Follow along: IG @birectifier
There is nothing too remarkable in here, but we do see the use of quite a lot of dunder.
A chemical control test in a rhummerie in Martinique
By M. SAUSSINE
In 1897, on the proposal of an industrialist from Martinique, I tried to set up a chemical control test in a rum distillery in Saint Pierre. The trial lasted two months; I have only been able to devote to it the leisure time left to me by my teaching. I was diverted from this research for some time but I have since been able to come back to it and if I have deferred publication, it is because I did not yet judge the whole to be coherent enough. After the interesting work that M. Pairault has just published on the rhummeries of Guadeloupe, (1) I cannot delay presenting a few observations.
(1) Voir Bulletin de septembre 1899. [I skimmed this but did not feel it worth prioritizing for translation.]
We have never explained these two words: rum and tafia. I find in statistics dating from 1818 a description of the manufacture at that time where I note these definitions:
“The first water that comes from the distillation is the rhum; maintained at 25° it is sellable [merchant] (2).”
(2) It is obviously about degrees Cartier.
“The tafia comes next; we deliver it to be sold at 28°.”
“We call petite eau everything that happens outside this degree and which is more or less weak; we use it by redistillation to make spirit.”
“The rumier (sic) made good use of the substances supplied to him in syrups when he delivered a gallon of taffia for a gallon of syrup.”
Here, as in Guadeloupe, several kinds of rum are made, depending on whether raw cane juice, cooked cane juice or molasses from a sugar factory are used. I won’t talk about the gros sirups, but I will point out that the molasses obtained by turbines in the agricultural sugar factories which still use the old Wetzel evaporation devices cannot be compared to the molasses of the central sugar factories where the extraction of sugar is pushed further. They give an excellent quality rum which is almost entirely exported.
[I have seen the gros sirup term frequently, but not known how to interpret it. Here it is the old Wetzel evaporation process.]
As for cane juice rums, the one prepared with raw cane juice is consumed immediately on the spot under the name of tafia grappe blanche; it is said to be difficult to store. Fermentation is uncertain, a large number of fungi causing cane diseases, among others Thielavopsis ethaceticus (Went), intervening in the torula form to modify the result. This is why those who export rum from cane juice, and there are many of them in Martinique, take care to cook the cane juice, generally in the Père Labat crews, to bring it to 10°, 20° and even 40° Baumé; they then obtain the battery syrup which cooking has sterilized.
Tanks — The volume of the tanks in most rum distilleries are between 2,000 and 3,000 liters, fermentation in industrial rum distilleries lasts 5 to 6 days. A large factory which works continuously, distills and reloads a fairly large number of vats every day. In my tests, I chose a few at random to determine the density, temperature and acidity each day; microscopic examination each time completed these observations.
This is not the place to discuss whether a large number of small vats or a small number of large vats should be used. When you are not sure of the result, it is obviously unwise to ferment 10 to 20,000 liters at once. Another reason can be alleged: the ambient temperature is already very high; the exothermic work of fermentation elevates it still further; by multiplying the tanks, the cooling surfaces are increased; however, from the point of view of the bouquet, there is an advantage in lowering the temperature.
Molasses and composition — As a result of the increase in duties on foreign molasses, the importation of these has greatly decreased and products from Martinique and Guadeloupe are mainly used. Their richness is quite high and the proportion of glucose there is considerable: a third or sometimes a half of crystallizable sugar. Also the fermentation is established very quickly as soon as the molasses is brought to the suitable dilution by adding dunder (vinasse) and water.
Depending on the richness, we use 10 to 12% syrup; As for the proportions of dunder and water, they vary from one factory to another or during manufacture. Some agricultural rhummeries do not use dunder; in industrial rhummeries, 55 to 65% are used.
Density − Starting from a density close to 7° Baume, some cane juice rum manufacturers obtain a complete attenuation of the density, but in industrial rhummeries the densities are always higher. With 9 to 10° Baume at the start, the fermentation seems to stop around 4 or 5 (1.030) and the resulting dunder reaches up to 1.040 (at 28° C).
If we follow the density in the vats day by day, we find that with good fermentation it begins to decrease the day after loading and this rapid decrease in the first three days slows down. But there is often delay and other times a slow and steady decrease. These anomalies are related to those presented by the other elements temperature and acidity.
Temperature − It should be remembered that the average temperature of the island is 27° by the sea to the west. As cane hardly grows above 300 meters above sea level, the factories are by the sea or at a low altitude. However, 300 meters greatly changes the conditions of fermentation, as well as the mode of cultivation of the cane. While at the seaside the fermentations sometimes take place at 40°, in the high factories the average is always below 30°. during part of the year, in the high elevation the planter can direct his cultivation so as to have canes all year round. My observations focused mainly on industrial rhummeries. With an initial temperature of 33° or 31°, the graphs first show an increase from 4° to 5° during the first two or three days and then a slower decrease. But with initial temperatures of 38° I always observed a decrease first, then an increase. In cases of bad fermentation the temperature has always decreased.
Acidity — It has often been said that fermentations in industrial rhummeries occur in the presence of very high acidity; however, it never occurred to anyone to try to diminish it. It is largely to the dunder that it must be attributed since during fermentation it increases little, in view of its absolute value. Volatile acidity is not very considerable since rums in general are not more acidic than other consumption alcohols. There is therefore in the dunder a fixed acid, or at least not very volatile, which is transmitted from one operation to another. Almost all rhummeries use sulfuric acid but at very low doses 0.2 grams per liter, therefore out of proportion with the numbers found. In good fermentations the acidity has varied from 4 grams or 4.5 grams (as SO4H2) at the start to 6 or 6.5 for the grappe, 7.5 for the dunder and 0.5 for the tafia. In bad fermentations the acidity at the start is much more considerable. There are dunders that have more than 15 grams per litre.
Dunder — Left to cool in basins, dunder leaves a deposit in the long run which would be very valuable as fertilizer; but it is not given time to form. The still cloudy and barely cooled liquid is used immediately for the wash. There is sometimes a little alcohol there, but more often exhaustion has been complete. Industrial rhummeries work with a fairly high proportion of dunder; when it becomes too acidic, which we recognize by the progress of fermentation, we put less of it or we start with water alone. The excess of each day, or dunder that has become bad, is thrown back into the sea. The question of waste water does not exist; but if it does not exist from the point of view of hygiene, it is necessary to pose it from the point of view of fertilizer. There is indeed in this enormous volume of lost liquid, a source of fertilizing matters. I measured up to 15 grams of nitrogen per litre. It is an invaluable fertilizer for the agricultural rhummeries which can spread it without transport costs on the manures or in the fields; but for industrial rum, it remains to find an economical means of concentration or extraction.
The density is very high; I observed up to 1.045 at 28°. The acidity also; I found numbers varying between 10 and 20 gr (SO3) per liter. It’s a very reductive environment; it reduces silver nitrate and Fehling’s liquor well; it therefore contains aldehydes and it is easy to recognize furfurol in it. These tests are successful either with the liquid decolorized by means of lead acetate and phosphate of soda, or with the product of the redistillation of the dunder. By subjecting it to distillation, the boiling which is maintained for a long time at 100° yields a liquid hardly heavier than water (0.9968 at 29°) having absolutely the smell of ordinary tafias; this liquid is colored grenadine pink by addition of a drop of aniline without addition of acid. There are therefore other products with a stronger smell than furfurol and whose dunder is, so to speak, saturated. This appears as a reserve of aromatic products or, as the industrialists say, empyreum, not very volatile, and the alcoholic vapours, crossing the trays which are loaded with it, carry off a small quantity.
I stored dunder samples for several months in stoppered bottles. The well-filtered liquid is dark brown and keeps perfectly clear; its acetic odor diminishes and its aroma gradually improves. In the air, the liquid becomes cloudy again and becomes populated with bacteria.