Chapter 5
Downloaded by UNIV OF PITTSBURGH on April 11, 2017 | http://pubs.acs.org Publication Date: December 28, 2006 | doi: 10.1021/bk-2007-0952.ch005
Flavor Authenticity Studies by Isotope Ratio Mass Spectrometry: Perspectives and Limits Elke Richling, Markus Appel, Frank Heckel, Kathrin Kahle, Michael Kraus, Christina Preston, Wolfgang Hümmer and Peter Schreier* Food Chemistry, University of Wuerzburg, Am Hubland, 97074 Wuerzburg, Germany
In recent years authenticity assessment of flavor substances has gained increased importance by using multi-element isotope ratio mass spectrometry (IRMS). Whereas C/ C ratio determinations have already been performed previously by coupled gas chromatography (GC) in the combustion (C) mode, the technical prerequisites for GC-pyrolysis (P)-IRMS were made available only recently, Thus, the additional information about the H/ H and O/ O ratios of industrially attractive flavor compounds opened the way to multi-element approaches. From the current studies conclusions can be drawn concerning (i) general and (ii) element-specific requirements for GC-C/P-IRMS measurements. They comprise for (i) the availability of authentic reference material, statistically relevant sample numbers, the exclusion of isotope discrimination in the course of sample preparation and chromatographic steps, as well as continuous checks of system stability using certified standards. As to (ii), GC-C-IRMS measurements of C/ C ratios are routinely performed; the GC-P-IRMS determination of H/ H ratios is a highly promising technique provided that the dynamic linearity is checked carefully and, depending on the structure of the target molecule potential isotope exchange is excluded. GC-P-IRMS measurements of O/ O ratios still suffer from the empirical pyrolysis technique and the need to use tertiary standardization. In addition, similarly to H/ H determinations, check of dynamic linearity and potential isotope exchange is required. Considering these requirements, GC-P-IRMS is an additional helpful tool in the authenticity assessment, in particular, as predictions about the global δO and δH values of natural compounds on the basis of their biogenesis are increasingly available. 13
12
2
13
2
1
18
16
12
1
18
2
16
1
18
vsmow
2
vsmow
© 2007 American Chemical Society Ebeler et al.; Authentication of Food and Wine ACS Symposium Series; American Chemical Society: Washington, DC, 2006.
75
76
Downloaded by UNIV OF PITTSBURGH on April 11, 2017 | http://pubs.acs.org Publication Date: December 28, 2006 | doi: 10.1021/bk-2007-0952.ch005
The problem of authentication The quality of food is mainly determined by its aroma which is an important part of the flavor, i.e. the complex sensory impression of odor, taste, chemosensation and texture that appears during eating. Constituents of flavors are the volatile, chemically exactly defined flavor compounds. Approximately 8000 volatile substances have been found in foodflavorsto date (/); about 2000 industrially used materials will be integrated in a list according to El999/217/EWG by the EU (2). Once this 'positive list' is in effect, the EU will have taken a similar approach regarding the approval of flavor materials in foods and beverages, as the USA where the so-called FEMA-GRAS list (a positive list with a little more than 2000 flavoring materials) has been in force for many years. All constituents of industrially producedflavorsare governed by legislation and codes. In the 1998 directive, the EU defined six classes of flavoring substances, i.e. (i) natural compounds; (ii) nature-identical compounds; (iii) artificial compounds; (iv) flavor extracts; (v) processed flavors; and (vi) smoked flavors (3). If we focus our interest on the first two groups, we will define: Naturalflavorcompounds are substances that are obtained only from suitable natural raw materials of plant or animal origin by means of appropriate physical processes, (including distillation and solvent extraction), or through microbiological or enzymatic, i.e. biotechnological processes. Nature-identical flavor compounds are substances that are obtained via chemical synthesis or through isolation from natural products by means of chemical processes. The chemical structure of nature-identical flavor compounds is identical to that of the corresponding natural ones. In order to be considered nature-identical, the flavor compound has to have been identified in plant or animal material traditionally consumed by humans as food. In addition to the EU Directive the IOFI Guidelines are worth mentioning in which an interpretation in more detail is given about the conditions and processes used by the flavor industry on a global basis. Flavor compounds from these two categories differ in their market values. Between nature-identical substances and natural ones average price differences ranging from 1:10 to 1:100 exist. As a consequence, there is a certain temptation to increase the profit by unlawfully giving false declarations, e.g. declaring a nature-identical flavor compounds as natural. Both the industry and the consumer are confronted with this situation, the first when buying raw materials to be used in the composition of a food flavor, the latter when selecting the flavored product at the supermarket. How can the question be answered whether the content corresponds to the declaration? Principally, differences of selected parameters arising between nature and laboratory chemistry are analytically evaluated. Firstly, the wellknown selectivity of nature is used to biosynthesize preferably one of the
Ebeler et al.; Authentication of Food and Wine ACS Symposium Series; American Chemical Society: Washington, DC, 2006.
Downloaded by UNIV OF PITTSBURGH on April 11, 2017 | http://pubs.acs.org Publication Date: December 28, 2006 | doi: 10.1021/bk-2007-0952.ch005
77 enantiomers of a chiral compound, while chemical synthesis - except for asymmetric modifications - leads to an enantiomeric ratio of 50:50. Gas chromatographic (GC) techniques are established to exactly determine the enantiomeric ratio of flavor compounds (4), but they are limited to the small amount of chiral flavor substances. Secondly, the determination of the ratio of stable isotopes is the method of choice to obtain more comprehensive information. Once the analytical information has been collected, it needs to be carefully interpreted, and the decision needs to be made whether the respective natural material has been indeed made from natural starting materials in a process considered natural. The exact determination of the extremely small differences in isotope abundances can be realized by a precise quantitative mass spectrometric method, i.e. the isotope mass ratio spectrometry (IRMS) applicable to simple gases (5). They have to be produced from the organic compounds by combustion (C) or/and pyrolysis (P). The measuring gases needed are C 0 formed in the C-mode as well as hydrogen gas and CO, the latter two generated in the P-mode. The prerequisite for IRMS measurements of complex mixtures of flavor compounds in food is their separation and purification. Separation into pure compounds and IRMS analysis can be performed by online combination of GC with IRMS. The integrated interface consists of two different reactor types, one to realize the combustion of carbon into C 0 (for C/ C analysis) and two others to pyrolyze the compounds, forming H and CO as measuring gases for H/*H and 0 / 0 determinations, respectively (Fig. 1). The isotope ratios are expressed in per mil (%o) deviation relative to the Vienna Pee Dee Belemnite (VPDB) and Vienna Standard Mean Ocean Water (VSMOW) international standards. Mass spectrometrical measurements of isotope ratios were limited to "offline" determinations of C/ C and H / H ratios for a long time. In the past decade, the "on-line" coupling of GC with IRMS via a combustion interface has opened the acess to the C/ C ratios of individual constituents in complex flavorings (5). Recently, the measurement of 0 / 0 ratios was made available in both "off-line" and "on-line" modes using P-IRMS (6-10). The large variations known to exist in the H/*H ratio in nature have made it a very attractive target for IRMS studies. However, technical problems have precluded for a long time successful measurement of H / H ratios of individual peaks eluting from a capillary column. Recently, these problems have been overcome, and H / H determinations of GC peaks are possible using commercially available equipment (11,12). At present, multi-element isotope information is available about a number of 'key' flavor substances, e.g., citral (75), (£)-2-hexenal, (£)-2-hexenol, and decanal (14), linalool and linalyl acetate (15,16), as well as several aromatic volatiles like estragol and methyl eugenol (17). In the following, several selected examples of GC-IRMS analyses are represented and the recent progress of coupled IRMS modes is highlighted. 2
13
12
2
2
2
18
16
13
12
13
2
]
12
18
16
2
2
2
!
1
Ebeler et al.; Authentication of Food and Wine ACS Symposium Series; American Chemical Society: Washington, DC, 2006.
Downloaded by UNIV OF PITTSBURGH on April 11, 2017 | http://pubs.acs.org Publication Date: December 28, 2006 | doi: 10.1021/bk-2007-0952.ch005
78
Figure 1. GC-IRMS-System (scheme). The system consists of a gas chromatograph coupled to an isotope ratio mass spectrometer via a pyrolysis or combustion interface (a, b and c for oxygen, hydrogen, and carbon measurements, respectively) and is additionally equipped with an elemental analyzer (EA).
GC-IRMS measurements: From single to multi-element approach 13
12
If we focus first on the determination of C / C ratio we will notice that this technique has a traditional value in the differentiation of flavor compounds arising from C3 and C4 plants (18) (or C A M - performing the so called crassulacean acid metabolism, such as vanillin (19)). For instance, the Ô CVPDB values of 2-methylpropanol formed by alcoholic fermentation of sucrose from C3 and C4 sources show the expected (5) ranges of -24.7 to -21.4%o and -12.4 to -9.2%o for the product from C3 and C4 origin, respectively. By adding step-bystep cane sugar (C4) to the beet sugar (C3) before fermentation, an increase in the Ô CVPDB values of 2-methylpropanol results (Fig. 2) (20). Literature reveals that authenticity assessment performed solely by means of C / C ratio is in most cases not succesful within die area of flavorings from C3 origin, so the above-mentioned introduction of technical prerequisites for GC coupled IRMS of H / H ratio (11,12) was a milestone in flavor authenticity 13
13
13
12
2
!
Ebeler et al.; Authentication of Food and Wine ACS Symposium Series; American Chemical Society: Washington, DC, 2006.
79
-10 >
Ο
Ο -sj- -τ - o o g g ^ g
-15
"So S
ϋ ^οΟ υ
^ ^^ ο 2> ο ο ο "*~ *~
Downloaded by UNIV OF PITTSBURGH on April 11, 2017 | http://pubs.acs.org Publication Date: December 28, 2006 | doi: 10.1021/bk-2007-0952.ch005
-20
3fi νρ 4 ^ CO ο COνρ «ζ ο ο «ζ cl 10 ο _ ο Ε ^