Birectifier Analysis of an Experimental Geotrichum Rum

Follow along: IG @birectifier

This was an extremely exciting case study I was honored to explore with the birectifier. This rum, which revives a dormant idea from the late 1930’s, was produced by a startup team in New Zealand doing exciting things with yeast. All of this is in an experimental stage and my understanding is that alcoholic fermentation was conducted with a turbo yeast but a secondary aroma focused fermentation was conducted with the mildew yeast Geotrichum fragrans (their own isolate which is always very exciting). This yeast does not produce ethanol but rather metabolizes proteins to produce aroma compounds. It grows on the surface of a ferment as a film.

Arroyo pioneered the use of a similar yeast, saprochaete suaveolens, which was featured in Studies on Rum (thanks again Simon!). Arroyo used his yeast on cane juice, but the New Zealand yeast was used on molasses. Mildew yeasts were previously thought to be molds but are now categorized as yeasts. The dominant ester described by Arroyo was ethyl tiglate which has an apple-y character.

Cory Widmayer successfully isolated a suaveolens early in our collaborative rum studies and we were able to get it to produce what is likely ethyl tiglate, but we weren’t able to get the one strain to grow fast enough to be viable. More examples need to be isolated to find a stronger strain. We also had no fresh cane juice. Using apple juice as a proxy, the aroma from suaveolens was almost strawberry-like and distinctly attractive. I had that limited frame of reference before exploring this rum.

[Suaveolens isolated by Cory in 2019 growing on apple juice]

Due to being experimental and produced on the small scale, this distillate, likely had some minor issues that accumulated experience will eventually work out. The aroma of the rum has a tufo character which is likely from distilling on the lees as well as leaving too much tails in the distillate. This is not a big deal because the birectifier can sort this aroma, isolating the high value character in fraction 5 and separating tufo related aroma in later fraction (7,8).

One way to think of experimental distillates that feature no or minimal cuts is that you are only looking to identify assets and liabilities. Your distillate is not the produce yet, but only proving the fermentation and justifying a different mode of distillation. A ferment could be tweaked in a lot of ways before fraction recycling and distillation proof need to be considered. What is observed via the birectifier reinforces this.

I evaluated every fraction one by one in the classic way and then was able to take what remained of the fractions and vat a new micro rum from only fractions 1-6 to get a glimpse of the rum minus the tufo character in 7,8. The micro vatted rum was quite unique and almost drank like a fruit eau-de-vie. Definitely intriguing and singular, it begs to be matured for a few months.

This rum definitely has high value aroma, but does seem out of balance and somewhat inharmonious. It could either become a unique blending stock or fermentation changed to add more dimension to it. Pretty remarkable that there was so much high value aroma from a co-ferment as simple as a well selected mildew yeast. There is quite a lot of potential here. This rum concept is very clearly worthy of more investment. I can’t wait to see it again in the future.


Fraction 1: Fairly light fraction 1. Nothing odd.

Fraction 2: Extremely light as expected.

Fraction 3: Neutral as expected.

Fraction 4: Possibility of slight fraction 5 character or it has a unique fusel oil mix? Lighter and less offensive than I thought. I thought this rum may be high in fusel oil.

Fraction 5: A lot of aroma, but it seems estery rather than rum oil. A lot of persistence. Pleasant on the palate and not acrid which may imply a level of aroma and not so much type of esters. Nothing that obviously screams ethyl tiglate, but I’ve only ever smelt it in an ferment and not a distillate.

Fraction 6: Possibility of character carried over from fraction 5 implying a lot of high value aroma in the overall rum or something particular with a unique high boiling point. No distinct gustatory acidity.

Fraction 7: No distinct gustatory acidity. Hard to pinpoint stale character. 7 seems slightly more intense than 8.

Fraction 8: No distinct gustatory acidity. Hard to pinpoint stale character. Nothing obvious that is tufo.

Stillage: Slight tufo-esque character. Faint vanilla character. I believe this distillate saw brief maceration with oak chips as an attempt at maturation. Detectable acidity which may have been fixed acids from the oak treatment. I did not attempt to measure it because I likely could not differentiate the distillate from the oak influence.

Very cool! Congrats!

3 thoughts on “Birectifier Analysis of an Experimental Geotrichum Rum

  1. Thank you for that Stephen. You have been a tremendous help.
    It is fantastic to see it broken down by the birectifier into each fraction.
    We are only just scratching the surface.

  2. I would be interested in fermentation notes.
    In home distilling circles turbo yeast isn’t highly regarded, as it tends to throw off a lot of undesireable congeners.
    Surely for evaluation purposes a more neutral yeast would be beneficial?
    Anyway, glad to see people experimenting with alt yeasts, I’m looking forward to what they do next.

  3. Hi Sean,

    I don’t have much for notes, but I think it was just a matter of convenience because the mildew yeast was ready to go. They’ve got a good plan to continue experiments and I just hope I’m allowed to look at the rums!

    Rum is going to get interesting. Fission yeast is the queen yeast, but I’m also excited about the kveiks and mildew yeasts. And then of course, there is so much bacteria to tackle!

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

search previous next tag category expand menu location phone mail time cart zoom edit close