Study of the volatile acidity of French Antilles rums

Fahrasmane L., Parfait A., Jouret C., Galzy P., 1983. Etude de l’acidité volatile des rhums des Antilles françaises. Industries alimentaires et Agricoles. 100, 297–301.

Study of the volatile acidity of French Antilles rums
by L. FAHRASMANE*, A. PARFAIT*, C, JOURET**, P. GALZY*** with the technical collaboration of E. PACE*
*Station de Technologie I.N.R.A. 97170 Petit-Bourg, Guadeloupe.
** Station de Technologie I.N.R.A. CRA Toulouse, 31320 Auzeville.
*** Chaire de Génétique E.N.S.A.M., 34060 Montpellier Cédex.

Summary

Several types of rums are elaborated in the French West Indies. The raw material used (vesou, molasses, vinasse), the microorganisms involved in the fermentation (yeasts, bacteria), the distillation and storage conditions influence the quantitative and qualitative composition of the volatile fatty acids of these different rums.

INTRODUCTION

The fraction corresponding to the fatty acids of spirits and fermented drinks has been the subject of much work. Its composition can be mentioned: Dubois P. and Jouret C. (1965), Nykanen L. et al. (1968); its metabolism: Suomalainen H. et al. (1967). Clarke B.J. et al. (1981) specifically studied, in beer, some factors relevant to the formation of hexanoic, octanoic and decanoic acids.

The rum is, in the French West Indies, the eau-de-vie obtained from various raw materials derived from sugar cane (Saccharum sp): the juice of cane or vesou, the syrup, which is concentrate vesou, and molasses, a by-product of the manufacture of sugar. Depending on the raw material used for the production of rum, fermentation and distillation conditions, as well as possible aging in wooden barrels, different types of rums can be distinguished:

– The agricultural rum made from cane juice or syrup. The rum made from juice has a characteristic flavor reminiscent of that of Vesou.

– Industrial rum made from molasses diluted with water; sometimes the dilution is made with a mixture of water and vinasse. Its aroma is judged generally less fine, but more intense and more persistent than that of the agricultural rum.

– The rum grand arôme; the vinasse, entering into the dilution of the molasses, can undergo a pre-fermentation. This rum has a very high non-alcohol content. It is mainly used as an aromatic ingredient in blends and is made only in Martinique.

– Light rum, like industrial rums, is based on molasses; a rapid fermentation and a high degree of distillation make it possible to obtain distillates which are relatively lightly loaded with non-alcohol.

– Old rum is made from agricultural rum, industrial or light and very rarely rum grand arôme. It is kept in oak barrels with a maximum capacity of 650 liters, for a legal time of at least 3 years.

The development of gas chromatography, its coupling to mass spectrometry, as well as the contribution of nuclear magnetic resonance and infra-red spectrometry, related to the improvement of the extraction techniques of volatile compounds allowed , to many researchers, to analyze the aroma of rum: Shito E. et al. (1962), Baraud J. et al. (1963), Maurel A. (1964), Maurel A. et al. (1965), Stevens R. et al. (1965), Maarse H. et al. (1966), Nykanen L. et al (1968), more recently Ter Heide R. et al. (1981). At the same time, there has been a growing interest in the quantitative and qualitative knowledge of the fraction of volatile fatty acids because of its organoleptic importance in eaux-de-vie. In rum, to date, the following short-chain fatty acids have been identified: acetic, propenoic or acrylic, isobutyric, butyric, 2-methyl-butyric, isovaleric, valeric, 2-methyl-pentanoic, isocaproic, caproic, 2-ethyl-3-methyl-butyric.

Note that the latter was considered characteristic of rums. Unpublished results, obtained by one of us, show that this acid may be present in other alcoholic beverages.

Three main factors can influence the volatile acidity of eaux-de-vie: fermentation agents (yeasts and bacteria), fermentation temperature and distillation. The sugars of the raw material are degraded by alcoholic yeasts, mainly of the genus Saccharomyces (for the rum, grand arôme, there are also yeasts of the genus Schizosaccharomyces). In addition to indigenous yeasts, baker’s yeast, introduced several years ago, tends to take a predominant place.

The volatile acidity produced during the alcoholic fermentation is in close connection with the species and the yeast strain present. In addition, in the French West Indies, where fermentation media are poorly protected, bacteria of the Clostridium and Lactobacillus genera, brought in part by the raw material, can develop (Perfect A. et al., 1978) and participate in the production of volatile fatty acids.

Fermentation accidents due to bacteria of the genus Corynebacterium yield acrolein rums with an unpleasant flavor (Lencrerot P. et al., 1982).

The abnormal elevation of the fermentation temperature favors the production of volatile acidity (Kervegant P. 1946) by microorganisms.

Distillation is an important stage in the production of eaux-de-vie (Mejane J. and Piquois J., 1975); depending on the operating conditions, it gives rums more or less rich in volatile fatty acids.

Ethyl 3 methyl butyric acid, demonstrated in rums by Lethonen J. et al. (1977) and Ter Heide et al. (1981), considered characteristic of rums, revived interest in short chain fatty acids. The development of a simple and reliable chromatographic method for the determination of short fatty acids has enabled us to obtain usable results on rums: it is a necessary step to be able to carry out a systematic study on the formation of these compounds in this brandy.

MATERIALS AND METHODS

As part of a routine check, rum samples were selected for determination of their volatile acidity. As selection criteria, diversity of types and provenance were selected. Among the samples, pouring rums [over proof?] (distillate not brought back to the marketability alcoholic strength) and aged rums having been subjected to a longer or shorter aging period were analyzed. Rums from either good or bad fermentations were also studied.

The gas chromatographic assay of short fatty acids has two steps: the preparation of the sample and the actual dosage.

After determining the overall volatile acidity by steam distillation according to the technique of Jaulmes P. (1951), a sample volume of 60 ml is taken in the case of a spirits of equal volatile acidity. or less than 5 meq/l, 40 ml in the case of a volatile acidity of between 5 and 20 meq/l and 20 ml if the volatile acidity is greater than 20 meq/l.

[milliequivalents per liter (meq/L) Milliequivalents per Liter (mEq/L). An equivalent is the amount of a substance that will react with a certain number of hydrogen ions. A milliequivalent is one-thousandth of an equivalent.]

In the sample volume thus determined, 1 ml of ethyl 2-butyric acid dissolved in water at 1 g/l is added as internal standard. Steam is then distilled to recover about 250 ml of distillate, which is neutralized exactly at pH 8.3 with a deci-normal sodium hydroxide solution. Evaporated to dryness under vacuum, the temperature of the water bath does not exceed 45°C. It is taken up in 1 ml of normal phosphoric acid; it is adjusted with one or two drops of concentrated phosphoric acid so that the pH is 1.

The assay conditions by gas chromatography are as follows:

– Pyrex column of 1/4 inch in outer diameter and 1.80 m in length.
– Support: chromosorb W-AW 80/100 mesh.
– Phase: 10% SP 1200 + 1% phosphoric acid.
– Flow rates: nitrogen 50 ml / min, hydrogen 40 ml / min, air 350 ml / min.
– Temperatures: oven: 105 ° C; injector: 175 ° C; detector: 250 ° C.

The reference solution is composed of acetic acid, propionic acid, propenoic acid, isobutyric acid, butyric acid, isovaleric acid, valeric acid, 2-ethylbutyric acid, 4-methylvalenic acid, caproic acid with 1 g / l concentration in deci-normal phosphoric acid.

The acids serving as control components were obtained from the company Fluka A.G., the guaranteed purity of which is greater than 98% for one and 99% for the others.

RESULTS AND DISCUSSION

The method of determination of the short fatty acids used requires, beforehand, a treatment of the samples; the addition of an internal standard, before any manipulation, contributes to obtaining a quantitatively valid analysis (1% error). As for the chromatographic method, it is specific to volatile short-chain fatty acids (fig.1).

The calculation of the percentage of the sum of the short fatty acids, determined by chromatography with respect to the total volatile acidity, shows that the short fatty acids constitute the essential of the volatile acidity of the rums (table 1). For this calculation, rums with a volatile acidity of less than 1.6 meq/l, which generally correspond to light rums, have been distinguished from rums whose volatile acidity is greater than 1.6 meq/l; the pouring rums are examined separately, since their volatile acidity is generally higher than 1.6 meq/l and their alcoholic strength is 20-30 ° GL higher than the two previous groups of rums whose alcoholic strength is about 50 ° GL.

It is noted that the chromatographic method gives a good correlation for a rum of high alcoholic strength (greater than 70° GL) and whose volatile acidity is greater than 1.6 meq/l. For rums with a lower alcoholic strength (around 50° GL) and a volatile acidity of more than 1.6 meq/l, the correlation coefficient is less satisfactory; the percentage above 100% is not significant because, taking a safety factor of 99%, this percentage varies from 97.8 to 105.1. As for rums of volatile acidity lower than 1.6 meq/l, a slight loss of volatile acids is observed: on these rums of low acidity, the efficiency of the separation is less than that of the other samples. Nevertheless, the difference observed remains within tolerable limits and does not seem to be able to modify the interpretation of the results.

For the overall volatile acidity determined by chemical titration, the average was calculated for samples of different types of rums. It is very variable from one type to another (Table 2). Light rum has the lowest volatile acidity of 0.9 meq/l. There is a remarkable difference between the agricultural rum of Guadeloupe and that of Martinique, the latter is less acidic. This difference can be explained by sanitary prejudices generally better followed in Martinique, but especially by a different method of distillation.

The molasses rum, which is an important part of the production, has an average acidity of 5.5 meq/l. In the case of high-flavored grand arôme rum obtained from a particular fermentation medium, where the duration of fermentation is 8 to 12 days (instead of 36 to 48 hours for other types), with a high bacterial activity, gets a very acidic rum: 15 meq/l. For old rums, there is a rise in volatile acidity compared to the same types of young rums. This is normal, because during aging oxidation phenomena give rise to volatile acids (mainly acetic acid). Valaer P. (1937) was able to observe that after a two-year stay in warehouses in the United States the content of rums in volatile acids could, in certain cases, quadruple and even quintuple.

Finally, rums that have undergone a fermentation accident show a high volatile acidity, which is a clue to distinguish them. As a reminder, Nykanen L. et al. (1968) analyzed rums from Martinique and rum from Jamaica, and found volatile acidities of 4 to 10 meq/l; the types of rums analyzed have not been specified.

The chromatographic method used makes it possible to determine the free short fatty acids, except for formic acid. The acids found in the rum samples are acetic acid, propenoic acid, propionic acid, isobutyric acid, butyric acid, isovaleric acid, valeric acid, isocaproic acid, caproic acid and 2-ethyl-3-methyl-butyric acid. For the discussion, we will make a distinction between acids that are always present and those that are not. Let’s first look at the first group (Table 4).

– Acetic acid is a secondary product of fermentative metabolism of sugars by yeast. In general, it constitutes more than 90% of the volatile acidity of rums (Table 2), except for Martinique’s agricultural rum (76.4%). These percentages are higher than those observed for other spirits (Nykanen, L. et al., 1968).

– Propionic acid comes from the metabolism of yeast and that of bacteria possibly present in the fermentation medium. Its relative abundance in the high-flavor rum (Figure 2) could be attributed to the high bacterial activity in the fermentation medium.

– Isobutyric acid has the precursor valine (Suomalainen H. and Keranen, 1967). Its relative importance decreases in rums originating from fermenting media with high bacterial activity, such as grand arôme rums and rums that have suffered a fermentation accident (acrolein …) (figure 2).

– Butyric acid is produced by the lipid metabolism of yeast, but also by butyric bacteria sometimes present in fermentation media (Parfait A. et al., 1978). This is the case of acid rum (Figure 2, Table 4) which comes from a fermentation contaminated by these bacteria.

– Isovaleric acid has the precursor leucine (Suomalainen H. and Keranen, 1967); it is present in variable proportions but always weak (Fig. 2, Table 3). There is even less when the bacterial activity was strong during the fermentation.

– Caproic acid is, like butyric acid, a product of the lipid metabolism of yeast.

In the same table (4) are the acids of the second group.

– Propenoic acid: its presence seems random in agricultural and light rums; it is, on the other hand, very often present in the rums of molasses and in the corresponding old rums. The presence of this acid is most likely related to the activity of bacteria involved in the metabolism of acrolein. The increase in its frequency in old agricultural rums is due to the practice of choosing acid rums for aging. These have sometimes suffered fermentation accidents.

– Valeric acid is very often absent from light rums (Table 4) due to the distillation technique used; it is sometimes absent from the agricultural rums of Guadeloupe. This acid, when present, does not represent more than 3% of short fatty acids, apart from acetic acid, which represents concentrations of the order of one thousandth of a milliequivalent per liter (Table 4).

– Isocaproic acid is present only in light rums and in certain samples of grand arôme rum, in a small quantity of the order of one thousandth of a milliequivalent per liter.

– 2-ethyl-3-methyl butyric acid is very often absent from light rums and frequently present in other types of rum. It was considered characteristic of rums (Lethonen J. et al., 1977); recent work not yet published shows that this acid can be found in other alcoholic beverages at lower concentrations. Its absence in the light rum is due to the distillation technique.

CONCLUSION

In the light of these results, it appears that certain short volatile acids can be considered as normal constituents of rum; acetic, propionic, butyric and caproic acids are always present; isobutyric, isovaleric, valeric and 2-ethyl-3-butyric acids are in many cases. The balance of these acids and the overall volatile acidity are certainly the two essential factors in the quality of rums from the point of view of fatty acids.

It should be remembered that the grand arôme rums are particularly rich in propionic acid and have a relative fatty acid composition very different from other rums; likewise, the proportions of the various fatty acids seem to vary with aging and could be attributable, in particular, to oxidation and hydrolysis phenomena.

Butyric acid is present in most of the analyzed rums (73 out of 74); however, too much concentration may be evidence of a serious fermentation accident. Considering the samples analyzed and their origin, it can be estimated that more than 35% of butyric acid in short fatty acids, apart from acetic acid, can be considered as abnormal.

In the case of propenoic acid, it is clear that its presence even in very small amounts is a sign of significant bacterial activity in the fermentative medium unfavorable to the rum flavor; it is a good indicator of microbiological purity of fermentation.

BIBLIOGRAPHY

BARAUD J., MAURICE A., 1963. Ind. Aliment. Agr. (Paris), 80, 3.

CLARKE B.J., DAVINE D.F., HAWTHORNE D.B., KAVANAGH T.E., MOULDER P.J., 1981. MBAA Technical Quaterly 18, 188-194.

DUBOIS P., JOURET C., 1965. C.R. Acad. Agri. Fr., 595599.

JAULMES P., 1951. Analyse des vins. 386-398. 2nd éd. Librairie POULAIN, Montpellier.

KERVEGANT D., 1946. Les éditions du Golfe. Vannes, 512 pages.

LENCREROT P., PARFAIT A., JOURET C. Science des Aliments, sous presse.

LETHONEN J.M., GREF K. B., PUPUTTI. V.E., SUOMALAINEN H. 1977. J. Agric. Food Chem., 25, 953.

LIEBICH H.M., KOENIG W.A., BAYER E., 1970. Journal of Chromatographic Science, 8,527.

MAARSE H., ter NOEVER de BROWN M.C., 1966. J. Food Sci., 31, 951.

MAUREL A., 1964. Compt. Rend. Acad. agr. France, 50-52.

MAUREL A., SANSOULET O., GRIFFARD V., 1965. Ann. Fals. Expert. Chim. 58, 291.

MEJANE J., PIQUOIS J., 1975. Anh. Techn. Agric…, 24, 343-359.

NYKANEN L., PUPUTTI E., SUOMALAINEN H., 1968. J. Food Sci., 33, 88-92.

PARFAIT A., GANOU-PARFAIT B., 1978. Nouv. Agron. Antilles-Guyane, 4, 314, 261-273.

SHITO E., NYKANEN L., SUOMALAINEN H., 1962. Teknilsen Kernian Aikakauslehti 19, 753.

SUOMALAINEN H., KERANEN J.A., mistilehti B40-280.

SUOMALAINEN H., NYKANEN L., 1972. Wallerstein Laboratories Communication 35, 185-202

STEVENS R., MARTIN G.E., 1965. J. Assoc. Offic. Agr. Chemist., 48, 802.

Ter HEIDE K., SCHARP. H., WOBBEN H.J., de VALOIS P.J., TIMMER R., 1981. The Quality of foods and Beverages, vol. 1., Chemistry and Technology, 193-200, Academic Press.

VALAER F., 1937. Ind. Eng. Chem. XXIX, 988-1001.

YOKOTA. M., FAGERSON I.S., 1971. Journal of food Science, 36, 1091-1094.