The idea here is to use the birectifier to analyze vermouths and also answer the question: is the birectifier a useful tool for vermouth producers? The short answer is, yes, it is a useful tool, and the long answer is that I’ve had many false starts and made a bunch of mistakes developing a protocol. At the moment with my rushed schedule it is hard to gather the resources to properly elaborate everything.
The challenge with vermouth, as opposed to the distillates, I’ve been studying, is that you need to perform a stripping run. It simply adds hours to the process that we will solve with automation. Sadly, we are not there yet. The false starts are mainly from not having a proper heating profile for the larger stripping mantle. After stripping with the birectifier, the spirit has to be fractionated the traditional way. You cannot develop enough reflux to do it in one pass, but in the future that may change as we learn more about varying throughput with automation (we’d run it slower).
The case study I examine below has a major error and is basically a mulligan that will need redone, but there were enough insights to write it up. I erred in having more than 100 ml of absolute alcohol in the distillate. This skews the the fractions in theory, but in practice I think everything worked out. 100 ml of absolute alcohols aligns the fractions in a way that alcohol is exhausted in the very beginning of the 5th fraction and then a band of low volatility high value congeners comes across all in a short duration. In this case, this band came across not at the beginning, but at the middle of the 5th fraction. This would allow too much fusel oil into the fifth fraction and possibly not enough time to collect all the high value congeners (in this case products of aromatization). In practice, this fraction 5 still retained all the identity of the vermouth. The other fractions actually add no specific identity. The role model of course is none other than Noilly Prat dry vermouth.
Fraction 5, the quality regulating fraction containing all high value congeners, is also quite complex and just as beyond language as vermouth usually is. The birectifier does not help us get much further than “vermouth is a wine which looks and tastes like vermouth”, but hopefully peeling away the layers can physically isolate that concept to help achieve it.
Vermouth is probably the most complicated and elaborate product in all the culinary arts. As I’ve said before, it is the summation of oenology, distillation, and perfumology. The birectifier reveals details about oenology especially when we see appreciable acetaldehyde in Noilly Prat’s wine base. This could likely be because Noilly Prat is madeirized, oxidized, and basically tortured. We don’t see acetaldehyde to this degree in conventional distillates, but Noilly Prat seems to be able to rock it, and thus we know it is acceptable.
On the distillation side of things, we know Vermouths must have a certain percentage of wine which is basically as much as possible so the distillate occupies a small percentage. High proof fortifying spirits would be mainly neutral, but there would be room to use ideas like Arroyo’s super fractionated spirits concept where you basically birectify at production scale and assemble a distillate from parts. If you had a byproduct market for the parts you don’t use, you could create an extremely high proof fortifying spirit while including high value congeners that would typically be excluded. Noilly Prat, however, demonstrated no cloudiness, no emulsion, or anything resembling Cognac oil in its fraction 5.
Perfumology comes in because the aromatization of Noilly Prat is so sophisticated it has no culinary analog. If we wanted to create new botanical formulas and/or advance this art, we’d have to look to powerful new concepts like semantic odour space and the computation that was goes with it to correlate various words beyond (object comparisons) to chemical compounds and the botanical products that contribute them.
I have yet to perform any survey of conventional role model table wines using the birectifier. We may find patterns where exemplary wines have extraordinary fraction 5’s that demonstrate high value congeners. I have long described vermouth as the supernormal stimuli version of wine. Birectifier exploration of both exemplary tables wines and vermouth may contribute more insights to articulating this. With new constellations of understanding, we will arrive at new focal points for involvement and avenues to advance the craft. I have long thought the ancients made vermouth because their tastes were more advanced than their ability to commonly produce great wine and I still think I’m correct. Vermouth is an ingenuity that follows the lead of extraordinary wine.
Fraction 1: Ethyl acetate and distinct note of acetaldehyde. Far more concentrated than you’d think and the aldehyde content is higher than you see in distilled spirits. No extra notes, but things are possibly concealed by the above threshold ordinary congeners.
Fraction 2: Same notes as fraction 1, but less concentrated. No new notes, nothing that resembles vermouth.
Fraction 3: Fairly neutral. No distinct character
Fraction 4: Perceivable fusel oil, but not the level typical of spirit, possibly because the charge was above 100 ml of absolute alcohol.
Fraction 5: Recognizable as vermouth! Defies language like vermouth! Acrid taste on the palate like drinking other concentrated fraction 5’s. Possible character of fusel oil. Clear, not cloudy and no emulsion.
Fraction 6: Slightly stale character. No volatile acidity. Nothing to distinctly identify it as vermouth. Stewed character.
Fraction 7: Very little character.
Fraction 8: Very little character.
It is possibly not worth collecting the last fractions which may speed of analysis.