J Am Oil Chem Soc (2015) 92:783–790

DOI 10.1007/s11746-015-2635-2

ORIGINAL PAPER

Comprehensive Two‑Dimensional Gas Chromatography–Mass

Spectrometry Analysis of Different Types of Vegetable Oils
Olga Vyviurska1 · Nikoleta Jánošková1 · Tibor Jakubík1 · Ivan Špánik1 

Received: 21 July 2014 / Revised: 21 December 2014 / Accepted: 12 March 2015 / Published online: 22 April 2015
© AOCS 2015

Abstract Comprehensive two-dimensional gas chro- acid (ω6 fatty acid) and γ-linolenic acid (ω3 fatty acid)
matography coupled with time-of-flight mass spectrom- are not synthetized in the human body, thus must be sup-
etry (GC×GC-TOFMS) was applied for detailed charac- plied from other sources. Since a main task of fatty acids
terization of fatty acid profile of 8 vegetable oils. Due to is to maintain energy balance, side effects connected with
enhanced selectivity and sensitivity characteristics, the risk of possible obesity or cardiovascular disease could fre-
GC×GC method yielded more reliable quantification quently occur.
results compared to one dimensional gas chromatography, It is generally accepted that a ratio of saturated and
especially for medium-chain fatty acids and odd-carbon unsaturated fatty acids determines the levels of the “bad
number fatty acids, which are present only at trace level. cholesterol” (low density lipoproteins) and “good choles-
All problematic positional counterparts of unsaturated terol” (high density lipoproteins) in the blood. Overall,
fatty acids (e.g. C21:0–C20:3 ω6, C20:3ω3–C20:4 ω6 and saturated fatty acids (SFA) increase cholesterol levels in a
C20:5 ω3–C22:0), which commonly coeluted in the case of value proportional to the length of their carbon chains [1].
1D gas chromatography, were baseline resolved. The spe- Various conclusions were reported for cholesterolemic
cific compounds were found for particular vegetable oils, properties of stearic acid, since some studies [2] have indi-
such as γ-linolenic acid for hempseed oil, heneicosylic acid cated that this fatty acid is “neutral” concerning cholesterol
and tricosylic acid for olive pomace oil, and nervonic acid level change, but simultaneously it could affect thrombotic
for mustard oil. tendency or arrhythmia. On the other hand, it should be
accounted that a minimum amount of SFA is essential for
Keywords  Vegetable oils · Comprehensive two- healthy nutrition, due to their importance for membrane
dimensional gas chromatography function [3]. Monounsaturated fatty acids (MUFA) have
a positive influence on blood lipids, blood pressure and
insulin sensitivity and diminish risk of obesity [4]. In the
Introduction case of polyunsaturated fatty acids (PUFA), adverse effects
could take place when more than 10 % of whole organism
Nowadays, a healthy lifestyle based on an appropriate energy is provided with PUFA. ω6 and ω3 fatty acids from
nutrition is strongly promoted by the World Health Organi- vegetable oils are used as precursors of long chain PUFA
zation. Consequently, edible oils are recommended to be a (e.g. arachidonic acid, dihomo-γ-linolenic acid, eicosapen-
major source of fatty acids, since some of them e.g. linoleic taenoic acid and docosahexaenoic acid), which are essential
for normal brain development and synthesis of high bioac-
tive hormone-like compounds e.g. eicosanoids [5]. At the
* Ivan Špánik same time, an appropriate ratio of ω6/ω3 should be kept in
ivan.spanik@stuba.sk a diet, because of the opposite effects of ω6 and ω3 fatty
1 acids on inflammatory processes and their competition for
Institute of Analytical Chemistry, Faculty of Chemical
Technology, Slovak University of Technology, Radlinského the same enzymes. The Western diet is described with high
9, 81237 Bratislava, Slovak Republic ω6/ω3 ratio that could reach 15:1 value, because of high

13

784 J Am Oil Chem Soc (2015) 92:783–790

intake of animal fats and corn oil. Whereas the Mediterra- was received for ω9 and ω7 isomers on BP-20 column,
nean diet rich in fruits, vegetables, whole-grains, and with whereas C18:4 ω3 and C20:1 ω11 were overlapped on
ω6/ω3 ratio 2:1 and 3:1, is considered as more optimal. BPX-70 column. The non-polar columns have high ther-
An important problem encountered for vegetable oils is mal stability and chemical inertness, whereas separa-
reliable authentication, especially in the case of adultera- tion power for unsaturated fatty acids is limited. Kubinec
tion of virgin olive oil with cheaper edible oils. In general, et al. [15] tested methyl silicone OV-1 stationary phase
a specific content of the particular fatty acids or sterols (100 m × 0.25 mm × 0.25 μm) for identification of the
(e.g. brassicasterol, stigmastenol, avenasterol) and their methyl branched saturated FAME in tongue coating. The
dehydration products, such as 3,5-stigmadiene, 3,5-camp- separation of fatty acids with methyl group located near the
estadiene, 3,5,22-stigmastatriene, are commonly recom- middle of the carbon chain was problematic, and it required
mended for assessment of oil identity. Moreover alternative an additional deconvolution procedure. Another non-polar
methods based on a peculiar ratio of C18:3 at the 2-posi- column Zebron ZB-1 (15 m × 0.25 mm × 0.1 μm) with
tion on glycerol molecules to overall concentration of this polydimethylsiloxane stationary phase was applied for
acid, and composition of tocopherols or waxes have been analysis of biological samples. In this case, the positional
developed [6]. The most common method for analysis of isomers of C18:1 ω9 and C18:1 ω12 were not separated to
fatty acid content is gas chromatography, which develop- baseline [16].
ment has led to expanding of sample types. There are two Incomplete separation of geometrical and positional iso-
factors which influence on selection of stationary phases, mers of unsaturated fatty acids and limited quantification
its thermal stability and an appropriate resolution of fatty of odd-carbon number fatty acids are the major obstacles
acids [7, 8]. In general, high polar stationary phases espe- encountered in analysis of FA by classical gas chromatog-
cially those modified with cyano groups are preferable for raphy. Comprehensive two-dimensional gas chromatogra-
FA analysis, because of their high resolution of unsatu- phy (GC×GC) as a novel type of gas chromatography is
rated fatty acids. Harynuk et al. [9] compared polysilphe- characterized with enhanced separation power and selec-
nylene-siloxane stationary phases with 60–90 % content of tivity [17]. Due to application of two separation columns
cyanopropyl groups (30 m × 0.25 mm × 0.25 μm BPX60- with different polarity of stationary phases, a full separa-
BPX90 columns). The best results without important peak tion of structural and geometrical isomers could be easily
overlap were obtained on BPX-60 column, but C20:5 ω3 achieved. The selection of columns combination is tradi-
and C22:0 fatty acids were not baseline separated. HP-88 tionally based on orthogonal approach, where non-polar
(100 m × 0.25 mm I.D. × 0.2 μm) column with biscyano- columns (e.g. BPX5, HP-5MS, DB5-MS) are preferred in
propyl polysiloxane stationary phase was used for estab- the 1st dimension and polar columns (e.g. BP20, BPX-70,
lishment of fatty acid profile in human blood [10]. An Supelcowax-10) are applied in the 2nd dimension. The first
excellent separation was observed for cis and trans C18:1 dimension requires to use capillary columns similar to one
isomers, but positional isomers of C18:1 were not fully dimensional chromatography (15–60 m × 0.25–0.53 mm
resolved. In order to separate three problematic conjugated ID  × 0.25–1 µm df). Since the separation in the second
linoleic acid isomers (trans-ω7, cis-ω9; cis-ω9, trans-ω11; dimension should be rapid and effective, short and nar-
trans-ω8, cis-ω10), CP-Sil column (100 m × 0.25 mm row bore columns (0.5–1.5 m × 0.1–0.25 mm ID × 0.1–
i.d. × 0.2 μm) coated with cyanopropyl polysiloxane sta- 0.25 µm df) are preferred. Such column setup was success-
tionary phase was tested [11], however obtained results fully used for establishment of FA profile in oils [18–20],
demanded software deconvolution of isomers. At the same bacteria [21], human plasma [22], milk [23], insulin secret-
time, this column was sufficient for monitoring of fatty ing cells [24]. The reversed combination (polar × non-
acid variation of ewes’ colostrums [12]. Low thermal sta- polar) of stationary phases has been rarely applied for
bility of polar columns requires a slow temperature pro- FA analysis. Normal and reversed orthogonal systems of
gram for analysis of samples with a great variety of fatty columns were compared in a few studies [25–27], but it
acids. Whereas, stationary phases based on polyethylene was difficult to arrive to a single conclusion about a pri-
glycol or stationary phases with lower content of cyano- ority of the some approach. High data acquisition rate and
propyl groups could be used at higher temperature ranges. a small internal volume of detector in GC×GC provide
Milk fat composition was analyzed using polyethylene gly- an increased sensitivity of measurements. FID and μECD
col stationary phase (30 m × 0.32 mm × 0.25 μm Super- do not assert structural information, in contrast to a mass
wax 10) [13], and in this case low resolution of C18:1 and spectrometer. TOFMS with rate 50–500 mass spectra per
C18:2 was observed. Wasta et al. [14] compared polyeth- second allows to record full mass spectrum, followed by
ylene glycol column (50 m × 0.22 mm × 0.25 μm BP-20) spectral deconvolution of overlapping peaks. Furthermore,
and cyanopropyl column (60 m × 0.25 mm × 0.25 μm identification procedure is simplified due to formation of
BPX-70) for analysis of ω3 products. The poor separation group-type patterns of homologous series of compounds.

13
J Am Oil Chem Soc (2015) 92:783–790 785

Although a considerable amount of research has been characterize fatty acid methyl esters (FAME). Helium was
devoted to analysis of vegetable oils, most research was used as a carrier gas at constant flow mode (1.0 ml/min).
performed by 1D gas chromatography, where coelution The column setup included 30 m × 0.25 mm I.D. coated
of FAME often occurs and consequently accurate quanti- with 0.25 µm film thickness DB-FFAP column (Agilent
fication of odd carbon-number fatty acids and unsaturated J&W Column, Agilent Technologies, Palo Alto, CA, USA)
fatty acid counterparts was not always reached. The major in 1st dimension, and 1.5 m × 0.1 mm I.D. coated with
purpose of this work was to define completely FA profile 0.1 µm film thickness BPX-50 column in the 2nd dimen-
of the most common vegetable oils, such as pomace and sion (SGE Analytical Science, Victoria, Australia). The
extra virgin olive oils, rapeseed oil, sunflower oil, sesame oven temperature program was as follows: isothermal at
oil, mustard oil, hempseed oil, and a commercial mixture 40 °C for 2 min, 30 °C/min to 160 °C, 2 °C/min to 230 °C,
sunflower and extra virgin olive oil with comprehensive and held at 230 °C for 20 min.
two-dimensional gas chromatography. The temperature of secondary oven was set to 10 °C
offset compare to primary oven temperature. The follow-
ing modulation parameters were used: a 20 °C modula-
Materials and Methods tion temperature offset, 8 s modulation period with 1.5 s
hot pulse duration. The mass spectrometer transfer line
Samples and Chemicals was heated to 240 °C, the ion source temperature was
230 °C, the electron ionization energy 70 eV. Mass spec-
Rapeseed oil, sunflower oil, sesame oil, mustard oil, hemp- tra within mass range m/z from 45 to 650 and acquisition
seed oil, olive pomace and extra virgin olive oils and a com- rate of 100 spectra/s were recorded. Data processing was
mercial mixture of refined sunflower oil (60 %) and extra performed with LECO ChromaTof software (version 4.13),
virgin olive oil (40 %) were produced in Hungary, Czech and spectra were deconvoluted at a signal to noise (S/N)
Republic, Spain, France and Mexico, and purchased from ratio >200. In addition, the obtained chromatograms were
a local shop in 2011. A FAME standard mixture (Supelco evaluated using the US National Institutes of Standards
37 Component FAME mix, 47885-U) was purchased from and Technology (NIST08) mass library and Agilent FAME
Merck (Darmstadt, Germany). library. FAME standard mixture was used for quantifica-
tion of fatty acid in the samples. Reliability of the proce-
Sample Pretreatment dure was checked by analysis of vegetable oil reference
material used for the FAPAS proficiency test (code 14109).
Fatty acid methyl esters were prepared by transesterifica- The results were compared to results obtained by tests per-
tion of samples based on the procedure described in ISO formed in accredited laboratory.
55080 as follows: 0.2 g of the sample was dissolved in 5 ml
of hexane (Merck, Darmstadt, Germany), and then 1 ml of
KOH (Mikrochem, Pezinok, Slovakia) solution (2 mol/l) in Results and Discussion
methanol (Merck, Darmstadt, Germany) was added. The
obtained mixture was agitated and heated in water bath at Polar × medium-polar columns combination was used for
60 °C for 30 s, with the next addition of 2 ml of HCl (Mik- FA analysis in vegetable oils. Features of the ordered struc-
rochem, Pezinok, Slovakia) solution (1 mol/l). After agita- ture of chromatogram regarding carbon number, number
tion and standing until complete separation of the phases, and position of double bonds were found for the FA profile
50  μl of hexane solution was put into a volumetric flask (Fig. 1). Overall, a retention factor of analyte is defined by
and diluted to 1 ml with hexane. One microliter of the final stationary phase characteristics, as well as analyte param-
mixture was injected directly into the CIS injector in split eters (e.g., volatility, solute size, solute bipolarity/polar-
mode (1:50). izability, hydrogen bond acidity of the solute, excess of
polarizability of the solute) [28, 29]. As can be seen from
Gas Chromatography–Mass Spectrometry Analysis Fig.  1, fatty acids with a higher number of double bonds
being more polar are more strongly retained on a polar col-
LECO Pegasus 4D (LECO Corporation, St. Joseph, MI, umn in the first dimension, whereas a reverse tendency is
USA) consisting of an Agilent 6890 gas chromatograph observed for the separation in the second medium-polar
(Agilent, Agilent J&W Column, Agilent Technologies, column. Homologous unsaturated fatty acids with the
Palo Alto, CA, USA) and Gerstel MPS 2 autosampler same position of double bonds show up bands, and the
(Gerstel, Mülheim, Germany), equipped with a CIS injec- compounds with the lowest ω number elute later. Further-
tor, dual-stage thermal modulator and secondary oven con- more, the second column was effective for the separation
nected to a Time-Of-Flight Mass Spectrometer was used to of pairs C21:0–C20:3 ω6 and C20:3ω3–C20:4 ω6 which

13

786 J Am Oil Chem Soc (2015) 92:783–790

Fig. 1  GC×GC-TOFMS chro-
matogram of a FAME standard
mixture

are coeluted in the first dimension of the column setup. Due to the tendency of unsaturated fatty acids to be eas-
In comparison to cyanopropyl stationary phases, which ily oxidized, the higher content of monounsaturated fatty
mainly recommended for 1D gas chromatographic analysis acids is more preferable for long-term storage and frying.
of FAME, a problematic pair C20:5 ω3–C22:0 was com- Moreover, a commercial production of vegetable oils with
pletely resolved. an enhanced concentration of oleic acid has been developed
As it is shown in Table 1, olive pomace oil and extra vir- over the last decade [4]. Oleic acid was dominant among
gin olive oil, pumpkin and sesame oils are enriched with fatty acids (40.35–62.35 %) for two types of olive oils and
saturated fatty acids, whereas sunflower, rapeseed, mustard rapeseed oil (Table 1). The maximum content of palmit-
and hempseed oils are characterized by a higher content oleic acid was observed in olive oils (1.63–1.20 %). Eicose-
of unsaturated fatty acids (USFA/SFA 8.11–13.89). Fig- noic acid was detected at the minor level (less than 1 %) in
ure  2 demonstrates fatty acid profiles of the studied veg- all studied samples, whereas highest values were 1.27 and
etable oils. Overall, medium chain fatty acids were found 2.91 % for rapeseed and mustard oil, respectively. How-
only in trace amounts in some vegetable oils, e.g., caprylic ever, higher quantities of eicosenoic acid were reported for
acid (0.01–0.03 %) in sunflower oil, olive pomace oil, a mustard oil (6.89 %) and rapeseed oil (9.3 %) in other stud-
mixture of sunflower and extra virgin olive oils, pumpkin ies [32, 33]. The higher amount of erucic acid (22:1 ω9)
and sesame oils, capric acid (0.02 %) in rapeseed and mus- could make edible oils undesirable for human consumption
tard oils, and lauric acid (0.01 %) in rapeseed, mustard and [33, 34]. In our case, erucic acid was identified in rapeseed
pumpkin oils. Similarly a minor content (nearly of 1 %) of oil (0.32 %) and mustard oil (2.73 %). At the same time, a
myristic, margaric, arachidic, behenic acids were found in long unsaturated nervonic acid was present only in mustard
all studied samples. In contrast, a higher value (1.94 %.) of oil (1.77 %).
myristic acid was reported for pumpkin oil by Siang et al. The highest content of linoleic acid (18:2) was detected
[30], which could be partially related to application of hol- in sunflower oil (66.56 %), pumpkin oil (55.59 %) and
low fiber microextraction technique for sample preparation. hempseed oil (56.52 %), and the obtained results are in
There was no significant difference observed in the con- good correlation with those published in the literature [35,
centration of stearic acid (1.87–6.7 %) for all the studied 36]. Meanwhile, oleic acid and linoleic acid, as major com-
vegetable oils. An increased content of palmitic acid was ponents of fatty acids profile, are present in nearly equal
determined in extra virgin olive oil (17.85 %), whereas its amounts in mustard oil and sesame oil. Rapeseed oil, olive
content varied in the range from 3.77 to 12.48 % in other pomace oil and extra virgin olive oils were characterized by
edible oils. A slightly higher result of 15.97 % was pre- smaller content of linoleic acid, 29.59, 16.80 and 13.43 %
viously reported for palmitic acid in pumpkin oil [31]. respectively. Another long-chain ω6 fatty acid, γ-linolenic
Among saturated fatty acids with odd number of carbon acid (22:3), which improves nutrition values of the oil [37],
atoms, small amounts of pentadecylic acid (0.01–0.02 %) was presented only in hempseed oil. The special attention
were present in all samples except sunflower oil, olive should be given to α-linolenic acid, the most common ω3
pomace oil, extra virgin olive oil and sesame oil, whereas fatty acid, which was found for hempseed oil (18.77 %),
heneicosylic acid (0.02 %) and tricosylic acid (0.03 %) mustard oil (10. 43 %) and rapeseed oil (8.40 %). The
were identified only in olive pomace oil. results obtained correlate with data presented by other

13
J Am Oil Chem Soc (2015) 92:783–790 787

Table 1  The fatty acids composition of different vegetable oils

CAS number Name Fatty acid composition (%)
Sunflower Olive pomace Extra-virgin Mixturea Rapeseed Mustard Pump- Hempseed
oil oil olive oil oil oil kin oil oil

124-07-2 Caprylic acid (8:0) 0.01 0.03 0.01 0.02

334-48-5 Capric acid (10:0) 0.02 0.02
143-07-7 Lauric acid (12:0) 0.01 0.01 0.01
544-63-8 Myristic acid (14:0) 0.06 0.02 0.02 0.06 0.05 0.06 0.10 0.03
1002-84-2 Pentadecylic acid (15:0) 0.01 0.02 0.02 0.01 0.02
57-10-3 Palmitic acid (16:0) 5.50 12.48 17.85 8.16 4.72 3.77 9.02 6.73
2091-29-4 Palmitoleic acid (16:1, ω7) 1.20 1.63 0.31 0.24 0.24 0.11 0.13
506-12-7 Margaric acid (17:0) 0.03 0.11 0.09 0.05 0.03 0.03 0.05 0.05
29743-97-3 10-Heptadecenoic acid (17:1 0.18 0.18
ω7)
57-11-4 Stearic acid (18:0) 3.03 4.40 5.79 4.91 1.87 2.34 5.62 2.99
112-80-1 Oleic acid (18:1 ω9) 23.45 62.35 58.62 38.70 52.24 37.65 28.14 12.56
60-33-3 Linoleic acid (18:2 ω6) 66.56 16.80 13.43 45.95 29.59 36.55 55.59 56.52
506-26-3 γ-Linolenic acid (18:3 ω6) 0.59
463-40-1 α-Linolenic acid (18:3 ω3) 0.93 1.04 0.22 8.40 10.43 0.24 18.77
506-30-9 Arachidic acid (20:0) 0.22 0.68 0.72 0.37 0.61 0.65 0.38 0.78
5561-99-9 Eicosenoic acid (20:1 ω9) 0.14 0.42 0.43 0.22 1.27 2.91 0.14 0.38
2091-39-6 Eicosadienoic acid (20:2 ω3) 0.24 0.06
2363-71-5 Heneicosylic acid (21:0) 0.02
112-85-6 Behenic acid (22:0) 0.73 0.28 0.20 0.78 0.39 0.44 0.27
112-86-7 Erucic acid (22:1 ω9) 0.32 2.73
2433-96-7 Tricosylic acid (23:0) 0.03
557-59-5 Lignoceric acid (24:0) 0.28 0.11 0.26 0.23 0.28 0.18 0.11
506-37-6 Nervonic acid (24:1 ω9) 1.77
Saturated acids 9.86 18.15 24.67 14.62 7.96 7.19 15.81 10.98
Monounsaturated acids 23.59 64.16 60.85 39.22 54.08 45.30 28.38 13.08
Polyunsaturated acids 66.56 17.73 14.48 46.16 37.98 54.53 55.83 75.95
Unsaturated acids 90.15 81.88 75.33 85.39 92.06 99.84 84.21 89.02
USFA/SFA ratio 9.14 4.51 3.05 5.84 11.56 13.89 5.33 8.11
ω6/ω3 ratio 18.06 12.86 213.40 3.52 3.53 233.40 3.05
MUFA/PUFA 0.35 3.62 4.20 0.85 1.42 0.83 0.51 0.17
a
  Mixture of refined sunflower oil (80 %) and extra virgin olive oil (20 %)

scientific groups [32, 33, 37, 38]. Mustard oil and hemp- technology of the oils. Extra virgin oil was characterized a
seed oil also include minor amounts (0.06–0.24 %) of eico- by higher content of hexadecanoic acid and octadecanoic
sadienoic acid (20:2 ω3), and sunflower oil does not con- acid compared to olive pomace oil, while a vast variety
tain ω3 fatty acids at all. In spite of a low content of ω3 of fatty acids was detected in olive pomace oil. The cal-
fatty acids in most vegetable oils, a ratio of ω6/ω3 ~4:1 is culations showed a smaller ω6/ω3 ratio for olive pomace
recommended for healthy nutrition [39]. According to the oil. In general, extra virgin olive oil is more strongly pre-
results obtained, it appears that rapeseed oil, mustard oil ferred by customers because of its solvent-free production
and hempseed oil with increased amounts of ω3 fatty acids, technology.
should be incorporated into a well-balanced diet. The commercial mixture of refined sunflower oil (60 %)
Both olive pomace oil and extra virgin olive oil demon- and extra virgin olive oil (40 %) with an improved compo-
strated a similar fatty acid profile. The difference in satu- sition for domestic utilization was analyzed. As was men-
rated and unsaturated fatty acids contents were estimated tioned above, a high polyunsaturated fatty acids content
in the range 3.25–6.52 %, and resulted from the production of sunflower oil is in some cases problematic, and adding

13

788 J Am Oil Chem Soc (2015) 92:783–790

Fig. 2  The fatty acid profile of the studied vegetable oils [mixture*—mixture of refined sunflower oil (80 %) and extra virgin olive oil (20 %)]

13
J Am Oil Chem Soc (2015) 92:783–790 789

extra virgin olive oil with domination of monounsaturated α-linolenic acid. The higher content and variety of poly-
fatty acids could balanced the fatty acid profile of sunflower unsaturated fatty acid were investigated for hempseed oil.
oil. As a result, the received mixture was characterized by Among other fatty acids γ-linolenic acid was specific for
nearly equal amounts of oleic and linoleic acids. The con- hempseed oil, heneicosylic acid and tricosylic acid for
tent of saturated fatty acids was also increased, whereas the olive pomace oil, and nervonic acid for mustard oil. Since
dominance of ω6 upon ω3 fatty acids became more signifi- a ratio of ω6/ω3 ~4:1 is strongly recommended for healthy
cant. This fact has a positive influence on application of the nutrition it appears that rapeseed oil, mustard oil and hemp-
mixture for high temperature cooking. seed oil should be included in the daily diet. The results
The analysis of the reference material has proven that obtained by analysis of reference materials were accept-
the developed analytical procedure is applicable for deter- able for the application of the method for FAME analysis
mination of FAME in vegetable oils. The reliability of the in vegetable oils.
method, expressed as absolute value of:
Acknowledgments  This research was supported by the Slo-
100 × (reference value − obtained value)/reference value, vak Research and Development Agency under the contract No.
APVV-0797-11.
varies within a range 94–97 % for saturated FAME,
95–96 % for monounsaturated and polyunsaturated FAME. Conflict of interest None.
The “obtained value” was calculated as an average value
from three independent measurements, while standard Compliance with ethical requirements  This article does not con-
deviation from those three measurements did not exceed tain any studies with human or animal subjects.
15 % (a relative value). The major source of error is the
sample treatment procedure, since repeatability of meas-
urements obtained from 8 injections of the same solution References
was below 4 % (a relative value) for each FAME present
in the mixture. The results obtained by GC×GC-TOF–MS 1. Denke MA (2006) Dietary fats, fatty acids, and their effects on
were also compared to results obtained by GC–MS analysis lipoproteins. Curr Atheroscler Rep 8:466–471
of the same material in an accredited laboratory (State Vet- 2. Rzybylski R (1999) Development and processing of vegetable
oils for human nutrition. AOCS Press, Champaign
erinary Institute, Bratislava, Slovakia). The interlaboratory 3. Wilke MS, Clandinin MT (2005) Influence of dietary saturated
comparison showed good agreement with the standardized fatty acids on the regulation of plasma cholesterol concentration.
method since observed deviations did not exceed 2 %. Lipids 40:1207–1213
4. Gillingham LG, Harris-Janz S, Jones PJH (2011) Dietary
monounsaturated fatty acids and protective against meta-
bolic syndrome and cardiovascular disease risk factors. Lipids
Conclusion 46:209–228
5. Richardson AJ (2004) Long-chain polyunsaturated fatty acids
The comparison of fatty acid profile of 8 types of vegeta- in childhood developmental and psychiatric disorders. Lipids
39:1216–1222
ble oils was performed by comprehensive two-dimensional 6. Aparicio R, Aparicio-Ruíz R (2000) Authentification of veg-
gas chromatography. The special attention was paid on etable oils by chromatographic techniques. J Chromatogr A
the content of medium-chain fatty acids and odd-carbon 881:93–104
number fatty acids, since little is known about these minor 7. Eder K (1995) Gas chromatographic analysis of fatty acid
methyl ester. J Chromatogr B 671:113–131
constituents of vegetable oils. In comparison with 1D gas 8. De Zeeuw J (2002) Developments in stationary phase technol-
chromatography, complete separation of positional iso- ogy for chromatography. TrAC Trend Anal Chem 21:594–607
mers of unsaturated fatty acids was achieved, especially 9. Harynuk J, Wynne PM, Marriott PJ (2006) Evaluation of new
for such problematic pairs of fatty acids as C21:0–C20:3 stationary phases for the separation of fatty acid methyl esters.
Chromatographia 63:61–66
ω6, C20:3ω3–C20:4 ω6 and C20:5 ω3–C22:0, which per- 10. Bicalho B, David F, Rumplel K, Kindt E, Sandra P (2008) Cre-
mitted precise estimation of each of these fatty acids. As ating a fatty acid methyl ester database for lipid profiling in a
a result, monounsaturated acids were dominant in olive single drop of human blood using high resolution capillary
oils and rapeseed oil, whereas polyunsaturated fatty acids gas chromatography and mass spectrometry. J Chromatogr A
1211:120–128
were the major components of the other oils. The content 11. Blaško J, Kubineca R, Ostrovský I, Pavlíková E, Krupčík J,
of monounsaturated and polyunsaturated acids were nearly Soják L (2009) Chemometric deconvolution of gas chromato-
equal for sesame oil. The linoleic acid as a main source graphic unresolved conjugated linoleic acid isomers triplet in
of ω 6 fatty acid was the dominant fatty acid in sunflower milk samples. J Chromatogr A 1216:2757–2761
12. Górová R, Pavlíková E, Blaško J, Meľuchová B, Kubinec R,
oil (66.56 %), pumpkin oil (55.59 %) and hempseed oil Margetín M, Soják L (2011) Temporal variations in fatty acid
(56.52 %). In all studied samples, the content of ω3 fatty composition of individual ewes during first colostrum day. Small
acids was small and determined mainly by presence of Ruminant Res 95:104–112

13

790 J Am Oil Chem Soc (2015) 92:783–790

13. Contarini G, Povolo M, Pelizzola V, Monti L, Lercker G (2013) oils using comprehensive two-dimensional gas chromatography.
Interlaboratory evaluation of milk fatty acid composition by J Chromatogr A 1019:187–196
using different GC operation conditions. J Food Compos Anal 26. Manzano P, Arnáiz E, Diego JC, Toribio L, García-Viguera C,
32:131–140 Bernal JL, Berna J (2011) Comprehensive two-dimensional
14. Wasta Z, Mjøs SA (2013) A database of chromatographic prop- gas chromatography with capillary flow modulation to separate
erties and mass spectra of fatty acid methyl esters from omega-3 FAME isomers. J Chromatogr A 1218:4952–4959
products. J Chromatogr A 1299:94–102 27. Adahchour M, Jover E, Beens J, Vreuls RJJ, Brinkman UATh
15. Kubinec R, Blaško J, Górová R, Addová G, Ostrovský I, Amann (2005) Twin comprehensive two-dimensional gas chromato-
A, Soják L (2011) Equivalent chain lengths of all C4–C23 satu- graphic system: concept and applications. J Chromatogr A
rated monomethyl branched fatty acid methyl esters on methyl- 1086:128–134
silicone OV-1 stationary phase. J Chromatogr A 1218:1767–1774 28. Abraham MH, Ibrahim A, Zissimos AM (2004) Determination of
16. Quehenbergera O, Aaron AM, Denni EA (2011) High sensitivity sets of solute descriptors from chromatographic measurements. J
quantitative lipidomics analysis of fatty acids in biological sam- Chromatogr A 1037:29–47
ples by gas chromatography–mass spectrometry. BBA Mol Cell 29. Poole CF, Poole SK (2002) Column selectivity from the per-
Biol L 1181:648–656 spective of the salvation parameter model. J Chromatogr A
17. Tranchida PQ, Donato P, Cacciola F, Beccaria M, Dugo P, Mon- 965:263–299
dello L (2013) Potential of comprehensive chromatography in 30. Siang GH, Makahleh A, Saad B, Lim BP (2010) Hollow fiber
food analysis. TrAC Trend Anal Chem 52:186–205 liquid-phase microextraction coupled with gas chromatography-
18. De Geus HJ, Aidos I, de Boer J, Luten JB, Brinkman UATh flame ionization detection for the profiling of fatty acids in veg-
(2001) Characterization of fatty acids in biological oil samples etable oils. J Chromatogr A 1217:8073–8078
using comprehensive multidimensional gas chromatography. J 31. Rezig L, Chouaibi M, Msaasa K, Hamdi S (2012) Chemical
Chromatogr A 910:95–103 composition and profile characterization of pumpkin (Cucurbita
19. Western RJ, Lau SSG, Mariott PJ, Nicholos PD (2002) Posi- maxima) seed oil. Ind Crop Prod 37:82–87
tional and geometric isomer separation of FAME by comprehen- 32. Abul-Fadl MM, El-Badry N, Ammar MS (2011) Nutritional and
sive 2D GC. Lipids 37:715–722 chemical evaluation for two different varieties of mustard seeds.
20. Mondello L, Tranchida PQ, Dugo P, Dugo G (2006) Rapid,
WASJ 15:1225–1233
micro-scale preparation and very fast gas chromatographic sep- 33. Ramos MJ, Fernández CM, Casas A, Rodríguez L, Pérez Á

aration of cod liver oil fatty acid methyl esters. J Pharmaceut (2009) Influence of fatty acid composition of raw materials on
Biomed 41:1566–1570 biodiesel properties. Bioresour Technol 100:261–268
21. Gu Q, David F, Lynen F, Rumpel K, Xu G, Vos PD, Sandra P 34. Wendlinger C, Hammann S, Vetter W (2014) Various concen-
(2010) Analysis of bacterial fatty acids by flow modulated com- tration of erucic acid in mustard oil and mustard. Food Chem
prehensive two-dimensional gas chromatography with paral- 153:393–397
lel flame ionization detector/mass spectrometry. J Chromatogr 35. Jiao J, Li Z-G, Gai Q-Y, Li X-J, Wei F-Y, Fu Y-J, Ma W (2014)
1217:4448–4453 Microwave-assisted aqueous enzymatic extraction of oil from
22. Tranchida PQ, Costa R, Donato P, Sciarrone D, Ragonese C, pumpkin seeds and evaluation of its physicochemical properties,
Dugo P, Dugo G, Mondello L (2008) Acquisition of deeper fatty acids composition and antioxidant activities. Food Chem
knowledge on the human plasma fatty acid profile exploiting 147:17–24
comprehensive 2-D GC. J Sep Sci 31:3347–3351 36. Bansal G, Zhou W, Tan T-W, Neo F-L, Lo H-L (2009) Analy-
23. Hyötyläinen T, Kallio M, Lehtonen M, Lintonen S, Peräjoki P, sis of trans fatty acids in deep frying oils by three different
Jussila M, Riekkola ML (2004) Comprehensive two-dimensional approaches. Food Chem 116:535–541
gas chromatography in the analysis of dietary fatty acids. J Sep 37. Lin J-Y, Zeng Q-X, An Q, Zeng Q-Z, Jian L-X, Zhu Z-W (2012)
Sci 27:459–467 Ultrasonic extraction of hempseed oil. J Food Process Eng
24. Payeur AL, Lorenz MA, Kennedy RT (2012) Analysis of fatty 35:76–90
acids composition in insulin secreting cells by comprehensive 38. Leizer C, Ribnicky D, Poulev A, Dushenkov S, Raskin I (2000)
two-dimensional gas chromatography time-of-flight mass spec- The composition of hemp seed oil and its potential as an impor-
trometry. J Chromatogr B 893–894:187–192 tant source of nutrition. J Nutraceut Funct Med Foods 2:35–52
25. Mondello L, Casilli A, Tranchida PQ, Dugo P, Dugo G (2003) 39. Dunford M, Doyle JA (2012) Nutrition for sport and exercise.
Detailed analysis and group-type separation of natural fats and Belmont, Wadsworth

13