This step could be compared to wine aging, or cheese ripening, because the aromatic quality of the vanilla pods is clearly enhanced during this period.
This curing process is implemented in roughly the same way in all vanilla-producing countries worldwide. It is highly empirical, relatively uncontrolled, and closely dependent on the prevailing climatic conditions—which can lead to major pod quality defects in some circumstances—but it is tailored to the conditions in which farmers live in most producing countries, even in Madagascar—the top vanilla-producing country. Substantial handling is required throughout this vanilla curing process, thus providing a major source of income for many people.
UNCONVENTIONAL TECHNIQUES
The different faults in the conventional curing process, already mentioned, as well as its duration (over six months before a marketable product is obtained), have encouraged many authors to try to develop alternative curing methods. Moreover, it is known that the final vanillin contents are never higher than 3% relative to the dry matter (Bayle et al., 1982; Derbesy et al., 1982; Arnaud et al., 1983; Derbesy, 1989; Falque et al., 1992; Fayet et al., 1999; Derbesy and Charvet, 2000; Saltron et al., 2002; Gassenmeier et al., 2008), whereas the initial potential is usually over 5% (Arana, 1943; Ansaldi et al., 1988; Leong, 1991; Brunerie, 1993; Odoux, 2000; Havkin-Frenkel et al., 2005). These considerations have also prompted research aimed at boosting vanillin levels in the end product.
Some authors have focused on enhancing control of the basic parameters (temperature, relative humidity, processing time, etc.) that supposedly have an impact on the final vanilla quality, while also trying to create conditions that closely match conventional pod curing conditions. Finally, the goal is to imitate the conventional process while “industrializing” the different steps.
Consequently, the techniques proposed by these authors—three of which are patented—are relatively similar, while mainly differing with respect to the initial state of the raw material, that is, depending on whether the pods are whole, in pieces or ground.
Towt (1952) proposed to grind vanilla pods to obtain a uniform purée, based on the idea that the different chemical and biochemical reactions that take place during the vanilla quality development process would be facilitated and accelerated as compared to using whole pods. This purée is subjected for 48 h to a temperature ranging from 50°C to 55°C, with air injection into the mass to promote oxidation reactions, and it is then dried at 60°C until the moisture level has dropped below 20%.
Kaul (1967) used pods that were whole or cut into pieces so as to preserve the commercial “identity” of the end product, while effectively controlling the process to curb mold development, which is a common issue under conventional curing conditions. The process proposed by this author involves incubating pods at 38°C for one week, or at 65°C for 24 h, or any other intermediary time/temperature combination, in a container that is sealed to hamper moisture loss and promote chemical and biochemical reactions. This curing step is followed by a drying process under conditions that are not outlined in the patent description. The author has noted that this drying step can, in all cases, be conducted rapidly—contrary to conventional procedures—at low or high temperature without being detrimental to the product quality. The author considers that the described curing conditions must be closely followed to obtain a top quality end product.
Karas et al. (1972) proposed to work on pods cut into pieces and placed on racks. The racks are put in an oven at 60°C for 72 h. The pods are dried at 60°C until a moisture content of 35–40% is reached, and then at ambient temperature until the pod moisture content levels off at 20–25%.
These three techniques were patented by McCormick & Company, Inc., and to our knowledge one of them (Karas et al., 1972) is used in Uganda and also by an industrial group in Madagascar.
As early as 1949, Jones and Vicente (1949a) published a study in which relatively similar procedures were compared. Whole, cut up, or ground vanilla pods were oven dried at 60°C for 24 h, and then at 45°C (until the whole pods were flexible), and finally dried and conditioned at ambient temperature. The results of aromatic tests on ice creams flavored with the corresponding vanilla extracts did not reveal any significant differences between treatments.
Bourriquet (1954) also pointed out that in 1950, a Mexican company was already producing a very flavorful product called “vanilla fruit,” which was produced from a green vanilla purée that was heated for 48–60 h and then dehydrated in a dryer.
Théodose (1973) reported the findings of studies carried out at the Antalaha research station in Madagascar in the 1960s, which aimed at simplifying the vanilla curing procedure. One technique was developed that involved scalding and sweating the whole pods in the conventional way, cutting them into 2–3 cm pieces, and then drying them in batches for 12 days to obtain an end product with 20–25% moisture content. Vanilla importers considered that the results were interesting, as the vanillin contents were higher than normal and the aroma was stronger.
Other authors focused more on optimizing the reactions to hydrolyze glucovanil-lin into vanillin by endogenous glucosidase (see Chapter 10 and section “Relationship between Vanilla Curing and Aroma Development,” of this chapter), while using a specific treatment or technology to bring these components into contact. This treatment could be the initial step before a more conventional drying step or when manufacturing a flavor extract.
Growth hormones such as ethylene were studied long back by Arana (1944) and Bourriquet (1954). It was not clearly determined whether vanilla beans are climacteric or not, but their sensitivity to ethylene as a maturation-inducing agent was clearly demonstrated by these authors. This ethylene induction process leads to pod browning and glucovanillin hydrolysis. However, ethylene induces high pod dehiscence which, according to Arana (1944), could affect up to 47.5% of all pods. More recently, Havkin-Frenkel et al. (2005) generally confirmed these high dehiscence rates in pods treated with ethylene in the presence of air or oxygen. Treatments with ethylene, naphthalene acetic acid (hormone of the auxin family), or biotic elicitors have also been tested on pods that had undergone heat treatment (63°C for 3 min), combined or not with scarification (Sreedhar et al., 2007, 2009). According to the authors, these treatments had a positive impact on the formation of vanillin and other aromatic phenolic compounds, while also reducing the vanilla curing time. The results of these experiments were somewhat inconsistent, however, so it is hard to draw clear conclusions. In addition, the physiological impact expected from these molecules on pods that had undergone a senescence-inducing heat treatment could be questioned.
Ultrasound has also been studied as a technique for bringing enzymes and substrates into contact (Obolenski, 1957, 1958, 1959) and, according to the author, this technique can boost the vanillin contents (but the results were quite confusing overall).
Cernudat and Loustalot (1948), cited in Bourriquet (1954), tested the use of infrared radiation treatments as an initial vanilla curing step, and the quality of the end product was considered fairly acceptable.
Crushing vanilla pods—without squashing or breaking them—by around 10 runs through rollers at 20 kPa pressure was also found to enhance the hydrolysis of gluco-vanillin into vanillin (Perera and Owen, 2010).
A technique involving freezing/thawing of green pods has been patented (Balls et al., 1942; Ansaldi et al., 1988) and has been the topic of different publications (Jones and Vicente, 1949a; Odoux et al., 2006). According to Ansaldi et al. (1988), this technique enables hydrolysis of around 80% of the glucovanillin present in green pods. In similar conditions, our findings showed that it is even possible to obtain hydrolysis rates of over 90% (Odoux et al., 2006).