2. SURVEY ON GAS CHROMATOGRAPHY–MASS
2.2. Gas Chromatography–Mass Spectrometry
Essential oils are complex mixtures of flavor and fragrance compounds originat- ing from plants. The analysis of essential oils, which consist mainly of mixtures of monoterpene and sesquiterpene hydrocarbons and their oxygenated deriva- tives, often requires a prefractionation before the chromatographic analysis. Oth- erwise, identification is difficult even with the use of GC–MS, since mass spectra of the components of the same class present very similar patterns. Consequently, in order to unequivocally identify each essential oil constituent, it is necessary to obtain the spectrum of an extremely pure substance. Among the on-line meth- ods of fractionation, LC–GC allows the separation and analysis of complex mix- tures in fully automated mode. Applications of this technique to citrus oil were recently reported [23,24]. The LC–GC–MS technique was proposed for the anal- ysis of monoterpenes, sesquiterpenes, and their oxygenated derivatives in berga- mot essential oil [23], and mono- and sesquiterpenes in the volatile fraction of eight essential oils: bergamot, lemon, mandarin, sweet orange, bitter orange, grapefruit, clementine, and Mexican lime [24]. After on-line preseparation of the oil by LC into compound classes and further separation by capillary GC, compo- nent identification was carried out by MS, using ion-trap detection (ITD). The advantage of using this on-line technique lies in the ability to obtain mass spectra of the components of a fraction eliminating the interferences that occur when the whole essential oil is analyzed.
In the field of essential oils, separation of individual enantiomers and deter- mination of enantiomer excess play an important role in the characterization of plant material, in the investigation of the origin of the essential oil, and in the search of possible adulterations. Reliable assessment of the genuineness of essen- tial oils is a difficult task, since synthetic analogs of essential oil components are commercially available. Therefore, suitable specific methods in the authenticity control of essential oils are of fundamental interest. Two-dimensional GC is a useful technique for both analytical and preparative purposes. After a separation of aroma components on a first column, they are on-line transferred to a second
Analysis of Flavors and Fragrances 417
column coated with a different stationary phase for further separation. Enantiosel- ective two-dimensional GC is based on the combination of a nonchiral precolumn and a chiral column, thus proving useful in evaluating origin-specific enantio- meric ratios and for differentiating natural flavor and fragrance substances from those of synthetic origin. By using this system (Fig. 3), the enantiomeric distribu- tion of monoterpene hydrocarbons (β-pinene, sabinene, limonene) and monoter- pene alcohols (linalool, terpinen-4-ol,α-terpineol) has been determined in order to evaluate the genuineness of mandarin essential oils [25]. Besides the character- ization of mandarin oil, the method developed could afford the determination of extraneous oils added to or contaminating the oils.
Other authors determined the enantiomeric excess of limonene and limo- nene-1,2-epoxide by two-dimensional GC and GC–MS with trimethyl-γ-cyclode- xtrin as chiral selector [26]. They found (R)-(⫹)-limonene with an enantiomeric purity between 97.1 and 97.4% and (1S,2R,4R)-(⫹)-limonene-1,2-epoxide with an enantiomeric purity between 88.0 and 91.9%.
On-line two-dimensional GC–MS using a combination of J&W DB-Wax and heptakis(2,6-di-O-methyl-3-O-pentyl)-β-cyclodextrin columns has been pro- posed in a study of enantiodifferentiation of edulans I and II, the key flavor com- ponents of the purple passion fruit [27]. The investigation was applied to a variety of purple passion fruit extracts and distillates of different origins (Kenya, Chile and Ivory Coast). In all samples analyzed, the 2Senantiomers prevailed, as shown in Figure 4.
The results of enantio-two-dimensional GC–MS in authenticity analysis of geranium oil have been recently described [28]. This essential oil, which is ob- tained by distillation of leaves ofPelargoniumspecies, is used in the fragrance industry and also as a flavoring material in foods and beverages. An important olfactive ingredient of the oil is rose oxide (cis-/trans-2-(2-methyl-1-propenyl)- 4-methyltetrahydropyran). Using SPME and enantio-two-dimensional GC–MS, the mechanistic aspects of rose oxide formation using mixed specifically labeled precursors were studied. These findings proved helpful for authenticity assess- ments of geranium oils.
Enantioselective GC–IRMS and two-dimensional GC–IRMS represent other powerful techniques in the authenticity control of natural flavors and fra- grances; using IRMS, identical δ13C ratios are expected for enantiomers from genuine substances. Interesting investigations have been recently accomplished by using GC–IRMS in the authenticity control of essential oils [29–31].
Authenticity assessment of genuine substances of coriander oil (Coriander sativumL.) has been the object of a work dealing with the analysis of 10 authentic coriander essential oil samples of different origins [29]. The techniques used were GC–IRMS and enantioselective two-dimensional GC using a chiral cyclodextrin derivative as the stationary phase. The enantiomer ratio and theδ13C values of 12 characteristic components were compared with those of commercially available
418 Careri and Mangia
Figure 3 Pneumatic and electronic scheme of the GC–GC system. (From Ref. 25.)
Figure 4 Two-dimensional GC–MS enantiodifferentiation of edulans 1a to d in an ex- tract of purple passion fruit from Chile: (a) preseparation on achiral column (J&W DB- Wax), cut 1, 16.8 to 17.1 minutes, cut 2, 19.0 to 19.3 minutes; (b) separation on a chiral column [heptakis(2,6-di-O-methyl-3-O-pentyl)-β-cyclodextrin]. (From Ref. 27.)
420 Careri and Mangia
species and essential oils in order to establish their authenticity. An interesting aspect of this work was the use of suitable internal isotopic standards to perform capillary GC–IRMS. This approach allowed the elimination of the influence of isotopic effects on theδvalues during CO2fixation by photosynthesis in such a way that isotopic effects among authentic substances are only limited by the influence of enzymatic reactions during secondary biogenetic pathways. The iso- topic fingerprint characteristic of the coriander essential oil is illustrated in Figure 5.
Faulhaber et al. [30,31] analyzed mandarin essential oils by measuringδ13C ratios of characteristic flavor compounds of this product. Mandarin oils, which are obtained by cold-pressing the peel of the fruits ofCitrus reticulataBlanco, are used in the food industry and in perfume compositions. The main constituents of mandarin essential oil are limonene (approximately 69%) andγ-terpinene (ap- proximately 20%), even though the fragrance of this oil is mostly determined by minor components such as methyl N-methylanthranilate (approximately 0.4%) andα-sinensal (approximately 0.3%). Evaluation of genuineness of this product is of special interest, since synthetic analogs of the essential oil components are commercially available. In this context, measurements ofδ13C andδ15N of methyl N-methylanthranilate [30] and ofδ13C of characteristic flavor components in this oil proved helpful in the authenticity assessment of cold-pressed mandarin oil [31]. The characteristic profile of mandarin essential oil so established could be applied to the authenticity control of commercially available mandarin oils.
Figure 5 Isotope fingerprint of the essential oils from authentic samples (n⫽10) ofC.
sativumL. with maximal and minimal values;γ-terpinene is the internal isotopic standard:
limonene (a),γ-terpinene (b),p-cymene (c), linalol (d), geraniol (e), myrcene (f), geranyl acetate (g),β-pinene (h), camphene (i), terpinolene (k), and sabinene (m). (From Ref. 29.)
Analysis of Flavors and Fragrances 421