Separation of Fatty Acid Methyl Esters by GC-Online Hydrogenation× GC

Dec 20, 2012 - The separation of fatty acid methyl esters (FAME) provided by a 200 m × 0.25 mm SLB-IL111 capillary column is enhanced by adding a sec...
0 downloads 0 Views 899KB Size
Subscriber access provided by UNIV OF ALABAMA BIRMINGHAM

Article

Separation of fatty acid methyl esters by GC-online hydrogenation x GC. Pierluigi Delmonte, Ali Reza Fardin-Kia, and Jeanne I. Rader Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac302707z • Publication Date (Web): 20 Dec 2012 Downloaded from http://pubs.acs.org on December 24, 2012

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1 1 2 3 4 5 6

Separation of fatty acid methyl esters by GC-online hydrogenation × GC.

Pierluigi Delmonte*, Ali Reza Fardin-Kia, Jeanne I. Rader

7 8 9 10

Office of Regulatory Science, Center for Food Safety and Applied Nutrition, Food and Drug Administration, College Park, MD, USA.

11 12

Running title: Separation of fatty acid methyl esters by GC-online hydrogenation x GC.

13 14

Key words:

15

SLB-IL111, GC×GC, fish oil, PUFA, fatty acids, hydrogenation, gas chromatography.

16 17

*Corresponding author:

18 19

Pierluigi Delmonte, Ph.D.

20

HFS-717

21

US Food and Drug Administration

22

5100 Paint Branch Pkwy

23

College Park, MD, 20740

24

Phone: 301-436-1779

25

Fax: 301-436-2622

26

[email protected]

27 28

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

2 29

ABSTRACT

30 31

The separation of fatty acid methyl esters (FAME) provided by a 200 m x 0.25 mm SLB-

32

IL111 capillary column is enhanced by adding a second dimension of separation (2D) in

33

a GC×GC design. Rather than employing two GC columns of different polarities or using

34

different elution temperatures, the separation in the two dimensional space is achieved by

35

altering the chemical structure of selected analytes between the two dimensions of

36

separation. A capillary tube coated with palladium is added between the first dimension

37

of separation (1D) column and the cryogenic modulator, providing the reduction of

38

unsaturated FAMEs to their fully saturated forms. The 2D separation is achieved using a

39

2.5 m x 0.10 mm SLB-IL111 capillary column and separates FAMEs based solely on

40

their carbon skeleton. The two dimensional separation can be easily interpreted based on

41

the principle that all the saturated FAMEs lie on a straight diagonal line bisecting the

42

separation plane, while the FAMEs with the same carbon skeleton but differing in the

43

number, geometric configuration or position of double bonds lie on lines parallel to the

44

1D time axis. This technique allows the separation of trans fatty acids and

45

polyunsaturated FAs (PUFA) in a single experiment and eliminates the overlap between

46

PUFAs with different chain lengths. To our knowledge, this the first example of GC×GC

47

in which a chemical change is instituted between the two dimensions in order to alter the

48

relative retentions of components and identify unsaturated FAMEs.

49 50 51

INTRODUCTION

52 53

Oils and fats extracted from products of marine origin such as fish and algae contain a

54

large number of saturated, monounsaturated and polyunsaturated fatty acids1,2. Gas

55

chromatography using mid and high polarity capillary columns was shown to provide

56

only a partial separation of the FAMEs prepared from these oils and fats1,2. Santercole et

57

al.1 recently compared the separations of menhaden oil FAMEs using a mid polarity GC

58

column (polyethylene glycol phase, PEG) and a high polarity column (100% cyanopropyl

59

siloxane phase, CPS). The CPS column used by the authors provided a more detailed

ACS Paragon Plus Environment

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

3 60

separation of positional and geometric isomers of unsaturated fatty acids compared to the

61

PEG counterpart but none of the sets of chromatographic conditions evaluated could

62

separate all of the FAMEs contained in the samples investigated1.

63 64

About a decade ago, Armstrong et al. showed that room temperature ionic liquids (RTIL)

65

possess the volatility, viscosity, solubility and polarity proprieties required for use as

66

stationary phases for capillary GC3,4. Used as stationary phases in GC, they showed dual

67

nature retention selectivity, separating polar molecules as a polar stationary phase and

68

non polar molecules as a non polar stationary phase3,4,5. The development of

69

dicationic6,7,8, trigonal tricationic9, and cross-linked10 RTILs led to the novel IL capillary

70

columns currently used for the separation of FAMEs11-14. Capillary columns coated with

71

RTILs are increasingly used for GC×GC separations due to their unique selectivity and

72

thermal stability15-27. The use of a column coated with an RTIL

73

(trihexyl(tetradecyl)phosphonium bis(trifluoromethane)sulfonamide) for GC×GC

74

separations was first tested by Seeley et al, using standard mixtures of organic

75

compounds with a wide range of functional groups15. Used for 1D or 2D, RTIL columns

76

have been used successfully for the separation of phosphorous–oxygen (P–O) containing

77

compounds16, FAMEs17-19, PCBs20, 21, perfume allergens22, pesticides in water23,

78

petroleum fractions containing nitrogen and sulphur compunds24,25. A GC×GC×GC

79

application for the separation of phosphorous–oxygen (P–O) containing compounds in

80

diesel samples has also been reported26. The use of RTIL columns for GC separations has

81

recently been reviewed in greater detail by Ragonese et al 27.

82 83

Delmonte et al. tested the novel SLB-IL111 ionic liquid column and showed that the

84

higher polarity of its stationary phase provides a higher selectivity for FAMEs based on

85

the number, geometric configuration and position of double bonds than other phases used

86

for FAME analysis12. Based on these findings, the authors developed a procedure capable

87

of separating most FAMEs prepared from milk fat using a 200 m x 0.25 mm SLB-IL111

88

capillary column13. Among other improvements, this methodology allowed the separation

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

4 89

of almost all positional/geometric isomers of 18:1 and 18:2 milk fat FAMEs. However,

90

certain FAMEs differing in chain length and number of double bonds still co-eluted. All

91

of the trans-18:1 FAMEs up to t15-18:1 eluted before cis 9-18:1. The biologically active

92

trans vaccenic acid (t11-18:1) was clearly separated from other trans-18:1, and in the

93

conjugated linoleic acid region (CLA) t7,c9- was separated from c9,t11-18:213.

94 95

Comprehensive two-dimensional gas chromatography can provide a more detailed

96

separation of FAMEs by combining the selectivity of two different gas chromatographic

97

separations 28. The orthogonality between the two separations is commonly achieved by

98

employing GC columns of different polarities, different elution temperatures for 1D and

99

2D, or a combination of both. Several combinations of experimental conditions have been

100

developed targeting FAMEs prepared from different lipid sources and biodiesel blends17-

101

19, 29-37

102

lengths, and are often preferred for the 1D separation. Highly polar capillary columns

103

provide selectivity based on the number, geometric configuration and position of double

104

bonds and are preferred for the 2D separation. Gu et al.17 tested several combinations of

105

capillary columns for the GC×GC separation of FAMEs prepared from samples of marine

106

origin. The authors obtained a group-type GC×GC separation of FAMEs with different

107

carbon numbers and number of unsaturations by using an apolar polydimethyl siloxane

108

capillary column for 1D combined with a polar ionic liquid column for 2D. Tranchida et

109

al.18 separated FAMEs prepared from menhaden oil using an SLB-IL5ms for 1D

110

combined with an SP2560 for 2D. These platforms could separate a large number of

111

PUFAs and divide them into groups based on their carbon number but provided limited

112

separation based on the geometric configuration and position of double bonds17,18.

. Non polar capillary columns separate FAMEs based primarily on their chain

113 114

Online hydrogenation of FAMEs using a capillary fused silica tube coated with Pt or Pd

115

before or after GC separation was used in the early stages of lipid research38-40.

116

Schomburg et al.38 prepared a 300 mm x 0.35 mm I.D. capillary reactor by dynamic or

117

static coating, loading Pt or Pd on the capillary tube as an acetyl-acetonate salt dissolved

118

in MeCl2. The salt was successively dried on the inner surface of the tube, and

119

decomposed into metallic Pt or Pd by heating at 200 °C in a mild flow of carrier gas. The

ACS Paragon Plus Environment

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

5 120

efficiency of the Pt capillary reactor was tested by analyzing several types of unsaturated

121

compounds, including unsaturated FAMEs. The authors recommended using Pd over Pt

122

for the reduction of double bonds in FAMEs because the stronger reactivity of the latter

123

might cause the reduction of the ester group. Le Quere et al.40 characterized unsaturated

124

cyclic fatty acid monomers using a post column 60 cm x 0.32 mm capillary tube coated

125

with Pd prior to the MS detector.

126 127

This study describes the development of a novel approach to the GC×GC separation of

128

FAMEs in which the analytes undergo chemical reduction between 1D and 2D, and 2D

129

separates the products of the reaction. Unsaturated FAMEs (U-FAME) are reduced to

130

their fully saturated form by passing them through a capillary tube coated with Pd

131

(reactor) in the presence of H2 as the carrier gas. The GC×GC separation is achieved by

132

using two capillary columns coated with the same highly polar stationary phase, and

133

maintained at the same temperature. To our knowledge, this is the first GC×GC

134

application in which the two dimensional separation is achieved by modifying the

135

chemical structure of the analytes instead of by applying different chromatographic

136

conditions.

137 138 139

MATERIALS AND METHODS

140 141

Fatty acid methyl ester reference mixture GLC 463 and other FAMEs reference materials

142

used for identification purposes were purchased from Nu-chek Prep, Inc. (Elysian, MN).

143

Menhaden oil and Pd acetyl acetonate were purchased from Sigma Aldrich (St. Louis,

144

MO). The FAMEs prepared from human colon adenocarcinoma cells HT-29 were kindly

145

provided by Christian Degen of University of Jena (Germany). The origin and

146

preparation of this sample are described in a previous study41.

147 148

Capillary reactor preparation. Fifty milligrams of Pd acetyl acetonate were weighed in

149

a 10 ml volumetric flask and dissolved with MeCl2. The solution was loaded into a 4 m x

150

0.15 mm deactivated fused silica capillary tube using a 1 ml glass syringe, allowing

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

6 151

several drops to exit from the end side. The capillary tube was dried for 2 days under

152

vacuum with an end plugged with silicone vacuum grease and then maintained for 2

153

hours at 200 °C in the GC oven under a mild flow of hydrogen. If the crystallization and

154

thermal decomposition of the acetyl acetonate salt produce a homogeneous layer of Pd,

155

the expected thickness is 5 nm. A piece of uncoated silica capillary tube was placed

156

between the injector port and the Pd reactor to avoid contamination. The capillary tube

157

coated with metallic Pd was cut in 50 cm portions to be used as capillary reactors.

158 159

GC×GC conditions. Separations were achieved with an Agilent 7890A gas

160

chromatograph (Agilent Tech., Wilmington, DE) combined with a Zoex ZX2 dual stage

161

cryogenic modulator (Houston, TX). The injection port of the gas chromatograph was

162

connected to a 1034 kPa electronic pressure control module and hydrogen was used as

163

the carrier gas. The fluidic path of the system is shown in Figure 1. A Supelco SLB-

164

IL111 capillary column (200 m x 0.25 mm I.D., 0.2 µm coating, Bellefonte, PA) was

165

used for 1D separation and the same capillary column but with different dimensions (2.75

166

m x 0.10 mm I.D., 0.08 µm coating) was used for the 2D separation. The cryogenic

167

modulator first loop consisted of a 2 m x 0.10 mm deactivated capillary tube and the

168

modulation (first stage) was set at 25 cm from the beginning of the tube. The second

169

stage of modulation was achieved using part of the 2D column and the modulation spot

170

was set at 25 cm from the column front end leaving 2.50 m of 2D column for separation.

171

The 50 cm x 0.15 mm Pd capillary reactor was placed between the 1D column and the

172

modulator. The timing of the modulation was set to 3.5 s, the temperature of the hot jet

173

was set at 350 °C with a pulse of 350 ms. The column oven temperature was set to 170

174

°C, and the injector port at 300 °C. Chromatography was acquired in constant pressure

175

mode at 690 kPa and the split flow was set to 125.69 ml/min. The FID was maintained at

176

250 °C and fed with 30 ml/min of nitrogen make up gas, 400 ml/min air, and 25 ml/min

177

of hydrogen. All capillary connections were made with Ultimate unions (Agilent

178

Technologies).

179 180

GC×GC conditions, temperature program. The separation was achieved using the

181

same instrumental configuration used for isothermal separation, with the exception of the

ACS Paragon Plus Environment

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

7 182

100 m x 0.25 mm SLB-IL111 used for 1D. The elution temperature was set to 130 °C per

183

23 minutes, then increased 1.3 °C/min to 220 °C. The inlet pressure was maintained at

184

483 kPa per 23 minutes, and then raised at 1.514 kPa/min until 588 kPa. The modulation

185

time was set to 2 s, and the chromatograms were adjusted by shifting the 2D phase -1.8 s.

186 187

GC-MS identification of FAMEs. Identities of FAMEs were confirmed or achieved by

188

mono-dimensional GC-MS using a Micromass GCT TOF mass spectrometer

189

(Manchester, UK) coupled with an Agilent 6890N GC. The separation was achieved

190

using a 100 m x 0.25 mm SLB-IL111 capillary column maintained at 170°C and H2

191

carrier gas at the pressure of 80 kPa. Experiments were carried out in CI+ mode using

192

isobutane as the ionization gas. The source was maintained at 150°C, the electron energy

193

was 70 eV, the emission current 250 µA, and the CI gas flow was set to 10. The

194

molecular ion of FAMEs identified the chain length and the number of unsaturations.

195 196 197

RESULTS AND DISCUSSION

198 199

The current preferred approach for quantitation of FAMEs is mono-dimensional GC

200

using the longest and most polar capillary columns available. Methods for the

201

quantitation of trans fatty acids and PUFAs require the separation of FAMEs based on

202

the number and geometric configuration of their double bonds. Rather than enhancing

203

these separations by adding a second dimension of separation, most GC×GC methods use

204

a low polarity column for 1D, thus separating FAMEs based primarily on their chain

205

length, combined with a short polar 2D column providing limited separation based on the

206

number, position and geometry of double bonds.

207 208

In this study, we selected for 1D separation the same chromatographic conditions we

209

prefer for mono-dimensional GC analyses: a highly polar 200 m x 0.25 mm SLB-IL111

210

column maintained at 170°C. After being separated by the 1D column, the U-FAMEs are

211

reduced to their carbon skeleton by the capillary reactor and the products of reaction are

212

separated using a 2.5 m x 0.10 mm SLB-IL111 maintained at the same temperature as

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

8 213

1D. This GC×GC system requires a very simple hardware configuration: the two

214

capillary columns are maintained in the same oven and there is no need for supplying

215

extra carrier gas at the mid-point before the 2D column, as in the case of using larger

216

diameter columns for 2D separation (i.e., flow modulation).

217 218

Fatty acid methyl esters prepared from samples of marine origin are characterized by a

219

high sample dimensionality, primarily derived from the presence of double bonds on their

220

alkyl chains42. The number, position (individual, or cumulative using n-x nomenclature)

221

and geometric configuration of double bonds generate a number of sample dimensions

222

higher than the number of separation dimensions that can be combined in a single

223

chromatographic system. The hydrogenation of double bonds reduces the dimensionality

224

of the sample solely to chain length, or to the chain length and the position of the methyl

225

group substituent if iso- and anteiso-FAMEs are also considered. While the 1D column

226

provides a complex peak distribution by retaining FAMEs based on all sample

227

dimensions, the 2D provides an ordinate separation based only on one sample dimension

228

(or two, including iso-/anteiso- FAMEs). The result is a GC×GC separation in which

229

FAMEs are eluting from 1D in a “disordered” manner, and are then “ordered” by 2D into

230

lines based on their chain length dimension. If the desired separation is limited to straight

231

chain C8-C26 FAMEs, a theoretical 2D peak capacity of 19 can be sufficient to separate

232

all the products of hydrogenation. The low peak capacity required for performing the 2D

233

separation allows the rapid modulation time of 3.5 s.

234 235

The highly polar column used for 2D was selected based on its lower retention for long

236

chain saturated FAMEs relative to short/medium chain saturated FAMEs, in comparison

237

with lower polarity columns. It also allowed the use of a single GC oven, thus

238

simplifying the hardware configuration. The disadvantage of this selection is the lower

239

solubility of the FAMEs in the highly polar liquid phase, resulting in a lower loading

240

capacity. Capillary reactors were prepared by loading either 0.5 or 1.0% (w/v) Pd acetyl

241

acetonate in MeCl2 onto the capillary tube. Capillary reactors coated using the 1.0%

242

solution as short as 5 cm were observed to provide quantitative reduction of U-FAMEs.

243

The 50 cm version prepared using the 0.5% solution proved to be less dependent on the

ACS Paragon Plus Environment

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

9 244

uniformity of the Pd coating layer and was preferred for this study. Combinations of

245

excessive Pd load and/or length of the capillary reactor, such as use of a capillary of at

246

least 1 cm prepared using a 5% Pd acetyl acetonate solution, were observed to generate

247

1D peak tailing and the degradation of FAMEs. This was recognized by the appearance

248

of peaks in proximity to the 2D unretained band.

249 250

Figure 2 shows the separation of the FAMEs contained in reference mixture GLC 463.

251

The two dimensional separation can be easily interpreted using two basic rules: 1) all the

252

saturated FAMEs lie on a straight diagonal line bisecting the separation plane, as

253

expected from the lack of orthogonality between the 1D and 2D chromatographic

254

conditions, and 2) the FAMEs with the same carbon skeleton but differing in the

255

number/geometric configuration/position of double bonds lie on lines parallel to the 1D

256

time axis, based on the fact that the reduction converted them into the same compound.

257

The result is a two dimensional separation in which the 1D separation shows limited loss

258

of resolution compared to its mono-dimensional equivalent, as shown by the fact that the

259

18:1 and 20:1 FAMEs contained in this reference mixture are still separated from each

260

other. The most critical co-elutions affecting the same separation by mono-dimensional

261

GC are instead resolved by 2D: the n:1 FAMEs are separated from the saturated (n+1):0

262

(eg, 18:1 vs. 19:0), the 18:2 FAMEs are separated from the 20:0 and 19:1, and the 18:3

263

FAMEs are separated from the 20:1 and 21:0.

264 265

The pressure of 690 kPa applied to the injection port provided an average linear velocity

266

of 17 cm/s for 1D and 230 cm/s for 2D, based on the retention times of hexane, an un-

267

retained compound. Two dimension chromatograms were adjusted by shifting the 2D

268

phase by the estimated holdup time, 1.1 s. The 0 s time of the chromatogram corresponds

269

to 1.1 s of separation, while 3.5 s correspond of 1.1 s of the subsequent modulation cycle.

270

The 3.5 s modulation time allowed the separation of FAMEs with up to 23 carbons in a

271

single modulation cycle, while FAMEs with 24 or more carbons if present elute during

272

the subsequent modulation cycle. If the sample does not contain FAMEs with more than

273

21 carbons, as in the case of several vegetable oils, the 2D modulation time can be

274

reduced to 2 s. The combination of the long 1D column (200 m) and the 3.5 s modulation

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

10 275

time allowed a modulation ratio (MR) of at least 3 for FAMEs with at least 10 carbons.

276

The isothermal and constant flow elution conditions applied in this study caused broad

277

late eluting peaks, and the last eluted FAME (22:6) showed an MR of 15 or more

278

according to its load on the column.

279 280

Figure 3 shows the separation of FAMEs prepared from menhaden oil. Identification of

281

FAMEs was achieved by combining information obtained from the separations of

282

available reference materials, fractionation using silver ion chromatography13, GC-MS

283

analysis, and available literature1,2,12,13. The FAMEs prepared from menhaden oil were

284

fractionated by silver ion chromatography into saturated FAMEs, trans-monounsaturated

285

FAMEs (MUFA), cis-MUFAs, and PUFAs as previously described13. The FAMEs in the

286

saturated FAMEs fraction were identified by comparison with available reference

287

material and separations reported in the literature1,2,12,13.The FAMEs isolated in the cis-

288

and trans-MUFAs fractions were identified using previously prepared and characterized

289

reference mixtures12,13,43,44. The identification of the FAMEs in the PUFA fraction was

290

achieved by integrating identifications reported in the literature1,2,13 using different

291

separation columns with GC-MS data obtained in this study. The molecular ion of PUFA

292

indicated the n:x structure of the FAME, and, for each n:x series, the location of double

293

bonds was obtained by comparison with identifications reported in the literature1,2. The

294

separation of PUFA up to 22:6 requires 165 min, and FAMEs with different chain lengths

295

are separated by 2D on different lines. The separation on the two dimensional space

296

shows that several C16, C18 and C20 U-FAMEs require a second dimension of

297

separation in order to be resolved. Valuable information provided by this separation

298

technique is the identification of the FAME carbon skeleton, including the presence of

299

branches. Branched chain n:x FAMEs (i.e. iso and anteiso isomers) are eluted by 2D on

300

lines located between the C(n-1) and Cn FAME. If branched chain FAMEs with

301

unsaturations are present, the 2D retention time indicates the iso or anteiso structure and

302

the 1D retention time provides information regarding the number and geometric

303

configuration of double bonds. Obtaining the same information utilizing other analytical

304

techniques including mass spectrometry might be challenging, especially if the targeted

305

FAME is present in low concentration and/or it co-elutes with other FAMEs.

ACS Paragon Plus Environment

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

11 306 307

Figure 4 shows the separation of the FAMEs prepared from menhaden oil from 14:0 to

308

18:3. The two dimensional separation of the 16:1 and 17:0 FAME isomers provides a

309

dramatic improvement over the most refined separation obtained by mono-dimensional

310

GC: the 16:1 isomers are separated by 1D along the C16 line, while the 17:0 branch

311

chain isomers are eluted on the saturated FAME diagonal line. The x1 compound

312

analyzed by mass spectrometry showed a skeleton constituted of 17 carbons and one

313

unsaturation or a ring. The 2D retention time indicates that this FAME is reduced by Pd

314

because it elutes below the diagonal line bisecting the plane, and it is not an n-/iso-

315

/anteiso-17:1 FAME because its 2D retention time is different. The 17:1 and 16:2

316

FAMEs are separated on different lines and, as in the case of several other co-elutions,

317

they could be resolved only by taking advantage of the 2D separation. Of particular

318

interest was the identification of the 16:4 FAME coeluting with c9,c12-18:2 in 1D, of the

319

16:4 FAME co-eluting with c11-20:1 and 21:0, and the 16:2 FAME co-eluting with c5-

320

18:1 in the middle of the 18:1 cluster. The separation described here not only provides a

321

unique separation capability but it also allows a simple identification of the separated

322

FAMEs.

323 324

The application presented in this study is expected to be suitable for the separation of

325

FAMEs prepared from complex samples of various origins. Figure 5 shows the

326

separation of FAMEs prepared from human colon adenocarcinoma cells HT-29 incubated

327

with conjugated fatty acids. More details regarding the composition and the biological

328

significance of this sample can be found in the original study41. In this separation, the

329

diagonal line of saturated FAMEs bisecting the plane is accompanied by at least two

330

similar other lines, probably not belonging to FAMEs. No attempt was made to identify

331

these compounds. The identification of the FAMEs contained in this sample is

332

challenging, but the model of interpretation described in this study simplifies the

333

identification of most FAMEs based on the 2D lines on which they are eluted. A single

334

dimension of separation can provide a very limited resolution of the FAMEs prepared

335

from this sample, while the chromatographic system described here can separate these

336

FAMEs from each others and from other interfering compounds. As an example, while

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

12 337

the C16 PUFAs are separated in the two dimension plane from other compounds

338

contained in this sample, their separation and identification by mono-dimensional GC is a

339

very challenging task.

340 341

The GC×GC separation of marine oil FAMEs can be further optimized by applying an

342

elution temperature program. Figure 6 shows the separation of FAMEs prepared from

343

menhaden oil using a 100 m x 0.25 mm SLB-IL111 for 1D and a temperature program

344

from 130°C to 220°C. The chromatogram is adjusted by shifting the phase 1.8 s. The

345

saturated FAMEs with 14 or more carbons are eluted in an almost straight line parallel to

346

the time axis, and, for each chain length, the unsaturated FAMEs are eluted on crescents.

347

The temperature elution program provided an optimized utilization of the separation

348

space, shortened the separation to 90 minutes, and allowed the reduction of the

349

modulation time to 2 s. The GC×GC separation shown in Figure 6 is characterized by a

350

reversed pattern compared with the separations provided by the system constituted by a

351

non-polar column for 1D and a polar column for 2D17,18: for each chain length, the

352

saturated FAME elutes first at the top of the crescent and unsaturated FAMEs follow in

353

order of increasing number of double bonds. The result is still a group-type GC×GC

354

separation based on the FAME chain length and number of unsaturations, but the groups

355

are arranged in reversed 2D elution order.

356 357 358

CONCLUSION

359 360

The GC×GC chromatographic system described here provides an enhancement of our

361

preferred method for trans fatty acid separation by adding a second dimension of

362

separation capable of resolving most of the FAMEs not separated by 1D. The FAMEs

363

can be easily identified based on a simple interpretation model: saturated FAMEs lie on a

364

diagonal line bisecting the separation plane and FAMEs with the same carbon skeleton

365

lie on lines parallel to the 1D time axis. The chromatographic conditions described have

366

been shown to allow the separation of the FAMEs prepared from the most complex

367

samples, including fish oil and biological extracts. The proposed approach, reducing

ACS Paragon Plus Environment

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

13 368

unsaturated FAMEs to their saturated form between the two dimensions of separation,

369

provided an enhanced separation compared to GC×GC systems based solely on the

370

orthogonality of the two separations. Reducing U-FAMEs prior to 2D separation also

371

allowed faster modulation times, because the reaction products (saturated FAME) show

372

significantly lower retention times when they are separated on highly polar columns. We

373

invite the users of GC×GC to consider whether or not the described chromatographic

374

system can be considered orthogonal, despite the fact that there is no orthogonality

375

between the separation columns selected for 1D and 2D.

376 377 378

AKNOWLEDGEMENTS

379 380

We are grateful to Dr. John Kramer for contributing to the identification of PUFA and to

381

Edward B. Ledford Jr. of Zoex Corporation for helpful discussions and guidance in the

382

optimization of the GC×GC instrumentation. We also thank Christian Degen (University

383

of Jena, Germany) for generously providing the FAMEs prepared from human colon

384

adenocarcinoma cells HT-29 incubated with conjugated fatty acids.

385 386 387

REFERENCES

388 389

(1) Santercole, V.; Delmonte, P.; Kramer, J.K. Lipids 2012, 47, 329-344.

390 391

(2) AOCS Official Method Ce 1i-07, AOCS Press, Urbana, IL, 2007.

392 393

(3) Armstrong, D.W.; He, L.; Liu, Y.S.; Anal. Chem. 1999, 71, 3873-3876.

394 395

(4) Anderson, J.L.; Armstrong, D.W. Anal. Chem. 2003, 75, 4851-4858.

396 397

(5) Anderson, J.L.; Ding, J.; Welton, T.; Armstrong, D.W. J. Am. Chem. Soc. 2002, 124,

398

14247-14254.

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

14 399 400

(6) Anderson, J.L.; Ding, R.; Ellern, A.; Armstrong, D.W.; J. Am. Chem. Soc. 2005, 127,

401

593-604.

402 403

(7) Huang, K.; Han, X.; Zhang, X.; Armstrong, D.W.; Anal. Bioanal. Chem. 2007, 389,

404

2265-2275.

405 406

(8) Breitbach, Z.S.; Armstrong, D.W.; Anal. Bioanal. Chem. 2008, 390, 1605-1617.

407 408

(9) Payagala, T.; Zhang, Y.; Wanigasekara, E.; Huang, K.; Breitbach, Z.S.; Sharma, P.S.;

409

Sidisky, L.M.; Armstrong, D.W. Anal. Chem. 2009, 81, 160-173.

410 411

(10) Anderson, J.L.; Armstrong, D.W. Anal. Chem. 2005, 77, 6453-6462.

412 413

(11) Ragonese, C.; Tranchida, P.Q.; Dugo, P.; Dugo, G.; Sidisky, L.M.; Robillard, M.V.;

414

Mondello, L.; Anal. Chem. 2009, 81, 5561-5568.

415 416

(12) Delmonte, P.; Fardin Kia, A.R.; Kramer, J.K.; Mossoba, M.M.; Sidisky, L.; Rader,

417

J.I. J. Chromatogr. A 2011, 1218, 545-554.

418 419

(13) Delmonte, P.; Fardin-Kia, A.R.; Kramer, J.K.; Mossoba, M.M.; Sidisky, L.;

420

Tyburczy, C.; Rader, J.I. J. Chromatogr. A 2012, 1233, 137-146.

421 422

(14) Destaillats, F.; Guitard, M.; Cruz-Hernandez, C. J. Chromatogr. A 2011, 1218,

423

9384-9389.

424 425

(15) Seeley, J.V ; Seeley, S.K. ; Libby, E.K. ; Breitbach, Z.S. ; Armstrong, D.W. Anal.

426

Bioanal. Chem. 2008, 390, 323-332.

427 428

(16) Reid, V.R. ; Crank, J.A. ; Armstrong, D.W. ; Synovec, R.E. J. Sep. Sci. 2008, 31,

429

3429-3436.

ACS Paragon Plus Environment

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

15 430 431

(17) Gu, Q.; David, F.; Lynen, F.; Vanormelingen, P.; Vyverman, W.; Rumpel, K.; Xu,

432

G.; Sandra, P. J. Chromatogr. A 2011, 1218, 3056-3063.

433 434

(18) Tranchida, P.Q.; Franchina, F.A.; Dugo, P.; Mondello, L. J. Chromatogr. A 2012

435

1255, 171-176.

436 437

(19) Villegas, C.; Zhao, Y.; Curtis, J.M. J. Chromatogr. A 2010, 1217, 775-784.

438 439

(20) Zapadlo, M.; Krupcík, J.; Májek, P.; Armstrong, D.W.; Sandra, P. J. Chromatogr. A

440

2010, 1217, 5859-5867.

441 442

(21) Zapadlo, M.; Krupčík, J.; Kovalczuk, T.; Májek, P.; Spánik, I.; Armstrong, D.W.;

443

Sandra, P.; J. Chromatogr. A 2011, 1218, 746-751.

444 445

(22) Purcaro, G.; Tranchida, P.Q.; Ragonese, C.; Conte, L.; Dugo, P.; Dugo, G.;

446

Mondello, L. Anal. Chem. 2010, 82, 8583-8590.

447 448

(23) Purcaro, G.; Tranchida P.Q., Conte, L.; Obiedzińska, A.; Dugo, P.; Dugo, G.;

449

Mondello, L. J. Sep. Sci. 2011, 34, 2411-2417.

450 451

(24) Dutriez, T.; Borras, J.; Courtiade, M.; Thiébaut, D.; Dulot, H.; Bertoncini, F.;

452

Hennion, M.C. J. Chromatogr. A 2011, 1218, 3190-3199.

453 454

(25) Mahé, L.; Dutriez, T.; Courtiade, M.; Thiébaut, D.; Dulot. H.; Bertoncini, F. J.

455

Chromatogr. A 2011 1218, 534-544.

456 457

(26) Siegler, W.C. ; Crank, J.A. ; Armstrong, D.W. ; Synovec, R.E. J. Chromatogr. A

458

2010, 1217, 3144-3149.

459

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

16 460

(27) Ragonese, C.; Sciarrone, D.; Tranchida, P.Q.; Dugo, P.; Mondello, L. J.

461

Chromatogr. A 2012, 1255, 130-144.

462 463

(28) Adahchour, M., Beens, J., Vreuls, R. J. J., and Brinkman, U. A. T. Trends Anal.

464

Chem. 2006, 25, 438– 454.

465 466

(29) Mondello, L.; Casilli, A.; Tranchida, P.Q.; Dugo, P.; Dugo, G. J. Chromatogr. A

467

2003, 1019, 187-196.

468 469

(30) Adahchour, M.; Jover, E.; Beens, J.; Vreuls, R.J.J.; Brinkman, U.A.T. J.

470

Chromatogr. A 2005, 1086, 128-134.

471 472

(31) de Geus, H.J.; Aidos, I.; de Boer, J.; Luten, J.B.; Brinkman, U.A.T. J. Chromatogr.

473

A 2001, 910, 95-103.

474 475

(32) Western, R.J.; Lau, S.S.G.; Mariott, P.J.; Nichols, P.D. Lipids 2002, 37, 715-724.

476 477

(33) Harynuk, J.; Vlaeminck, B.; Zaher, P.; Marriott, P.J. Anal. Bioanal. Chem. 2006,

478

386, 602-613.

479 480

(34) Seeley, J.V.; Seeley, S.K.; Libby, E.K.; McCurry, J.D. J. Chromatogr. Sci. 2007,

481

45, 650-656.

482 483

(35) Tiyapongpattana, W.; Wilairat, P.; Marriott, P.J. J. Sep. Sci. 2008, 31, 2640-2649.

484 485

(36) Adam, F.; Bertoncini, F.; Coupard, V.; Charon, N.; Thiebaut, D.; Espinat, D.;

486

Hennion, M.C. J. Chromatogr. A 2008, 1186, 236-244.

487 488

(37) De Koning, S.; Janssen, H.G.; Brinkman, U.A.T. LC-GC Eur. 2006, 18, 590-600.

489

ACS Paragon Plus Environment

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

17 490

(38) Schomburg, G.; Hubinger, E.; Husmann, H.; Weeke, F. Chromatographia 1982, 16,

491

228-232.

492 493

(39) Henneberg, D.; Henrichs, U.; Schomburg, G. J. Chromatogr. 1975, 112, 343–352.

494 495

(40) LeQuere, J.L.; Semon, E.; Lanher, B.; Sebedio, J.L. Lipids 1989, 24, 347-350.

496 497

(41) Degen, C.; Ecker, J.; Piegholdt, S.; Liebisch, G.; Schmitz, G.; Jahreis, G. Biochim.

498

Biophys. Acta 2011, 1811, 1070-1080.

499 500

(42) Giddings, J.C. J. Chromatogr. A 1995, 703, 3-15.

501 502

(43) Delmonte, P.; Kia, A.R.; Hu, Q.; Rader, J.I. J. AOAC Int. 2009, 92, 1310-1326.

503 504 505

(44) Delmonte, P.; Hu, Q.; Kia, A.R.; Rader, J.I. J. Chromatogr. A 2008, 1214, 30-36.

506 507 508 509 510 511 512 513 514 515

C20 C19 C18 C17 C16 C15

C24

516

For TOC only

ACS Paragon Plus Environment

Analytical Chemistry

Inj. Port

FID

690 kPa

GC oven 170 ºC Modulator SLB-IL111

200 m x 0.25 mm

2.50 m x 0.10 mm

1D (200 m)

Reducer

25 cm

SLB-IL111 25 cm

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

Fused silica 2 m x 0.10 mm

2D (0.25 + 2.50 m)

Figure 1. Scheme of GC×GC apparatus.

ACS Paragon Plus Environment

Page 19 of 23

2.50

c1322:1

22:0

22:2 n-6

22:3 n-3

22:4 n-6

22:5 n-3

22:6 n-3

2.00

21:0 20:1

20:2 n-6

c5 c8 c11

1.50

20:3 n-6

20:0

18:1*

17:0 16:0

t9 c9

18:2

t9,t12

18:0

n-3

n-6

18:3 n-6

n-3

c10-17:1 16:1

0.50

24:0 15:0 14:0

20:5 n-3

20:4 n-6

c719:1

19:0 1.00

2D, 2.5 m x 0.10 mm SLB-IL111, seconds

c10-15:1

c15-24:1

c9-14:1 13:0 12-13:1 12:0 11-12:1 11:0 10-11:1 10:0 0.00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

18.96

30.64

42.31

53.98

65.65

77.32

88.99

100.66

112.33

124.00

135.67

147.34

1D, 200 m x 0.25 mm SLB-IL111, minutes

Figure 2. GC×GC separation of the FAME contained in reference mixture GLC 463 (Nu-chek Prep., Inc.). Instrumental configuration as shown in Figure 1. *In order of elution, t9-, t11-, c6-, c9- and c11-18:1.

ACS Paragon Plus Environment

159.01

Analytical Chemistry

c11 22:1 c13 c15

22:0

2.50 1.50 1.00 2.00 0.50 2D, 2.5 m x 0.10 mm SLB-IL111, seconds

22:5 n-6 22:4 n-3

22:4 n-6

22:5 n-3

22:6 n-3

21:5 n-3 21:0 20:2

20:1

NMI n-6

20:0

20:3 n-6 20:4 n-4 n3

20:5 n-3

n-3

19:1

19:0 18:0

n-6

18:1 18:2

18:3

n-3 n-6 n-4

18:4 n-6

n-3

n-1

17:1 17:0 16:0

c15-24:1

16:1 16:2 16:3 16:4

24:0

15:0 14:1 14:0

0.00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

18.96

30.64

42.31

53.98

65.65

77.32

88.99

100.66

112.33

124.00

135.67

147.34

159.01

1D, 200 m x 0.25 mm SLB-IL111, minutes

Figure 3. GC×GC separation of the FAME prepared from a menhaden fish oil sample. Instrument configuration as shown in Figure 1. NMI, non-methylene interrupted.

ACS Paragon Plus Environment

Page 21 of 23

20:1 1.60

20:0

c7 c9

c11 c13 c15

1.40 1.20 1.00 0.60 0.80 2D, 2.5 m x 0.10 mm SLB-IL111, seconds

c1119:1 19:0 c9

18:1 c11

t8 c7 t6 c5 /t9

18:0

18:2 x2

c5, c11

c13 c14

c8, n-6 c11

18:3 n-4

n-6

n-4

n-3

i-18:0

17:1

17:0 y2 i-17:0

16:0

c5

y1 i-16:0

15:0

c7 c9

ai-15:0 i-15:0

0.40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

14:0

c7

24.69

c9

26.90

t6

ai17:0

c7

c9

17:2

c11 c13

16:3

n-6 16:2 n-4 +NMI?

16:1 c9 x1 t6 t9 c7 c11 -12

n-4

n-3

n-3

16:4

n-1

15:1

14:1

29.13

31.34

33.56

35.78

38.00

40.21

42.44

44.66

46.88

49.00

51.31

53.54

1D, 200 m x 0.25 mm SLB-IL111, minutes

Figure 4. Partial GC×GC separation from C14 to C20 of the FAME prepared from a menhaden oil sample. Instrument configuration as shown in Figure 1. Y1, pristanic acid methyl ester; y2, phytanic acid methyl ester; x1, FAME with 17 carbons and one unsaturation; x2, FAME with 19 carbons and one unsaturation NMI, nonmethylene interrupted.

ACS Paragon Plus Environment

55.75

Analytical Chemistry

3.50 23:1

23:0 3.00

22:4 n-6 22:1

22:0

2.50 2.00 1.50 0.50 1.00 2D, 2.5 m x 0.10 mm SLB-IL111, seconds

22:5 n-3

22:5 22:4 n-6 n-3

22:6 n-3

21:5 n-3 21:1

21:0

20:3 20:1

20:2

n-6

20:0 19:0

16:0

20:4 n-6

20:4 n-3

20:5 n-3

19:1

18:1 18:0

n-3

18:2

17:1 16:3 16:1 16:2

18:3 n-6

18:2 18:4 n-3 conj. n-3 n-1

16:4

24:0

24:1

14:0

0.00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 23

18.97

30.64

42.31

53.98

65.65

77.32

88.99

100.66

112.33

124.00

135.67

147.34

159.01

170.68

1D, 200 m x 0.25 mm SLB-IL111, minutes

Figure 5. GC×GC separation of the FAME prepared from human colon adenocarcinoma cells HT-29 incubated with conjugated fatty acids. Instrument configuration as shown in Figure 1.

ACS Paragon Plus Environment

Page 23 of 23

2.00 y2

y1

1.75

24:0 16:0

1.50 14:0

1.00 0.75 0.50 1.25 2D, 2.5 m x 0.10 mm SLB-IL111, seconds

i-16:0 ai15:0 i- 15:0 15:0

i17:0

ai17:0

i- 18:0 18:0

22:0

17:0

20:0

19:0

21:0

c15 c5

i14:0

t6 c7 c7

14:1

16:1

t6 c9

c9

c915:1

c11

x1

c9

13:0

c11

c5

17:1

c7 c9 18:1 c11 19:1

c11 c13

c13 n-6 +NMI?

16:2

n-4

c9 c7

20:1

24:1

22:1

c17

c13

c11 c13

c15

c15

c13 x2

NMI

c14/c15 c5, c11 c8, c11

17:2

NMI

20:2

n-4

n-6 n-4 n-6 n-6

18:3 n-4

n-4 16:3 n-3 n-3

22:2

n-6

n-6

18:2

12:0

0.25

16:4 n-1

n-3 18:4 n-6 n-3 n-1

20:3

n-4 n3 20:4 n-6 20:4 n-3

22:4 n-6 22:5 n-6 22:4 22:5 n-3 n-3 20:5 n-3 21:5

22:6 n-3

18:5 n-3

0.00

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

9.00

15.67

22.33

29.00

35.67

42.33

49.00

55.67

62.33

59.00

75.67

82.33

89.00

1D, 100 m x 0.25 mm SLB-IL111, minutes

Figure 6. GC×GC separation of the FAME prepared from a menhaden fish oil sample. Instrument configuration as shown in Figure 1, with modifications: 1D column 100 m x 0. 25 mm SLB-IL111, and elution temperature program. Y1, pristanic acid methyl ester; y2, phytanic acid methyl ester; x1, FAME with 17 carbons and one unsaturation; x2, FAME with 19 carbons and one unsaturation; NMI, non-methylene interrupted.

ACS Paragon Plus Environment