Handbook of Vanilla Science and Technology

Vanilla extracts show at least 20 organic acid peaks in gas chromatographic analysis (Fitelson and Bowden 1968). Of the eight major peaks, only malic acid is positively identified; citric acid, isocitric acid, and succinic acid are suspected to be among other major acids.

Quantitatively, the organic acids content of vanilla extracts is expressed as “lead number”. Lead number is determined by precipitating lead salts of organic acids with the addition of lead acetate solution to the vanilla extract. Unreacted excess lead acetate is back-titrated to determine the amount of precipitated lead. Often, the lead number is used to determine the purity as well as the strength of vanilla extract. The lead number also indicates whether the extract has been prepared with the correct amount of beans and it used to be the standard method for such a determination. Typically, Bourbon extracts have lower lead numbers, in the range of 0.65 to 0.85 per fold of extract, compared to a 1.1 to 1.4 per fold range for Indonesian extracts. It is not uncommon to see the lead numbers falling out of either side of these ranges for both types of authentic extracts. For practical reasons, it is assumed that a pure single-fold vanilla extract should have a minimum lead number of 0.7. The procedure to measure lead number is described in the publication Official Methods of Analysis of AOAC International, 18th edition (2005), section 36.2.1.

This method of ascertaining the purity of vanilla extract is easy to circumvent by the addition of foreign acids to enhance the lead number. However, if used in conjunction with the gas chromatography method of Fitelson and Bowden (1968), it can be very useful in detecting adulteration of vanilla with foreign organic acids. The volatile trimethylsilylated (TMS) derivatives of vanilla organic acids show a typical pattern of 8 major peaks upon separation by gas liquid chromatography. The TMS pattern for extracts made from beans of different geographic areas varies in peak heights as well as certain peak ratios. A detailed procedure to determine the organic acids pattern in vanilla is described in the publication Official Methods of Analysis ofAOAC International, 18 th edition (2005), Section 36.2.22.

Many vanilla companies have stopped using the lead number method, because of environmental considerations regarding lead disposal.

9.3.2.6 Resin content

Single strength Bourbon extracts contain slightly more than 0.1% resins. Similar strength Indonesian extracts contain 0.04 to 0.06% resins. These pleasant taste producing polysubstituted polymerized phenols are determined quantitatively and their authenticity confirmed by a paper chromatography method, as described in AOAC procedure (2005).

9.3.2.7 Microbial limits

Chances of microbial contamination coming from the use of vanilla products are very small. But it is a concern to most food manufacturers and is of particular concern to the manufacturers of cultured dairy products. Most of vanilla buyers have adapted microbial contaminant limits (Table 9.5):

Table 9.5 Recommended limits for microbial contaminants

Test | Recommended Limits, Counts/g

Total Plate Count | 1000-10,000

Yeast | <10

Mold | <10

E. coli 3 Tube MPN | <3/g

Salmonella | Negative

Thermophilic Aerobic Spores | <10

Mesophilic Aerobic Spores | <10

9.4 DETERMINATION OF AUTHENTICITY OF VANILLA EXTRACTS

Since vanilla is an expensive product and the cost of vanilla beans can fluctuate significantly over a relatively short period of time, vanilla products are tempting targets for sophisticated adulteration schemes. It is therefore a continuing challenge to verify authenticity of vanilla extracts and natural vanilla products. Since addition of synthetic or semi-synthetic vanillin is the simplest and easiest method to reduce the product cost, vanillin is the main target for authenticity determination.