC at an equivalence ratio of

527 0.4 and the in-cylinder pressure about 54 bar (under conventional hydrogen-diesel dual fuel

528 mode). The in-cylinder temperature at the end of compression stroke is reached to the range of

529 600 oC to 700 oC. This temperature is enough for initiation of autoignition of hydrogen-air

530 charge in the engine cylinder. These experimental results are in agreement with the literature

531 data of autoignition temperature of the charge is about 650 oC to 700 oC at equivalence ratios of

532 0.4 to 0.6 [30, 44]. Daeyup and Hochgreb reported autoignition of premixed hydrogen-air

533 mixture could take place in reactant’s pressure range from 4 bar to 40 bar and temperature

534 range from 950 K -1050 K [52]. However, the relationship between autoignition temperature and

535 in-cylinder pressure with respect to equivalence ratio for dual fuel engines needs to be studied.

536

537 A summary of in-cylinder pressure and autoignition temperatures with respect to different

538 hydrogen energy shares under different experimental test cases is given in Table 6. Experimental

539 tests were conducted on the same test engine with retarded diesel injection timing under dual fuel

540 mode in the earlier study [8]. A conclusion emerged from these results is that the premixed

541 hydrogen-air charge gets autoignition at the in-cylinder temperature is about 953 K ± 8 K with

542 the corresponding in-cylinder pressure of 56 bar ± 3 bar in the engine. Wong and Karim studied

543 the effect of in-cylinder temperature variation on autoignition of three gaseous fuels such as

544 methane (CH4), propane (C3H8), and hydrogen (H2) in a diesel engine [36]. The fuel-air charge is

545 able to reach a higher peak temperature with hydrogen due to its higher polytropic index as

546 compared to other two gaseous fuels. They also found that the hydrogen-air charge did not

547 autoignite at low in-cylinder temperature whereas the propane and methane fuels were

548 autoignited even at low temperatures [36]. Hence, the problem of autoignition at high amount of

24
 ACCEPTED MANUSCRIPT

549 hydrogen substitution could be resolved with low temperature combustion strategies such as

550 retarded injection timing of diesel fuel, addition of diluents (nitrogen and carbon dioxide), water

551 injection, and compression ratio reduction.

552

553 4.2.2. Effect of free radicals concentration on autoignition of hydrogen-air charge

554 It is reported that the increase in concentration of hydrogen increases the potential for free

555 radicals (H, O, and OH) production during its oxidation process [50]. High amount of hydrogen

556 substitution in the dual fuel engine may produce a larger pool of H, O, and OH radicals at an ear-

557 lier stage of the combustion process. High concentration of these radicals have significant effects

558 on autoignition of hydrogen-air mixture and overall combustion reaction rate during the engine

559 operation. These effects include extension of the lean limit, increasing EGR/dilution tolerance,

560 and shortening of the flame nucleation period, thereby increasing the heat release rate [50].

561 Increase in the in-cylinder temperature could be one of the major cause for production of high

562 levels of free radicals in the dual fuel engine.

563

564 4.2.3. Effect of residual gases on autoignition of hydrogen-air charge

565 The other reasons for autoignition of hydrogen-air charge in the dual fuel engine could be the

566 gaseous fuel concentration in residual gases of the previous cycle and high temperature of the

567 residual gases [36, 44]. These existing species in the residual gases can play important chemical

568 and thermal roles in the preignition reaction processes of the next cycle. The presence of residual

569 gas could alter the in-cylinder temperature of the hydrogen-air charge at the beginning of

570 compression process. The kinetic effect of these residual gases may cause an increase in the

571 preignition reaction activity of the gaseous fuel-air charge which leads to autoignition of the

25
 ACCEPTED MANUSCRIPT

572 charge [36]. At higher load and higher concentration of the hydrogen, flame which is initiated

573 from the various ignition centers of the diesel fuel, propagates at a faster rate and consumes the

574 most of the cylinder gaseous fuel-air mixture in the preceding cycles which results in higher

575 residual gas temperature in the following cycles [50]. The residual gas with some partial

576 oxidation products from the preceding cycles would be a source of active radicals for the

577 following cycle. It is evident from research findings of Wong and Karim that the residual gas

578 (EGR) has high potential to enhance the preignition reaction rate in dual fuel engines [36].

579

580 4.3. Knock limited factor for maximum hydrogen energy share

581 From the wide range of experimental tests carried out with different engine operating conditions

582 (Case I to Case III), it is found that the knock limited factor (KLF) for maximum hydrogen

583 energy share under dual fuel mode is a function of hydrogen energy share, in-cylinder pressure,

584 equivalence ratio, compression ratio, in-cylinder temperature, and activation energy of hydrogen-

585 air mixture. The following critical conditions for KLF were observed from the experimental test

586 results under hydrogen dual fuel mode.

587

588 If KLF < 1 ----> Less probability for knocking

589 KLF > 1 ----> More probability for knocking

590 For Case (I) operation, the KLF was increased to 0.9 with 16.7% hydrogen energy share and then

591 it reached to 1.1 with 18.8% hydrogen energy share as shown in Figure 13. Similarly with Case

592 (II) and Case (III) operations, the KLF was under limit until the hydrogen energy shares of 18%

593 and 56.5% and beyond these energy shares the KLF was crossed the limit exponentially as

594 shown in the figure.

26
 ACCEPTED MANUSCRIPT

595

596 5. Conclusions

597 The effect of high percentage of hydrogen energy share on autoignition of premixed hydrogen-

598 air charge was analyzed in a single cylinder hydrogen based dual fuel engine. The following

599 conclusions are drawn based on the experimental results.

600  High in-cylinder temperature which is the predominant factor, influences the autoignition

601 of premixed hydrogen-air charge. The premixed hydrogen-air charge could get self-

602 ignition (autoignition) without external ignition aid (diesel pilot fuel) when the

603 temperature of the reactants is about 953 K ± 8 K with the corresponding in-cylinder

604 pressure of 56 bar ± 3 bar. Similarly, the auto-ignition temperature and pressure of any

605 compression ignition engine under dual fuel mode (Diesel-Hydrogen) could be found out

606 using the methodology emerged from this study. Hence, the critical energy share of

607 hydrogen in a dual fuel engine working under specific design and operating parameters

608 could be calculated within reasonable accuracy using this auto-ignition temperature and

609 pressure predicted using this study’s methodology.

610  Increase in hydrogen energy share in the engine enhances the degree of advancement in

611 start of combustion and the reactant’s in-cylinder temperature. Too advance in start of

612 combustion with high amount of hydrogen energy share (beyond a critical limit) leads to

613 more probability of auto-ignition which may result in severe knocking problem.

614  Knock limited factor (KLF) is a notable outcome emerged from this study as the

615 dimensionless indicator could be useful to predict qualitatively the maximum hydrogen

616 energy share in the engine under dual fuel mode. If the KLF is greater than 1, the

617 probability of knock would be higher whereas less than one means lower probability.

27
 ACCEPTED MANUSCRIPT

618  The probability of auto-ignition of hydrogen-air charge can be minimized by reducing

619 reactant’s temperature during ignition period using suitable techniques including water

620 injection. For example, it is confirmed from the experimental results that knock limited

621 maximum hydrogen energy share in a dual fuel diesel engine (7.4 kW at 100% load) was

622 found as 18.8% under conventional dual fuel mode and this critical energy share could be

623 enhanced to 60.7% with water injection in the dual fuel engine for the same engine

624 operating conditions.

625

626 Abbreviations

627 Ea : Activation energy of hydrogen, J/mol

628 R : Characteristic gas constant, J/mol

629 SWC : Specific water consumption, g/kWh

630 T : Absolute temperature/In-cylinder temperature, K

631 p : In-cylinder pressure, N/m2

632 V : Instantaneous cylinder volume, m3

633

634

635 References

636 1. Chintala V, Subramanian KA. Experimental investigations on effect of different

637 compression ratios on enhancement of maximum hydrogen energy share in a compression

638 ignition engine under dual-fuel mode. Energy. 2015;87:448-62.

28
 ACCEPTED MANUSCRIPT

639 2. Adnan R, Masjuki HH, Mahlia TMI. Performance and emission analysis of hydrogen

640 fueled compression ignition engine with variable water injection timing. Energy.

641 2012;43(1):416-26.

642 3. National Hydrogen Energy Road Map - 2006 (Abridged Version, 2007), National

643 Hydrogen Energy Board, Ministry of New and Renewable Energy Government of India.

644 2007.

645 4. Shafiei E, Davidsdottir B, Leaver J, Stefansson H, Asgeirsson EI. Comparative analysis of

646 hydrogen, biofuels and electricity transitional pathways to sustainable transport in a

647 renewable-based energy system. Energy. 2015;83:614-27.

648 5. Karim GA. Hydrogen as a spark ignition engine fuel. International Journal of Hydrogen

649 Energy. 2003;28(5):569-77.

650 6. Yang Z, Wang L, Zhang Q, Meng Y, Pei P. Research on optimum method to eliminate

651 backfire of hydrogen internal combustion engines based on combining postponing ignition

652 timing with water injection of intake manifold. International Journal of Hydrogen Energy.

653 2012;37(17):12868-78.

654 7. Chintala V, Subramanian KA. Assessment of maximum available work of a hydrogen

655 fueled compression ignition engine using exergy analysis. Energy. 2014;67:162-75.

656 8. Chintala V, Subramanian KA. An effort to enhance hydrogen energy share in a

657 compression ignition engine under dual-fuel mode using low temperature combustion

658 strategies. Applied Energy. 2015;146:174-83.

659 9. Subramanian KA, Chintala V. Reduction of GHGs emissions in a biodiesel fueled diesel

660 engine using hydrogen ASME 2013 Internal Combustion Engine Fall Technical

29
 ACCEPTED MANUSCRIPT

661 Conference; October 13-16; Dearborn, Michigan: Paper No. ICEF2013-19133,

662 doi:10.1115/ICEF2013-19133; 2013.

663 10. Chintala V, Subramanian KA. Hydrogen energy share improvement along with NOx

664 (oxides of nitrogen) emission reduction in a hydrogen dual-fuel compression ignition

665 engine using water injection. Energy Conversion and Management. 2014;83:249-59.

666 11. Yadav VS, Soni SL, Sharma D. Engine performance of optimized hydrogen-fueled direct

667 injection engine. Energy. 2014;65:116-22.

668 12. Deb M, Paul A, Debroy D, Sastry GRK, Panua RS, Bose PK. An experimental

669 investigation of performance-emission trade off characteristics of a CI engine using

670 hydrogen as dual fuel. Energy. 2015;85:569-85.

671 13. Wu H-W, Wu Z-Y. Using Taguchi method on combustion performance of a diesel engine

672 with diesel/biodiesel blend and port-inducting H2. Applied Energy. 2013;104:362-70.

673 14. Edwin Geo V, Nagarajan G, Nagalingam B. Studies on dual fuel operation of rubber seed

674 oil and its bio-diesel with hydrogen as the inducted fuel. International Journal of Hydrogen

675 Energy. 2008;33(21):6357-67.

676 15. Szwaja S, Grab-Rogalinski K. Hydrogen combustion in a compression ignition diesel

677 engine. International Journal of Hydrogen Energy. 2009;34(10):4413-21.

678 16. Chintala V, Subramanian KA. Experimental investigation on effect of enhanced premixed

679 charge on combustion characteristics of a direct injection diesel engine. International

680 Journal of Advances in Engineering Sciences and Applied Mathematics. 2014;6(1-2):3-16.

681 17. Varde K, Frame G. Hydrogen aspiration in a direct injection type diesel engine-its effects

682 on smoke and other engine performance parameters. International Journal of Hydrogen

683 Energy. 1983;8(7):549-55.

30
 ACCEPTED MANUSCRIPT

684 18. Saravanan N, Nagarajan G. An experimental investigation on manifold-injected hydrogen

685 as a dual fuel for diesel engine system with different injection duration. International

686 Journal of Energy Research. 2009;33(15):1352-66.

687 19. Mathur HB, Das LM, Patro TN. Hydrogen fuel utilization in CI engine powered end utility

688 systems. International Journal of Hydrogen Energy. 1992;17(5):369-74.

689 20. de Morais AM, Mendes Justino MA, Valente OS, Hanriot SdM, Sodré JR. Hydrogen

690 impacts on performance and CO2 emissions from a diesel power generator. International

691 Journal of Hydrogen Energy. 2013;38(16):6857-64.

692 21. Nguyen TA, Mikami M. Effect of hydrogen addition to intake air on combustion noise

693 from a diesel engine. International Journal of Hydrogen Energy. 2013;38(10):4153-62.

694 22. Bose PK, Deb M, Banerjee R, Majumder A. Multi objective optimization of performance

695 parameters of a single cylinder diesel engine running with hydrogen using a Taguchi-fuzzy

696 based approach. Energy. 2013;63:375-86.

697 23. Christodoulou F, Megaritis A. Experimental investigation of the effects of separate

698 hydrogen and nitrogen addition on the emissions and combustion of a diesel engine.

699 International Journal of Hydrogen Energy. 2013;38(24):10126-40.

700 24. Saravanan N, Nagarajan G. Performance and emission studies on port injection of

701 hydrogen with varied flow rates with Diesel as an ignition source. Applied Energy.

702 2010;87(7):2218-29.

703 25. Shin B, Cho Y, Han D, Song S, Chun KM. Hydrogen effects on NOx emissions and brake

704 thermal efficiency in a diesel engine under low-temperature and heavy-EGR conditions.

705 International Journal of Hydrogen Energy. 2011;36(10):6281-91.

31
 ACCEPTED MANUSCRIPT

706 26. Szwaja S, Naber JD. Dual nature of hydrogen combustion knock. International Journal of

707 Hydrogen Energy. 2013;38(28):12489-96.

708 27. Papagiannakis RG, Kotsiopoulos PN, Zannis TC, Yfantis EA, Hountalas DT, Rakopoulos

709 CD. Theoretical study of the effects of engine parameters on performance and emissions of

710 a pilot ignited natural gas diesel engine. Energy. 2010;35(2):1129-38.

711 28. Abdelaal MM, Rabee BA, Hegab AH. Effect of adding oxygen to the intake air on a dual-

712 fuel engine performance, emissions, and knock tendency. Energy. 2013;61:612-20.

713 29. Chintala V, Subramanian KA. A CFD (computational fluid dynamics) study for

714 optimization of gas injector orientation for performance improvement of a dual-fuel diesel

715 engine. Energy. 2013;57:709-21.

716 30. Karim GA. Combustion in Gas Fueled Compression Ignition Engines of the Dual Fuel

717 Type. Journal of Engineering for Gas Turbines and Power. 2003;125(3):827.

718 31. Sivabalakrishnan R, Jegadheesan C. Study of Knocking Effect in Compression Ignition

719 Engine with Hydrogen as a Secondary Fuel. Chinese Journal of Engineering. 2014;2014:1-

720 8.

721 32. Szwaja S, Bhandary K, Naber J. Comparisons of hydrogen and gasoline combustion knock

722 in a spark ignition engine. International Journal of Hydrogen Energy. 2007;32(18):5076-87.

723 33. Naber J. Hydrogen combustion under diesel engine conditions. International Journal of

724 Hydrogen Energy. 1998;23(5):363-71.

725 34. Miyamoto T, Hasegawa H, Mikami M, Kojima N, Kabashima H, Urata Y. Effect of

726 hydrogen addition to intake gas on combustion and exhaust emission characteristics of a

727 diesel engine. International Journal of Hydrogen Energy. 2011;36(20):13138-49.

32
 ACCEPTED MANUSCRIPT

728 35. Polk AC, Gibson CM, Shoemaker NT, Srinivasan KK, Krishnan SR. Analysis of Ignition

729 Behavior in a Turbocharged Direct Injection Dual Fuel Engine Using Propane and Methane

730 as Primary Fuels. Journal of Energy Resources Technology. 2013;135(3):1-10.

731 36. Wong YK, Karim GA. A kinetic examination of the effects of the presence of some

732 gaseous fuels and preignition reaction products with hydrogen in engines. International

733 Journal of Hydrogen Energy. 1999;24(5):473-8.

734 37. SinghYadav V, Soni SL, Sharma D. Performance and emission studies of direct injection

735 C.I. engine in duel fuel mode (hydrogen-diesel) with EGR. International Journal of

736 Hydrogen Energy. 2012;37(4):3807-17.

737 38. Saravanan N, Nagarajan G. An experimental investigation of hydrogen-enriched air

738 induction in a diesel engine system. International Journal of Hydrogen Energy.

739 2008;33(6):1769-75.

740 39. Heywood JB. Internal Combustion Engines Fundamentals. New York: McGraw-Hill, Inc.;

741 1988.

742 40. Tsujimura T, Mikami S, Achiha N, Tokunaga Y, Senda J, Fujimoto H. A Study of Direct

743 Injection Diesel Engine Fueled with Hydrogen. SAE Paper No 2003-01-0761. 2003.

744 41. Lata DB, Misra A, Medhekar S. Investigations on the combustion parameters of a dual fuel

745 diesel engine with hydrogen and LPG as secondary fuels. International Journal of

746 Hydrogen Energy. 2011;36(21):13808-19.

747 42. High Speed Diesel Specifications, Indian Oil Corporation Ltd. (IOCL), India 2005.

748 43. Chen K, Karim GA, Watson HC. Experimental and Analytical Examination of the

749 Development of Inhomogeneities and Autoignition During Rapid Compression of

33
 ACCEPTED MANUSCRIPT

750 Hydrogen-Oxygen-Argon Mixtures. Journal of Engineering for Gas Turbines and Power.

751 2003;125(2):458.

752 44. Liu Z, Karim GA. Knock characteristics of dual-fuel engines fuelled with hydrogen fuel.

753 International Journal of Hydrogen Energy. 1995;20(11):919-24.

754 45. Chintala V, Subramanian KA. CFD analysis on effect of localized in-cylinder temperature

755 on nitric oxide (NO) emission in a compression ignition engine under hydrogen-diesel

756 dual-fuel mode. Energy. 2016;116, Part 1:470-88.

757 46. Maghbouli A, Saray RK, Shafee S, Ghafouri J. Numerical study of combustion and

758 emission characteristics of dual-fuel engines using 3D-CFD models coupled with chemical

759 kinetics. Fuel. 2013;106:98-105.

760 47. Liu C, Karim GA. Three-Dimensional Computational Fluid Simulation of Diesel and Dual

761 Fuel Engine Combustion. Journal of Engineering for Gas Turbines and Power.

762 2009;131(1):012804.

763 48. Santoso WB, Bakar RA, Nur A. Combustion Characteristics of Diesel-Hydrogen Dual Fuel

764 Engine at Low Load. Energy Procedia. 2013;32:3-10.

765 49. Karim GA, Liu Z. Simulation of combustion processes in gas-fuelled diesel engines.

766 Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and

767 Energy. 1997;211(2):159-69.

768 50. Sherif SA, Goswami DY, Stefanakos EK, Steinfeld A. Handbook of hydrogen energy.

769 London, New York: CRC Press, Taylor & Francis Group; 2014.

770 51. Leishear RA. A Hydrogen Ignition Mechanism for Explosions in Nuclear Facility Piping

771 Systems. Journal of Pressure Vessel Technology. 2013;135(5):1-5.

34
 ACCEPTED MANUSCRIPT

772 52. Daeyup L, Hochgreb S. Hydrogen autoignition at pressures above the second explosion

773 limit (0.6–4 MPa). Int J Chem Kinet. 1998;30:385–406.

774

775 Figures:

776

777 Figure 1 Photographic view of experimental setup for dual fuel mode

778

35
 ACCEPTED MANUSCRIPT

779

780 Figure 2 Water injection system for the engine under diesel-hydrogen dual-fuel mode

781

782 Figure 3 Schematic diagram of combustion analysis system

36
 ACCEPTED MANUSCRIPT

783

784

785 Figure 4 (a) Hexahedral mesh of piston and cylinder head for sector of 72 degree (b) piston bowl

786 geometry

787

37
 ACCEPTED MANUSCRIPT

788

789 Figure 5 Determination of autoignition of hydrogen-air mixture under dual fuel mode

790

791 Figure 6 Variation of ignition delay with respect to temperature with diesel-hydrogen dual fuel

792 mode at 100% load

38
 ACCEPTED MANUSCRIPT

793

794

795 Figure 7 Autoignition of hydrogen-air fuel charge for base dual fuel mode (case I) [1]

39
 ACCEPTED MANUSCRIPT

796

797 Figure 8 Validation of CFD results with experimental in-cylinder pressure and in-cylinder

798 temperature results [45]

40
 ACCEPTED MANUSCRIPT

799

800 Figure 9 Contours of in-cylinder temperature for different hydrogen energy shares in the engine

801 under dual fuel mode at 100% load

41
 ACCEPTED MANUSCRIPT

802

803 Figure 10 Diesel injection pressure and heat release rate curves under diesel-hydrogen dual fuel

804 mode at 50% load

805

806 Figure 11 Autoignition of hydrogen-air fuel charge with low temperature combustion strategies

42
 ACCEPTED MANUSCRIPT

807

808 Figure 12 In-cylinder temperature variation at 19% hydrogen energy share for different test cases

809

810 Figure 13 Knock limited factor variation with respect to hydrogen energy share at different test

811 conditions
43
 ACCEPTED MANUSCRIPT

812 Tables:

813 Table 1 Literature details of the maximum hydrogen energy share in hydrogen dual-fuel engines

Reference Engine details Amount of hydrogen substitution

Saravanan et al. [18] Nc=1, CR=16.5:1, 6.7% energy share

BMEP =5.4 bar

Edwin et al. [14] Nc=1, CR=17.5:1, 12.7% energy share

BMEP =5.3 bar

Mathur et al. [19] Nc=1, CR=17.5:1, 14.8% energy share (without power loss)

BMEP =4.9 bar

de Morais et al. [20] Nc=4, CR=17:1, 20% energy share

BMEP =6.5 bar

Nguyen and Mikami Nc=1, CR=16.7:1, 10% volume of intake air (or) 15% energy

[21] BMEP =7.3 bar share (approx.)

Bose et al. [22] Nc=1, CR=17.5:1, hydrogen flow rate of 0.15 kg/h (or)17.6%

BMEP =6.4 bar energy share (approx.)

Yadav et al. [11] Nc=1, CR=17.5:1, 16.4% energy share

BMEP =5.3 bar

Christodoulou and Nc=4, CR=18.2:1, 8% volume of intake air (or) 12.8% energy

Megaritis [23] BMEP =9.2 bar share (approx.)

Saravanan et al. [24] Nc=1, CR=16.5:1, 10% energy share

BMEP =5.4 bar

Shin et al. [25] Nc=4, CR=17.3:1, 10% energy share

BMEP =4.9 bar

44
 ACCEPTED MANUSCRIPT

Varde et al. [17] Nc=1, CR=17.4:1, 14% hydrogen energy share at 5.8 bar BMEP

BMEP : 5.8 bar and 17% share at 4.7 bar BMEP

Szwaja et al. [15] Nc=2, CR=17:1, 17% hydrogen energy share (knock limited

BMEP : 11.2 bar hydrogen share)

814

815

816 Table 2 Technical specifications of the engine

Parameter Description

Type of engine Compression ignition engine

Number of cylinders 1

Displacement volume, cc 947.4

Rated power output, kW 7.4

Rated speed, rpm 1500

Compression ratio 19.5:1

Intake valve opening and closing, degree CA 43 before TDC & 67 after BDC

Exhaust valve opening and closing, degree CA 87 before BDC & 39 after TDC

Liquid fuel injection timing by spill, degree CA 8 before TDC

Nozzle opening pressure, bar 250

817

818 Table 3 Properties of the fuels used in the experimental study [3, 42]

Fuel characteristics Diesel Hydrogen

Lower heating value, MJ/kg 44.05 120

Stoichiometric air fuel ratio 14.5 34.2

45
 ACCEPTED MANUSCRIPT

Auto ignition temperature, K 530 858

Laminar burning velocity, m/s 0.3 2.65 -3.25

Cetane number 51 --

Density, kg/m3 821.5 0.083

Viscosity, cSt 2.64 --

819

820

821 Table 4 Details of measurement range and resolution of instruments/sensors

Instrument/ Measuring Measuring Resolution Accuracy Uncertai

sensor Name parameter range nty (%)

Dyno- Engine 0-150 N-m 0.1 N-m 0.2% of Full scale 2.42

controller torque

Dyno- Engine 0-10000 rpm 1 rpm 0.1% of Full Scale 0.195

controller speed

Piezoelectric In-cylinder 0-250 bar Sensitivity: ± 0.3% to ± 0.6% of 0.846

pressure pressure 45 pC/bar value (Peak

sensor pressure)

Optical Degree 720 pulses 0.1 degree --- ---

encoder crank angle per CA

revolution

Air flow Intake air 0-330 m3/h 0.2 m3/h ± 2% of flow 1.7

meter volumetric

flow rate

46
 ACCEPTED MANUSCRIPT

Gas mass Gaseous 0-10 kg/h 0.01 kg/h ± 0.35% of flow 1.4

flow meter fuel flow

rate

Calculated Hydrogen --- --- --- 0.33

parameter energy

share

822

823

824 Table 5 Summary of models used in CFD simulation

Description Model used

Solver Pressure based type (Transient)

Dynamic mesh Layering (In-cylinder options)

Spatial discretization Second order upwind

Turbulence Standard k-Epsilon

Turbulence chemistry interaction (Combustion) Finite rate/Eddy dissipation

Diesel pilot fuel injection (spray model) Discrete phase mode (DPM); Solid cone

825

826

827 Table 6 Summary of autoignition temperatures for different experimental test cases [8]

Operating condition Fuels used H2 energy In-cylinder Autoignition

share pressure (bar) temperature (K)

Dual fuel mode (Case I) Diesel-H2 18.8 54.4 944.5

Water injection (Case II) Diesel-H2 60.7 59.1 959.4

47
 ACCEPTED MANUSCRIPT

Retarded diesel injection Diesel-H2 21.2 56.1 951.2

timing [8]

Retarded diesel injection Diesel-H2 24.5 57.2 954.6

timing [8]

828

829

830

48
 ACCEPTED MANUSCRIPT

Highlights

 Maximum H2 energy share in a CI engine at 100% load is limited due to knocking

 Autoignition of hydrogen-air charge leads to knocking during combustion

 Increase in in-cylinder temperature is main reason for autoignition of the charge

 Start of combustion advanced with H2 addition in the engine under dual fuel mode

 Maximum H2 energy share increased with reduction in in-cylinder temperature