1

2

45

6

7 Q1

89

10

1 2

1314151617

18192021222324

2 5

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

Q2Q1

Fuel xxx (2014) xxx–xxx

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

Contents lists available at ScienceDirect

Fuel

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

Short communication

Water solubility characteristics of current aviation jet fuels

http://dx.doi.org/10.1016/j.fuel.2014.04.0910016-2361/� 2014 Published by Elsevier Ltd.

⇑ Corresponding author. Tel.: +44 1179367028.E-mail address: joseph.lam@airbus.com (J.K.-W. Lam).

Please cite this article in press as: Lam J-KW et al. Water solubility characteristics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/1j.fuel.2014.04.091

Joseph K.-W. Lam a,⇑, Mark D. Carpenter b, Colleen A. Williams b, Janice I. Hetherington b

a Airbus Operations Limited, New Filton House, Filton, Bristol BS99 7AR, UKb Cranfield Defence and Security, Cranfield University, Shrivenham, Swindon SN6 8LA, UK

a r t i c l e i n f o

2627282930313233343536

Article history:Received 25 November 2013Received in revised form 25 April 2014Accepted 29 April 2014Available online xxxx

Keywords:Water solubilityAviation fuelDissolved waterAromaticsAircraft fuel systems

a b s t r a c t

Water solubility-temperature curves were created experimentally for five kerosine fuels and one wide-cut fuel. The Karl Fischer Coulometer method was used to determine the saturated water content of eachfuel over a range of temperatures between �5 �C and 25 �C. The affinity for water of each fuel withrespect to the other fuels was measured and ranked. The experimental solubility curves are comparableto the water solubility curves of the Jet A-1 and JP-4 fuels reported in the CRC Handbook of Aviation FuelProperties. There are subtle differences between the experimental solubility curves and the CRC curves.This may be attributed to batch-to-batch variation or recent changes in fuel chemistry owing to the needto exploit less favourable crude oil reserves. A generic equation has been proposed to estimate the watersolubility when aromatics content and flash point are known.

� 2014 Published by Elsevier Ltd.

37

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

1. Introduction

Jet fuel does not have an exact chemical composition; it is pro-duced to meet a set of properties defined by various specifications[1–5]. Its chemical composition will depend on the chemical spe-cies present in the parent crude oil that distil during the physicalseparation in the kerosine boiling range and on further chemicaltreatments that may be required to meet the jet fuel specifications.The lack of precise property data is due to the variation of jet fuels,based on the composition of the original crude oil and the refinerymethods used.

The three major constituent groups of jet fuel are paraffins (sat-urated alkanes), naphthenes (saturated cyclic alkanes) and aromat-ics [6–8]. The proportion of each chemical species varies frombatch to batch, but paraffins and naphthenes together accountfor a minimum of 70%v/v [8]. Aromatics are limited to a maximumof 25%v/v [1] by specifications and to an industry accepted mini-mum of 8%v/v [6] for jet fuels containing synthetic componentsconforming to the ASTM D7566-13 [9] specification. Olefins(unsaturated hydrocarbons or alkanes) and trace impuritiesaccount for less than 5%v/v [8] and are minor constituents of jetfuel. Baena-Zambrana et al. [10] provide an overview of the struc-ture of these classes of hydrocarbons. Further details may be foundin Hemighaus et al. [7].

Aromatics are unsaturated hydrocarbons and have characteris-tic benzene rings with delocalized p electron clouds. They are

90

91

92

mainly alkyl-benzenes and alkyl-naphthalenes [8]. It has beenrecognised that extended p electron systems such as thoseencountered in aromatics, form hydrogen bonds with water [11–13]. The formation of attractive electrostatic interactions betweenwater and extended p electron systems account for the enhancedsolubility of water in aromatics relative to both paraffins andnaphthenes.

Water contamination of jet fuel can result in fuel system icing[14], microbial contamination, corrosion and tank gauging prob-lems. Mathematical models have been developed for studyingand predicting the dispersal of water in aircraft fuel systems[15,16], and are used in design to optimise fuel–water manage-ment and ensure that water build up is minimised throughoutthe system. These models compute the water contamination, dueto the principal mechanisms of dissolution, suspension, condensa-tion and transportation. Among other parameters, it is importantto have accurate data on the water solubility behaviour in jet fuelto ensure confidence in the model results.

Some data on the water solubility behaviour in jet fuel are avail-able in the open literature but this mostly dates back to the 1950s.Even data given in the CRC Handbook of Aviation Fuel Properties [6,Figs. 2–32], were found to pre-date 1984; the actual source of thesedata could not be verified but look similar to data presented by[17]. To take account of recent changes in fuel chemistry owingto the need to exploit less favourable crude oil reserves [18] andthe introduction of non-petroleum-derived fuels, such as Fischer–Tropsch (FT), hydrotreated esters and fatty acids (HEFA) togetherwith other alternatives [19–24], data on the water uptake andsolubility in jet fuels needed to be updated. This study was used

0.1016/

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

Abbreviations

AVTAG aviation wi de-cut turbine fuelAVTUR aviation turbine fuelCOA Certificate of AnalysisCRC Coordinating Research Council, Inc.CTL coal to liquidDi-EGMME Di-Ethylene Glycol Mono-Methyl EtherFBP final boiling pointFSII fuel system icing inhibitorFT Fischer–TropschGTL gas to liquid

HEFA hydrotreated esters and fatty acidsIBP initial boiling pointKF Karl FischerMSEP MicroseparometerNATO North Atlantic Treaty OrganisationPPMCC Pearson product-moment correlation coefficientRH relative humiditySDA static dissipater additiveUV ultraviolet

2 J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

to assess the water solubility of current aviation jet fuels to fill thatknowledge gap.

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

2. Test fuels

The study evaluated the water solubility characteristics of fivejet fuels made to different national standards, which either com-plied with or were very similar to the Jet A-1 fuel specification.The fuels were freshly sourced since fuel compositions may changesubtly with storage due to the formation of polar oxidation com-pounds such as organic acids. One of the fuels was fully synthetic,i.e., not derived from conventional crude oil stocks. The work pro-gramme also included limited tests on a wide-cut fuel conformingto the JP-4 specification; this contains a proportion of lighterhydrocarbon molecules (naphtha or gasoline fraction). The lighterfraction improves the fuel relight ability at altitude and reducesits freezing point but also contributes significantly to the fire riskdue to a low flash point and higher vapour pressure. It is now onlyused in consistently cold climates and availability is limited.

The fuels used in this study are summarised in Table 1. Fuels Ato E may be collectively referred to as kerosine fuels; Fuel F may bereferred to as a wide-cut fuel.

A Certificate of Analysis (COA) was supplied with each fuel asevidence that the fuel conformed to the relevant specification.Chemical composition data and physical test data were given inthe COA. Some key characteristics of the test fuels are listed inTable 2. The fuels were supplied to the normal commercial specifi-cation and included static dissipater additive (SDA), where shown.Fuel F was actually supplied to the NATO F-40 (AVTAG/FSII) spec-ification which includes requirements for lubricity improver, SDAand fuel system icing inhibitor, FSII.

157

158

159

160

161

162

163

3. Experimental

Fuels were filtered using nitrogen gas pressure through aSartorius filter assembly fitted with Millipore HAWP04700(0.45 lm) mixed cellulose filter; fuels were filtered in batches of

164

165

166

167

168

169

170

171

Table 1Test fuels.

ID Fuel & Specification Provenance

A Sasol Fischer–Tropsch fullysynthetic Jet A-1

Sasol, South Africa

B Chinese No. 3 jet fuel PetroChina International (Hong Kong)Ltd., Hong Kong

C Coryton high aromatics Jet A-1 Coryton Advanced Fuels Ltd., UKD Air BP filton airfield Jet A-1 Air BP UK Ltd., Filton, UKE Russian TS-1 aviation kerosine Blended by Coryton; tested/approved

in RussiaF JP-4 wide-cut fuel Coryton Advanced Fuels Ltd., UK

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

approximately 1.2 l. Filtered fuels were stored in individual desic-cators and dried over silica gel for at least 24 h. This procedureensured that all the fuels had low concentrations of water initiallyand that re-humidification below ambient temperatures could becarried out without precipitating water. The desiccators werestored inside a fume cupboard to minimise exposure to ambientlight. Fuel B was noted to be particularly susceptible to UV degra-dation, turning a noticeable yellow after a few days. The cause isunknown but many studies have shown that photo-induced degra-dation may be a result of photo-oxidation and microbial degrada-tion [25–29].

The fuels were re-humidified in a closed vessel (a desiccatorcontaining a small amount of distiled water throughout the pro-cess) at the appropriate test temperature; the aim was to achievea 100% relative humidity (RH) environment without introducingfree water. A Mercia Scientific Humidicab was used for tests at orabove 5 �C, as shown in Fig. 1, and a Fisons FE 300H cabinet fortemperatures below 5 �C. To avoid over-saturating the fuels, eachdried fuel sample was cooled to below the nominal conditioningtemperature, transferred to the desiccator containing a smallamount of distiled water and allowed to warm to the test temper-ature. Approximately 300 ml of fuel was treated in this way foreach test. Fuel samples were gently stirred with a magnetic stirrerand the sample left to reach water saturated equilibrium at the rel-evant set temperature for at least 16 h. Previous work had shownthis to be sufficient time for the system to reach equilibrium[30]. The desiccators were fitted with septum caps to allow sam-pling of the fuel without disturbing the equilibrium. A thermocou-ple was located in the fuel so that an accurate reading of the fueltemperature could be taken when sampled.

Karl Fischer (KF) coulometric titration with a Metrohm KF831Coulometer was used to make the measurement of water in fuel.There are a number of test standards covering the use of KFCoulometers, but the most relevant in the context of fuel testingare ASTM E203 [31] and ASTM D6304 [32]. An SGE gas-tight2.5 ml glass syringe with a long stainless steel luer needle was usedfor all injections. The Metrohm Karl Fischer Coulometer with anSGE glass syringe is depicted in Fig. 2. A multiple injectionsequence with the needle tip below the surface of the anolytewas used for making the measurements (see A). The multiple injec-tion procedure had been shown to give a smaller spread of resultsthan the standard single injection technique. Type AG-H anolytesolution manufactured by Sigma Aldrich under the HydranalCoulomat brand was used in the coulometer cell. It gave the mostconsistent results, with the least tendency to form a biphasicsolution of fuel and anolyte.

172

173

174

4. Results and discussion

To generate the water solubility curves, the saturated watercontent of each fuel was measured at nominal temperatures of

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208210210

211

212

213

Table 2Some key characteristics of the test fuels.

Parameter A B C D E F

Aromatics, %v/v 10.7 16.4 24.1 13.7 13.0� 10.3Density @ 20 �C, kg m�3 810.4 785.1 810.8⁄ 792.6⁄ 794.4 779.5⁄

Freezing point, �C �58.1 �54.0 �57.0 �57.0 �64.0 �58.0Kinematic viscosity @ �20 �C, mm2 s�1 5.229 3.661 3.547 3.992 6.238� –MSEP, rating 92 99 92 96 – 62Naphthalenes, %v/v – – 1.89 – – 0.8Olefins, %v/v – 0.7 – – – –Smoke point, mm 29.0 25.0 21.0 27.0 25.2 27.0Sulphur, %m/m <0.01 0.0359 0.029 0.0461 <0.1 <0.01Flash point, �C 46.0 41.0 38.5 41.0 37.0 –IBP, �C 155.0 151.9 147.2 147.6 146.5 38.810%v/v recovered, �C 176.0 167.1 167.7 165.3 159.0 107.150%v/v recovered, �C 197.7 190.4 193.2 198.0 180.0 192.6FBP, �C 261.5 246.2 262.5 280.2 245.0} 244.0

} 98%v/v recovered.� Viscosity @ �40 �C.⁄ Density @ 15 �C.� %m/m.

Fig. 1. Mercia Scientific Humidicab with fuel samples and desiccators in position.

Fig. 2. Metrohm Karl Fischer Coulometer KF 831, with SGE glass syringe and needlein foreground.

J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx 3

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

�5, 0, 5, 15 and 25 �C. Each fuel was conditioned at the appropriatetemperature and the water content was determined using thestandard procedure given in Section 3. When fuel samples weretaken at test temperatures below 5 �C, the fuel in the conditioningdesiccator tended to gradually warm up, typically by 1–3 �C overthe course of a set of measurements; this was due to the condition-ing chamber door being opened multiple times to allow samples tobe taken. Although individual readings of fuel temperature weremade, for accuracy the KF measurements were always averagedover ten readings and therefore an average temperature is givenfor a set of readings at a particular temperature.

Some of the data sets were repeated because there appeared tobe a larger degree of scatter in the results than was consideredsatisfactory. There are a number of reasons for this; these mightbe related to laboratory temperature and/or humidity and possibly

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

condensation on the desiccator septum or syringe needle. It mustalso be remembered that the concentrations of water beingmeasured are very low, so that even traces can significantly skewthe results. For example, there was difficulty discerning the watercontent of the Sasol Fully Synthetic Jet A-1 (Fuel A) fuel attemperatures below 0 �C due to the very low concentration presentand the limitations of the Karl Fischer apparatus.

The KF measurements are displayed graphically in Fig. 3 usingbox plots. Each box plot provides a synopsis of the data distributionof presenting the 5-number summary which consists of the mini-mum, maximum, the upper (or 3rd) quartile, lower (or 1st) quartileand the median [33]. The box plots are plotted with deviation fromthe median as the horizontal axis. It aligns the medians of the boxplots at zero on the horizontal axis for visual comparison.

Since the water absorption characteristics of aviation fuels fol-lows an exponential trend with increasing temperature, expressingwater content on a log-lin graph (a logarithmic scale on the y-axis,and a linear scale on the x-axis) yields a straight line relationship:

S ¼ a exp bTð Þ ð1Þ

S is the solubility of water in fuel expressed in ppm m/m and T is thefuel temperature in �C. The base coefficient, a, is the water solubilityin fuel at 0 �C expressed in ppm m/m whilst the exponent

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

4 J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

coefficient, b, is the slope of the straight line relationship of the log-lin plot expressed in �C�1.

Data from the five kerosine fuels, Fuels A to E, were plotted withtrend lines in Fig. 4. For comparative purposes, the typical watersolubility of Jet A-1 from the CRC Handbook of Aviation Fuel Proper-ties [6, Figs. 2–32] (the fuel, hereafter, referred to as CRC Jet A-1)has also been added to Fig. 4. Since Fuel F is a wide-cut fuel, it isnot directly comparable to the kerosine fuels and is not includedin Fig. 4.

The ranking of the five kerosine fuels with respect to their affin-ity for water is quite clear from Fig. 4. Fuel A has the lowest affinityfor water with the smallest base coefficient, followed by Fuel D,then Fuel E and Fuel B, with Fuel C having the highest affinity

Fig. 3. Box plots of the KF meas

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

and the largest base coefficient. The solubility curves for the fivekerosine fuels generated from the KF determinations have basecoefficients of similar order to that of the solubility curve for theCRC Jet A-1. However, the CRC Jet A-1 has a smaller exponent coef-ficient compared to the five kerosine fuels, i.e., the gradient of theline for CRC Jet A-1 is less than that for the five kerosine fuels. Asnoted in Section 1, the CRC data were found to pre-date 1984.The differences observed may be attributed to recent changes infuel chemistry, or the particular fuel or average of fuels, repre-sented by the CRC plot.

The water solubility curve for Fuel F, the wide-cut fuel, isshown in Fig. 5. For comparison purposes, the typical watersolubility of JP-4 from the CRC Handbook of Aviation Fuel

urements for the test fuels.

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

Fig. 4. Water solubility versus temperature for the five kerosine fuels. Fig. 5. Water solubility versus temperature for the wide-cut fuel.

J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx 5

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

Properties [6, Fig. 2-32] (the fuel, hereafter, referred to as CRCJP-4) has also been added to Fig. 5. Fuel F has a larger basecoefficient, but a smaller exponent coefficient compared to thefive kerosine fuels. A large base coefficient indicates that Fuel Fhas a higher affinity for water. The solubility curve for Fuel Fhas a similar base coefficient to the solubility curve for theCRC JP-4. However, the solubility curve for CRC JP-4 has a largerexponent coefficient compared to that of Fuel F.

Fuel F was made to the NATO F-40 (AVTAG/FSII) specification.This includes various additives that are used for military fuelsbut are not normally included in fuels for civil aviation purposes.The FSII interacts with free water to form a non-freezing solutionbut it may indirectly also increase water solubility in the fuel. Sucha hypothesis is reasonable given the additive is Di-Ethylene GlycolMono-Methyl Ether (Di-EGMME) which would offer some polarityeven at the low concentrations used in aviation fuel of 0.10–0.15%v/v [1]. It is also known that smaller hydrocarbon molecules, whichwould be present in the wide-cut jet fuel, tend to be slightly betterat holding water in solution [34]. It was clear from the KF tests thatthe Fuel F was less likely to shed water than the kerosine fuels butit was not possible to say whether the effect was mainly due to thesmaller hydrocarbons or the FSII.

Estimates of the water shed from each fuel at 0 �C and �20 �Con cooling from 25 �C are given in Table 4. The solubility valuesin the table were derived from trend lines on the log-lin plots givenin Figs. 4 and 5.

Published data for Russian fuels [35,36] are plotted with Fuel Eand the CRC Jet A-1 for comparison in Fig. 6. Similarly, published

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

data for some alternative aviation fuels [37] are plotted with FuelA and the CRC Jet A-1 for comparison in Fig. 7. It should be notedthat Fuel E is a Russian specification fuel and Fuel A is a syntheticfuel and they were the most appropriate fuels to be used for com-parison in Figs. 6 and 7, respectively.

The coefficients of water solubility for a range of fuels, calcu-lated using Eq. (1), are summarised in Table 3. Moses [36] notedthat water solubility increases with aromatics content anddecreases with higher boiling range fuels. Appropriate parametersto characterise higher boiling range fuels may include flash point,and distilation factors such as initial boiling point (IBP), 10%v/vrecovered, 50%v/v recovered and final boiling point (FBP). ThePearson product-moment correlation coefficients (PPMCC) for var-ious fuel parameters are given in Table 5; these relate the basecoefficient, a, and the exponent coefficient, b, with aromatics con-tent and physical characteristic for higher boiling range hydrocar-bons. Both a and b have a high correlation with aromatics content,a, a medium correlation with flash point, b, and IBP, but a weakcorrelation for 10%v/v recovered, 50%v/v recovered and FBP. Apositive correlation is found between a and a, but negativecorrelations between a; b and IBP. This supports the theory thatwater solubility increases with aromatics content and decreaseswith higher boiling range fuels.

The flash point, b, may be measured using a number of standardtest methods (ASTM D56-05 [38], ASTM D93-13 [39], ASTMD3828-12a [40], IP 170 [41]). Similarly, the boiling range distribu-tion (i.e., IBP, 10%v/v recovered, 50%v/v recovered and FBP), may bemeasured using the ASTM D86-12 [42] test or the ASTM D2887-13

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

296

297

298

299

300

301302

304304

305

307307

308

309

310

311

312

313314316316

317319319

320

321

322

323324326326

327329329

330

331

332

Fig. 6. Water solubility versus temperature for the Russian fuels. Fig. 7. Water solubility versus temperature for the alternative fuels.

Table 3Coefficients of the water solubility exponential function.

Fuel a (ppm m/m) b (�C�1)

A 19.87 0.05231B 29.42 0.04781C 35.72 0.04273D 25.59 0.04829E 27.10 0.04944F 58.05 0.01730

CRC Jet A-1 33.06 0.03049CRC JP-4 53.79 0.03110

CTL 26.60 0.04320GTL 23.14 0.04620GTL/naphthenic 20.70 0.04780

TS-1/RT 38.81 0.03878T-1 (sample 1) 42.74 0.02627T-1 (sample 2) 26.17 0.03621T-1 (sample 3) 36.90 0.02906

6 J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

[43] test. These test methods can produce different results [6,Sections 2.4.1 and 2.8.1]. This may have an negative impact onthe correlations.

The base coefficient, a, and the exponent coefficient, b, may beexpressed as functions of aromatics content, a, and flash point, b.The following equations are proposed for a and b:

a ¼a1ðaÞ � a2ðbÞ; if both a and b are knowna1ðaÞ; if only a is known

�ð2Þ

b ¼b1ðaÞ � b2ðbÞ; if both a and bare knownb1ðaÞ; if only a is known

�ð3Þ

a1 is an approximation function from the linear regression betweena and a; b1 is an approximation function from the linear regressionbetween a and b; a2 and b2 are correction factors that are functionsof b. The simple linear regression between a and a, and a and b, isshown in Fig. 8a and b, respectively. The regression equations fora1 and b1 are given by:

a1 ¼ 1:1327aþ 9:2721 ð4Þ

b1 ¼ �0:00066782aþ 0:058620 ð5Þ

The correction factors, a2 and b2, are determined from simple linearregression between b and Da, and b and Db, respectively. Theparameters, Da and Db, are defined as Da ¼ a1 � a andDb ¼ b1 � b. The regression equations for a2 and b2 are given by:

a2 ¼ 0:18516b� 7:7072 ð6Þ

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

b2 ¼ �0:00011776bþ 0:0048928 ð7Þ

The water solubility for kerosine fuels may be estimated from Eq.(1) with the base coefficient, a, and the exponent coefficient, b, com-puted from Eqs. (2) and (3).

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390391392393394395396397398399400401

Table 4Estimated water dropout from fuels on reducing temperature based on KF results.

Fuel Saturated water solubility (ppm m/m) Water dropout (ppm m/m)

@ 25 �C @ 0 �C @ �20 �C Cooling from 25 �C

to 0 �C to �20 �C

A 73 20 7 53 66B 97 29 11 68 86C 104 36 15 68 89D 86 26 10 60 76E 93 27 10 66 83F 89 58 41 31 48

Table 5Correlation of the base coefficient, a, and the exponent coefficient, b, to variousparameters for the kerosine fuels.

Parameter a b

Aromatics content, a 0.9766� �0.9756�

Flash point, b �0.7224 0.6118IBP �0.6078 0.595610%v/v recovered �0.4003 0.271950%v/v recovered �0.2554 0.0302FBP �0.1500 �0.0963

� Since the specification for measuring and reporting the aromatics content for FuelE were different to that for the other fuels, Fuel E was therefore excluded from thecorrelation analysis.

Fig. 8. Simple linear regression between (a) a and a and (b) a and b.

J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx 7

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

5. Conclusions

Water solubility-temperature curves were created experimen-tally for five kerosine fuels and one wide-cut fuel. The Karl FischerCoulometer method was used to determine the saturated watercontent of each fuel over a range of temperatures between �5 �Cand 25 �C. The experimental solubility curves are comparable tothe water solubility curves of published data and of the Jet A-1

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

and JP-4 fuels reported in the CRC Handbook of Aviation Fuel Prop-erties [6].

The affinity for water of each fuel was measured; Coryton HighAromatics Jet A-1 (Fuel C) was found to have the highest and SasolFully Synthetic Jet A-1 (Fuel A) the lowest affinity for water. The JP-4 wide-cut fuel (Fuel F) had a moderately high affinity for water atambient temperatures, but showed the least tendency of any of thefuels to shed water as the temperature decreased. The presence ofFSII in this fuel was identified as a possible contributory factor.Coryton High Aromatics Jet A-1 (Fuel C) shed the greatest quantityof water as the temperature decreased which suggested that icingproblems in aircraft fuel systems will be greatest for fuels with ahigh concentration of aromatic hydrocarbons.

A generic equation is proposed to estimate the water solubilitygiven the aromatics content in % v/v and the flash point in �C

Acknowledgements

This work formed part of a European Aviation Safety Agency(EASA) funded research project titled ‘‘Water in Aviation Fuelunder Cold Temperatures conditions—WAFColT’’. The researchwork was conducted by Cranfield Defence and Security and theresearch programme was managed by Airbus Operations Limited.The authors wish to thank SASOL for the supply of the fully syn-thetic Jet A-1, PetroChina for the supply of the Chinese No. 3 jetfuel and Dr. David Parmenter of Airbus Operations for technicaldiscussions.

Appendix A. Determination of water content in fuel using themultiple injection technique with the Metrohm Karl FischerCoulometer KF 831

A clean, empty syringe of known weight was taken and the nee-dle inserted into the Karl Fischer (KF) cell via a septum; the anolytesolution was drawn into the syringe, then injected into the cell anda reading was taken of the water content. The syringe remained inposition in the cell and the anolyte solution was again drawn upinto the syringe and then re-injected into the cell; this processwas repeated until the water content measured was less than1 ppm m/m.

The syringe was removed from the KF cell and conditioned avi-ation fuel was drawn into the syringe and then weighed. The syr-inge containing fuel was inserted into the KF cell and the sampleinjected; a reading was taken of the water content. The calculatedweight of fuel injected was entered via the Metrohm Karl FischerCoulometer KF 831 keypad, so that the water content of the fuelcould be determined by mass. The syringe remained in the celland the anolyte solution was drawn up into the syringe, re-injectedand a further reading of the water content was taken. This processwas repeated a further two times to give a total of four readings.These readings were added together to give the total amount ofwater in that sample of aviation fuel. The water content of eachfuel was measured ten times at each temperature.

References

[1] DEF STAN 91-91. Turbine fuel, aviation kerosine type, Jet A-1 NATO Code: F-35Joint Service Designation: AVTUR. Ministry of Defence; UK, 7th ed.; 2011.Defence Standard 91-91. <http://www.dstan.mod.uk>.

[2] ASTM D1655. Standard specification for aviation turbine fuels. WestConshohocken, Pennsylvania, US: ASTM International; 2013. doi: http://dx.doi.org/10.1520/D1655, <http://www.astm.org>.

[3] GOST 10227-86. Jet fuels specifications. Moscow, Russia: Standards PublishingHouse; 1986.

[4] GB 6537. No. 3 Jet fuel. China: AQSIQ, SAC; 2006.[5] ExxonMobil Aviation. World jet fuel specifications with avgas supplement.

Leatherhead, UK: ExxonMobil Aviation International Ltd., TechnicalDepartment; 2008.

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/

402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468

469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534

535

8 J.K.-W. LamQ1 et al. / Fuel xxx (2014) xxx–xxx

JFUE 8085 No. of Pages 8, Model 5G

13 May 2014

Q1

[6] Coordinating Research Council. Handbook of aviation fuel properties. CRCreport no. 635; Warrendale, Pennsylvania 15096, USA: Society of AutomotiveEngineers, Inc. (distributor); 3rd ed.; 2004. <http://books.google.co.uk/books?id=0HxTAAAAMAAJ>.

[7] Hemighaus G, Boval T, Bacha J, Barnes F, Franklin M, Gibbs L, et al. Aviationfuels technical review. Houston, TX 77002, US: Chevron Global Aviation,Chevron Products Company; 2006. <http://www.chevronglobalaviation.com>.

[8] The American Petroleum Institute. Robust summary of information onkerosene jet fuel. American Petroleum Institute, Petroleum High ProductionVolume (HPV) Testing Group; 2010. <http://www.petroleumhpv.org>.

[9] ASTM D7566-13. Standard specification for aviation turbine fuel containingsynthesized hydrocarbons. West Conshohocken, Pennsylvania, US: ASTMInternational; 2013. doi: http://dx.doi.org/10.1520/D7566, <http://www.astm.org>.

[10] Baena-Zambrana S, Repetto SL, Lawson CP, Lam JKW. Behaviour of water in jetfuel – a literature review. Prog Aeros Sci 2013;60(0):35–44. http://dx.doi.org/10.1016/j.paerosci.2012.12.001. <http://www.sciencedirect.com/science/article/pii/S0376042112000875>.

[11] Feller D. Strength of the benzene-water hydrogen bond. J Phys Chem A1999;103(38):7558–61. http://dx.doi.org/10.1021/jp991932w. http://pubs.acs.org/doi/pdf/10.1021/jp991932w; <http://pubs.acs.org/doi/abs/10.1021/jp991932w>.

[12] Atwood JL, Hamada F, Robinson KD, Orr GW, Vincent RL. X-ray diffractionevidence for aromatic p hydrogen bonding to water. Nature1991;349(6311):683–4. http://dx.doi.org/10.1038/349683a0. <http://www.nature.com/nature/journal/v349/n6311/abs/349683a0.html>.

[13] Ruelle P, Kesselring U. Nonlinear dependence of the solubility of water inhydrocarbons on the molar volume of the hydrocarbon. J Solut Chem1996;25(7):657–65. http://dx.doi.org/10.1007/BF00972680.

[14] Lam JKW, Hetherington JI, Carpenter MD. Ice growth in aviation jet fuel. Fuel2013;113(0):402–6. doi: http://dx.doi.org/10.1016/j.fuel.2013.05.048, <http://www.sciencedirect.com/science/article/pii/S0016236113004602>.

[15] Oreshenkov A. Accumulation of water in jet fuels. Mathematical modeling ofthe process. Chem Technol Fuels Oils 2004;40(5):320–5. http://dx.doi.org/10.1023/B:CAFO.0000046266.83408.d7.

[16] Tomlinson S, Barker M, Venn D, Hickson L, Lam JKW. Mathematical model ofwater contamination in aircraft fuel tanks. SAE technical paper; 2011, 2011-01-2540. doi: http://dx.doi.org/10.4271/2011-01-2540.

[17] Crampton AB, Finn RF, Kolfenbach JJ. What happens to the dissolved water inaviation fuels? SAE technical paper; 1953, 530096. doi: http://dx.doi.org/10.4271/530096. <http://papers.sae.org/530096>.

[18] Solash J, Hazlett RN, Hall JM, Nowack CJ. Relation between fuel properties andchemical composition. 1. Jet fuels from coal, oil shale and tar sands. Fuel1978;57(9):521–8. http://dx.doi.org/10.1016/0016-2361(78)90036-4. <http://www.sciencedirect.com/science/article/pii/0016236178900364>.

[19] Kreutz TG, Larson ED, Liu G, Williams RH. Fischer–Tropsch fuels from coal andbiomass. In: 25th Annual international pittsburgh coal conference, vol.29(2.10); 2008. <https://www.princeton.edu/pei/energy/publications/texts/Kreutz-et-al-PCC-2008-10-7-08.pdf>.

[20] Daggett D, Hadaller O, Hendricks R, Walther R. Alternative fuels and theirpotential impact on aviation. Natl Aeronaut Space Admin NASA/TM2006;214365(6). <http://www.obsa.org>.

[21] Hileman JI, Wong HM, Ortiz D, Brown N, Maurice L, Rumizen M. The feasibilityand potential environmental benefits of alternative fuels for commercialaviation. In: Proceedings of the 26th international congress of the aeronauticalsciences; 2008. p. 5–8. <http://stuff.mit.edu:8001/afs/athena.mit.edu/dept/aeroastro/partner/reports/proj17/alt>.

[22] Hileman JI, Ortiz DS, Bartis JT, Wong HM, Donohoo PE, Weiss MA, et al. Near-term feasibility of alternative jet fuels. Tech Rep; Rand Corporation; 2009.<http://stuff.mit.edu:8001/afs/athena.mit.edu/dept/aeroastro/partner/reports/proj17/altfuelfeasrpt.pdf>.

[23] Hileman JI, Donohoo PE, Stratton RW. Energy content and alternative jet fuelviability. J Propul Power 2010;26(6):1184–96. http://dx.doi.org/10.2514/1.46232.

[24] Shonnard DR, Williams L, Kalnes TN. Camelina-derived jet fuel and diesel:sustainable advanced biofuels. Environ Prog Sust Energy 2010;29(3):382–92.http://dx.doi.org/10.1002/ep.10461.

Please cite this article in press as: Lam J-KW et al. Water solubility characterj.fuel.2014.04.091

[25] Karetnikova E, Rapoport V. Influence of UV irradiation on microbiologicaldegradation of petroleum products. Biol Bull 2008;35(5):533–5. http://dx.doi.org/10.1134/S1062359008050154.

[26] Pengerud B, Thingstad F, Tjessem K, Aaberg A. Photo-induced toxicity of northsea crude oils toward bacterial activity. Mar Pollut Bull 1984;15(4):142–6.http://dx.doi.org/10.1016/0025-326X(84)90235-2. <http://www.sciencedire-ct.com/science/article/pii/0025326X84902352>.

[27] Lee RF. Photo-oxidation and photo-toxicity of crude and refined oils. Spill SciTechnol Bull 2003;8(2):157–62. http://dx.doi.org/10.1016/S1353-2561(03)00015-X. <http://www.sciencedirect.com/science/article/pii/S135325610300015X>.

[28] Maki H, Sasaki T, Harayama S. Photo-oxidation of biodegraded crude oil andtoxicity of the photo-oxidized products. Chemosphere 2001;44(5):1145–51.http://dx.doi.org/10.1016/S0045-6535(00)00292-7. <http://www.sciencedirect.com/science/article/pii/S0045653500002927>.

[29] Atlas RM. Microbial degradation of petroleum hydrocarbons: anenvironmental perspective. Microbiol Rev 1981;45(1):180–209. <http://www.ncbi.nlm.nih.gov/pmc/articles/PMC281502>.

[30] Carpenter MD, Hetherington JI, Lao L, Ramshaw C, Yeung H, Lam JKW, et al.Behaviour of water in aviation fuels at low temperatures. In: Morris RE, editor.12th International conference on stability, handling and use of liquid fuels, vol.2. IASH 2011, Sarasota, Florida, USA. 16–20 October 2011; 2011. p. 1036–62[ISBN 978-1-61839-764-5].

[31] ASTM E203. Standard test method for water using volumetric Karl FischerTitration. West Conshohocken, Pennsylvania, US: ASTM International; 2008.doi: http://dx.doi.org/10.1520/E0203-08, <http://www.astm.org>.

[32] ASTM D6304. Standard test method for determination of water in petroleumproducts, lubricating oils, and additives by coulometric Karl FischerTitration. West Conshohocken, Pennsylvania, US: ASTM International; 2007.doi: http://dx.doi.org/10.1520/D6304-07, <http://www.astm.org>.

[33] Potter K, Hagen H, Kerren A, Dannenmann P. Methods for presenting statisticalinformation: the box plot. Vis Large Unstruct Data Sets 2006:97–106.

[34] Polak J, Lu BCY. Mutual solubilities of hydrocarbons and water at 0 and 25 �C.Can J Chem 1973;51(24):4018–23.

[35] Zherebtsov VL, Peganova MM. Water solubility versus temperature in jetaviation fuel. Fuel 2012;102(0):831–4. http://dx.doi.org/10.1016/j.fuel.2012.06.070. <http://www.sciencedirect.com/science/article/pii/S0016236112004905>.

[36] Moses CA. Properties of Russian jet fuels. Tech rep CRC Project No. AV-12-10.Coordinating Research Council (CRC). Alpharetta, GA 30022 USA; 2011. FinalReport.

[37] Curtin A. Testing to investigate the differences between the selected fuels inrelation to an aircraft fuel system and the in-tank materials. ALFA-BIRD –alternative Fuels and Biofuels for aircraft development. Willi-Bleicher-Str. 19,70174 Stuttgart, Germany: EU-VRi European Virtual Institute for IntegratedRisk Management; 2012. <http://www.alfa-bird.eu-vri.eu>.

[38] ASTM D56-05. Standard test method for flash point by tag closed cuptester. West Conshohocken, Pennsylvania, US: ASTM International; 2010. doi:http://dx.doi.org/10.1520/D0056-05R10, <http://www.astm.org>.

[39] ASTM D93-13. Standard test methods for flash point by Pensky-Martens closedcup tester. West Conshohocken, Pennsylvania, US: ASTM International; 2013.doi: http://dx.doi.org/10.1520/D0093, <http://www.astm.org>.

[40] ASTM D3828-12a. Standard test methods for flash point by small scale closedcup tester. West Conshohocken, Pennsylvania, US: ASTM International; 2012.doi: http://dx.doi.org/10.1520/D3828-12A, <http://www.astm.org>.

[41] IP 170. Determination of flash point – abel closed-cup method. 61 NewCavendish Street, London, W1G 7AR, UK: Energy Institute; 2009. [ISBN IP170-2935507, <http://www.energypublishing.org>].

[42] ASTM D86-12. Standard test method for distillation of petroleum products atatmospheric pressure. West Conshohocken, Pennsylvania, US: ASTMInternational; 2012. doi: http://dx.doi.org/10.1520/D0086-12, <http://www.astm.org>.

[43] ASTM D2887-13. Standard test method for boiling range distribution ofpetroleum fractions by gas chromatography. West Conshohocken,Pennsylvania, US: ASTM International; 2013. doi: http://dx.doi.org/10.1520/D2887, <http://www.astm.org>.

istics of current aviation jet fuels. Fuel (2014), http://dx.doi.org/10.1016/