Handbook of Vanilla Science and Technology

The quantitative conversion of glucovanillin to vanillin during the curing process was investigated by the Symrise group (Gatfield et al. 2007), which reported losses up to 50% of the vanillin during the bean curing process. Analysis indicated approximately 9 g glucovanillin per 100 g of dry substance initially. During the curing, vanillin was dimerized to divanillin accounting for some of the vanillin loss.

An interesting and intensive genetic investigation of the origin of Tahitian vanilla was reported by researchers in 2008 (Lubinsky et al.). They collected a variety of vanilla species from natural and cultivated populations. From their genetic analysis they concluded that V. planifolia and V. odorata were the parental species of the interspecific hybrid species V. tahitensis and that there has been little evolutionary divergence since its probable hybridization during the Late Classic (1350-1500) in Mesoamerica.

Waliszewski et al. (2009) extensively studied the activity of the vanilla enzyme polyphenol oxidase (PPO), which was considered to be a factor in the browning of vanilla beans during curing. The stability of the extracted and purified enzyme was studied under a variety of conditions, including pH and temperature. Also, the inhibitory effects of several chemicals were investigated. The optimum temperature was 37°C and the pH, 3.0 and 3.4, depending on substrate. These researchers proposed that because vanilla PPO utilizes mono- and diphenols in order to catalyze oxidation, that the enzyme may decrease the vanillin content in the cured bean.

French (La Reunion), Dutch and Malaysian scientists (Palama et al. 2009) applied ^:H-NMR and LC-MS in the study of the metabolic variation of developing V. planifolia pods. This metabolomic analysis was performed on the green pods at several developmental stages. Analysis of the data was performed with multivariate data analysis. Generally, younger pods contained more glucose, malic acid, and homocitric acid while older pods contained more sucrose, glucovanillin, vanillin, p-hydroxybenzaldehyde glucoside, and p-hydroxybenzaldehyde. They suggested their results may serve as a base to clarify the biosynthetic pathway for vanillin.

10.4 VANILLA QUALITY AND AUTHENTICATION ANALYSES

In order to assure the consumer that the product purchased is indeed vanilla, governments worldwide have tried to define what constitutes a product called vanilla. As discussed earlier, in the United States, vanilla products are defined by a “standard of identity” found in 21CFR169.175. Similar laws occur throughout the world in legal statements such as that found in New South Wales, Australia: “the quantity of soluble substances in their natural proportions that are extracted by aqueous Alcoholic solution...” (Archer 1989). These regulations require a means to police compliance and many such methods have been developed over the last half century.

The early means to detect the formulation or extension of vanilla products with other botanical extracts and juices used paper and thin layer chromatography, which could separate and identify the various adulterating botanicals such as licorice root, cascara, yarrow, cherry, and prune extracts. In addition, the analysis of the vanilla organic acids by ion-exchange chromatography, as well as the analysis of amino acids, helped to validate vanilla products. Interestingly, also used during this period was the questionable lead number (Pb#), which supposedly established the vanilla bean content with a lead precipitation. However, not only was this method dangerous to the analyst as well as the environment, but it was also notoriously inaccurate.

Botanical dilution and substitution could be detected by these early methods but the most common “extender” was the addition of synthetic vanillin as discussed earlier. Vanillin was identified by Gobley in 1858 as the main flavor constituent of vanilla. The synthetic form of vanillin quickly became the compound used to enhance poor quality products and/or produce a purely imitation product. This form of adulteration was difficult if not impossible to detect by the techniques available at that time. The analysis and quantitation of amino acids (Stahl et al. 1962) was also added to the variety of testing methods. In the mid-1970s, a variety of combined and instrumental techniques were investigated. Martin et al. at ATF (1977) investigated the use of constituent ratios in order to detect not only synthetic vanillin addition but also the strength or dilution of the extract. These ratios involved vanillin, potassium, inorganic phosphates, and nitrogen.

While all these methods helped in thwarting adulteration, clever adulterers were still circumventing all these efforts. In the mid- to late 1970s and into the 1980s, two techniques were intensely investigated focusing primarily on detecting the addition of non-bean derived