Trehalose, fixatives, “rendering”, and the limits of re-distillation

One of the drinks I made for my nine rounds of high concept modernist cocktails presentation featured a special edition Yellow Chartreuse where the acacia honey was “removed” and swapped for pungent, full flavored dandelion honey. I had also done this in the past with Green Chartreuse where the non-aromatic white sugar was replaced with palm sugar jaggery creating a sort of white, coconut Chartreuse. I even went further by re-distilling Der Lachs Danzig Goldwasser and using it to fortify a strawberry liqueur.

This is really fun stuff but all the while I referred to it as a rendering and noted some limitations. One of the big limitations is that a small percentage of aroma is left behind in the pot when you re-distill, but why?

First I should point out that the above liqueurs I am manipulating are the product of distillation and not infusion with a small exception for the honey quotient of Yellow Chartreuse which is added after distillation. One of the visiting food scientists at the presentation contested this distillation assertion based on some of his experiences with a vacuum still. The assertion was based on vacuum distilling chartreuse and finding significant amounts of aroma left behind. His experience raises some questions that hopefully I can answer but the experience does not indicate infusion. Even if these particular liqueurs were infused, distillation would separate most of the volatile aroma. If significant non-volatile compounds were left behind in the pot, like gustatory-bitterness or gustatory-acidity, you would have an infusion on your hands but in each case there is nothing, just gustatory-sweetness from the sugar.

What we are both encountering is the fixative effect of the sugars added to the distillate which turns it into a liqueur. Sugar has a capacity to hold on to aroma compounds and keep them from volatilizing. Fixatives are a wild subject that I am only unraveling in tiny pieces. In the context of this modernist cocktail-centric distillation, the fixative effect is working against our rendering, but in many other cases the fixative effect can be used to our advantage.

A great paper, Trehalose Addition to Dehydrated Strawberry Puree by D. Komes et al., explains how food scientists put the fixative effect to work quite well. Trehalose is a miracle sugar that I had first heard about a year ago from a friend who is my science mentor. I sourced some back then but so far haven’t used it. Trehalose is the sugar that allows the desert to bloom after 50 years without rain. It was even added to cosmetics as a preservative before they figured out it was actually functioning as the active ingredient. I didn’t see at first investigation how trehalose would fit into my culinary inquiries. Trehalose is used in Modernist Cuisine (full text search tool), but not regularly and not with an explanation.

The linked paper is concerned with dehydrating strawberry puree. When you dehydrate the puree alone, near all the volatile aroma is lost. When you add sucrose before dehydration, significantly more aroma is kept and when you add trehalose instead of sucrose, an astonishing amount of aroma is kept. The volatile aroma just doesn’t evaporate with the water like you would think. The strawberry experiment parallels the re-distillation of liqueurs with significant sugar contents. Sugar holds on to some of the aroma during vacuum distillation as well as typical atmospheric distillation where the heat is dramatically higher.

Explaining the Chartreuse re-distillation phenomenon has exposed a whole new area of exciting inquiry. There must be countless fixative applications in pastry where trehalose would be helpful. I bet there are also applications for chefs trying to push the limits of dry aging meats. Can new vacuum/pressure techniques allow us to quickly apply fixatives like trehalose? (Modernist Cuisine has great pressure marinading advice) Trehalose is only recently economically viable and you will definitely hear more about it in the future.

TKO in 9 rounds with Bostonapothecary

Recently I put together nine rounds of modernist cocktails for a few visiting food scientists. Here goes:

1. Green Apple Soda.

Acmeapple soda

The first drink was the green apple soda which I decided to leave non-alcoholic because there were so many drinks. Carbonation rang in at 8 g/L which was quite bubbly. De-aeration with the champagne bottle manifold keeps the juice from browning which is the main gimmick. No ascorbic acid or pectic enzymes were added (not that I’m opposed to them). It is simply a way to show off the de-aeration concept in a fairly beautiful context. The apples were even juiced with an Acme centrifugal juicer which whips extra air in them which the magic of the manifold successfully removes.

2. Aged St. Valentine’s Day Sparkling Magnum

sparkling magnum

This drink first appeared on NYE 2012 but I served the batch executed for Valentine’s day 2013 which meant it was well over six months old and was showing well with no evidence of oxidation or loss of carbonation. The drink is proof that cocktails carbonated with the Champagne Bottle Manifold, when well executed, can be aged.

Per serving:
1.5 oz. Pacific Rim Heirloom Framboise
.5 oz. Blanco Tequila
.5 oz. Aperol
.5 oz. Lime Juice
1.5 oz. Water (dilution)

I did not finish off the magnum for the tasting and have been slowly serving glasses from it ever since with no problems de-aerating after every usage (days are elapsing between uses).

3. Bees Knees

For the Bees Knees I broke out the Tabasco aromatized gin and the Ames Farm single source Bass Wood honey syrup. For the gin, a commercial gin is simply re-distilled with Tabasco that first has had it’s volatile acetic acid (vinegar) neutralized with baking soda. The distillate is wildly fun but still fairly low involvement. It is not cut quite right so there is the faintest louche at 45% alc. and bottle condensation develops on the shoulders. The slight defects could be corrected by being more involved through executing more generations of the recipe. The Bass Wood honey syrup is scooped from the jar and mixed 1:1 with vodka to preserve it as well as precipitate some of the trace amounts of wax which can ultimately be removed with the centrifuge.

Bees Knees
1.5 oz. Tabasco Aromatized Gin
.75 oz. Bass Wood Honey
.75 oz. Lemon Juice

The overall goal of the drink was to synthesize the character of the rare and astounding Strawberry Tree honey of Corsica, Sardinia, and the Al Garve in south Portugal. This honey can smell redolent of chilies. I had been able to work with Corsican Strawberry Tree honey for many years but it has since been unavailable.

4. Special Edition Cherry Campari

I had intended to serve this as a Boulevardier but opted to only serve it on the rocks because there were so many drinks. Cherry Campari is pretty simple, the orange aroma is removed and replaced with the aroma of Kirsch. When the orange aromas are removed so too are the bitter aromas so they have to be replaced as well. It turns out olfactory-bitterness is very important to Campari’s identity. I made two versions which were cherry/wormwood and cherry/yerba-mate. The goal was to see how well they stood alone and then possibly blend them to create the most extraordinary tonal bitter effect. I still haven’t sufficiently explored all the blending options.

To remove the aromas, Campari is simply de-hydrated in an Excalibur food dehydrator. The Kirsch aroma is derived from Hiram Walker Kirschwasser re-distilled with the botanicals but I would like to explore simply compounding the Kirschwasser with a steam distilled essential oil. The Kirschwasser reconstitutes the dehydrated Campari but some gentle math has to be done to make sure everything returns to the original volume and alcohol content.

The results are subtle because the orange aroma of Campari is subtle. The same treatment can also be given to Cynar where I enjoy using slivovitz with quinine. There is a subtlety to replacing fruit aromas with fruit aromas because they are fairly convergent with expectations based on color and prior experience with the real deal Campari, but it might be exciting and pleasurable to pursue slight divergence by replacing the orange aroma with benzaldehyde-almond aromas taken from re-distilling an amaretto.

5. Satan’s Whiskers: an alliteration of echoing orange aromas, oh my!

sour orange

The most important theory in the Culinary Arts is that all creative linkage aspires to create a super normal stimuli. People are starting to study creative linkage within flavor but so far have not caught on to my theory nor come up with their own. They might benefit from learning a little more about the nature of attention from the great book, Slights of Mind, which is an excellent, edutaining, and accessible neuroscience title. I boil down the attainment of a super stimuli by the linkage strategies of alliteration and collage. My chosen example of alliteration is the Satan’s Whiskers poetically rendered in equal parts with a special guest appearance.

Satan’s Whiskers
.75 oz. Gin (inherently imbued with orange)
.75 oz. Sweet vermouth
.75 oz. Dry vermouth
.75 oz. Joseph König’s 19th Century Curaçao
.75 oz. Dominican sour orange juice
2 dashes Regan’s orange bitters.

Tonal nudging back and forth by the repetition of orange components creates a timbre of sorts and from the existing response tendency for orange the drink elicits an exaggerated response; Super Orange! The beetle mates with the more orange beer bottle (this phenomenon is so crazy).

The special guest mentioned above is the rendering of a 19th Century Curaçao which illustrates some of the secrets of the first grand cru liqueurs. Their sugar content was the maximum of solubility and so was their aroma content. The 55% alcohol orange liqueur was poured from a bottle with trace amounts of rock candy growing on the bottom because at 55% alc., roughly only 285 g/L of sucrose is soluble. This old style of liqueur also only had as much aroma as it could hold before it louched. I was slightly disappointed that the visiting food scientists were not familiar with the work of König who is considered to be the father of food science.

6. Final Ward

The most elaborate drink I made was my high concept version of Phil Ward’s Final Ward.

Final Ward
.75 oz. Over proof Overholt rye (55%)
.75 oz. Historically accurate Maraschino cheater
.75 oz. Special edition Dandelion Yellow Chartreuse
.75 oz. De-aerated 5 day old lemon juice

The Overholt was manipulated to remove the water, increasing the proof to 110 which I had detailed long ago. This rendering illustrates that a higher proof version of Overholt would be pretty darn cool. The Maraschino cheater was constructed from blending sugared & cut Hiram Walker Kirschwasser with sugared & cut re-distilled amaretto which is essentially how Maraschino liqueurs are made. I used proportions from old chemistry texts that reference bottlings from the early 20th century. I would love to deepen my involvement and use more historically accurate benzaldehyde (almond aroma) levels. A lot of great Maraschino data exists from 1912. The Dandelion Chartreuse was constructed by essentially removing the lightly aromatic Acacia flower honey from the chartreuse and replacing it with very full flavored Dandelion honey from Roero in Italy from the exemplary producer, Pozzolo. Dandelion honey is particularly sensual and earthy, quite distinct and unforgettable. Here I got into a minor argument with one of the visitors who was sure Chartreuse was in part made by infusions because of his vacuum distillation experiments. The Chartreuses are not made by infusion, but you cannot capture all the of aroma because of a fixative effect of the sugar added to the distillates. The sugar basically holds on to a small percentage of the aroma making it important to acknowledge that the new creation is only a rendering. I borrow the term rendering from poets that often translate works from dead languages. They take liberties, some degree of something is lost, but the results are still wildly fun. The lemon juice was simply de-aerated using reflux de-aeration via the champagne bottle manifold.

Flavor wise I thought this was the most impressive of all the drinks.

7. Collage

The counterpart to alliteration is collage and I first took my inspiration from the curious six equal parts Savoy classic, the Charleston. The Charleston is by no means a collage and rather uses three rhyming pairs, but it looked like it could get there pretty quickly.

My Collage
.5 oz. Mezcal
.5 oz. Kirschwasser (Hiram Walker)
.5 oz. Sweet Vermouth
.5 oz. Manzanilla Pasada Sherry (La Cigarrera)
.5 oz. Yellow Chartreuse
.5 oz. Plymouth Sloe Gin

Repeating aroma compounds can be highly engaging, attentional, and pleasurable but so too can barely repeating aroma compounds, but using quite a few. You can get comfortably wrapped up in a dizzying array of facets as easily as you can by witnessing the most beautiful overtone. The Savoy Cocktail Book is a great place to ponder the super stimuli via alliteration / collage creative linkage theory.

This drink is staggeringly delicious but I’m not sure if the visiting scientists enjoyed it or got the concept. One visitor was late so I served it for one group of guests 20 minutes before the other. Teasingly, I predicted the late comer would ask what vermouth brand I used and then be disappointed I served him Martini & Rossi… What started out as a jab at the NYC culinary scene played out too exactly. I got the question which hijacked us from the whole point. I thought I explained everything pretty elegantly but I guess it was a miss and he needed to bring it back to territory he was more familiar with. #fail

8. Marmite Rye Sazerac


This drink is really fucking cool and an illustration in some of the biggest concepts in distillation scaled down to a size no one previously thought possible. I’ve made this for years now and it keeps getting better and better as I deepen my involvement. It is definitely in the realm of acquired tastes and I don’t think it went over well with the visitors. What I hoped for was some sort of cute Anthony Bourdain style comment, “You bastard, that is devilish!”, “I’m a real Marmite slut”, or even “I didn’t want to like it but I like it”. Beverage people likely have accumulated more acquired tastes than food people and these were food people.

Marmite Rye Sazerac
2 oz. Marmite aromatized Rye (50%)
.5 oz. Simple syrup
4 dashes Peychaud’s bitters
rinse of Yerba Mate based anise/sloe berry Absinthe (70%)
expressed and discarded lemon peel

One of the big concepts in distillation is that aroma is created in the still. The main process here is esterification where fatty acids react with alcohols in the presence of heat to form esters. This process is mostly ignored in texts on beverage distillation because it can get pretty complicated pretty fast. I’ve slowly synthesized various writing on the topic and collected and annotated a bunch of lost Australian research papers that cover the topic in simplified experiments.

In my recipe, Marmite, a yeast concentrate high in fatty acids, undergoes esterification catalyzed by added non-volatile acids which is part of the emphasis on high total acidity in the wines of Cognac or the sour mash process. In this simple re-distillation, water and malic acid are added to a commercial rye with the Marmite. Time under heat is important to aroma formation so the added water allows distillation to take place slower, providing more time under heat for aroma creation. The minimum of energy is also applied to the boiler to make distillation as slow as possible. The added malic acid is a catalyst for the esterification of fatty acids and its addition sits in for the acidity that would be found when distilling wine or a grain based sour mash. Previously, you could only learn this stuff by playing with big batches at huge expense. My little experiment here allows you how to vary the parameters with standardized inputs at only $20 a batch (and enjoy drinking the results!). When you graduate to a big rig, the aroma creation processes at work will seem intuitive and involvement will be deepened much quicker. The rye was fake aged with my barrel bouillion technique.

The Absinthe used here is the updated form of a project begun long ago. A commercial Turkish Raki is the base because Turks are the masters of anise. Prunelle Sauvage or sloe berry eau-de-vie is added to increase alcohol and lengthen the aroma. The inspiration for the anise/sloe berry combo is the basque Paxtarian liqueur. The bitter aroma of wormwood is traded for the also bitter aroma of yerba-mate which is tonally darker. Being based on commercial spirits makes the recipe easy to construct on the nano-scale.

9. Something like an Alexander with Cashew Derived Heavy Cream

nut milkcolloid mill

One of the visitors is a serious nut milk enthusiast and the inspiration for my own nut milk adventures. I thought a great final drink might be to make something rich featuring a new nut milk idea that I don’t think anyone else has done before.

Inverse Alexander
1.5 oz. Overproof Overholt
1 oz. Pineau des Charentes
.5 oz. Wray & Nephews “Berry Hill” Pimento Dram
1 oz. Cashew Milk “Heavy Cream”

Sometimes I call these drinks inverse Alexanders because instead of featuring Cognac they feature Pineau des Charentes. The nut milk heavy cream is made by blending nuts and water 5:1 then dividing the volume in four and centrifuging. The fat rises to the top and can be collected and weighed. The water based quotient can then be collected and added to the fat in a ratio where the fat content is dramatically higher than a typical nut milk. I then ran the fat and milk through the colloid mill to homogenize it. Homogenizing seems to work fairly well until the nut milk starts to ferment and turn into yogurt. I think a change in pH (which I did not measure) starts to re-separate the fat. I have not investigated this heavy cream fermentation too deeply and have typically used everything immediately. Separation only happened after a few days. Who knows, a yogurt like product might even be better in a drink or the yogurt product could be migrated to a kitchen application. Pasteurizing might prevent the yogurt effect but I have not investigated further.

The drink is really extraordinary, less for the effects of the cashew cream and probably more for the creative linkage of the other ingredients.

widely used but maybe also widely taken for granted, aroma fixatives are mysterious

I’m going to try and assemble some literature about aroma fixatives.  One goal is to capture search terms about the topic.  I want to unite a lot of literature under one place so people searching for the topic might see many references grouped together.  To do this I’m going to use many quotations so feel free to skim them. I will try to bolden key terms to draw attention to patterns within all the sources.

Fixatives are strange and seem to be poorly understood and poorly described.  Fixatives are acknowledge mostly for lowering the volatility of aroma compounds to better preserve them.  They are then acknowledged for having an aroma that is only perceivable to some and not all, then being used typically sub recognition threshold.  Claims are made that fixatives can lower volatility while simultaneously increasing intensity of the aromas they preserve which leads me to speculate they may do something very unique at the level of perception.  Strange spatial language often gets attached to the effects of fixatives.  Some how they can bring aromas “together” and I suspect that people look to their effect on volatility to explain effects they might actually have on the spatial perception of aromas.  I speculate that fixatives might trigger modes perception where contrast enhancement is altered and odor object recognition might be more recollection heavy which could be said another way as they give us a tendency to complete aromas based on memories making them seem fuller and more complete.

The second time I ever encountered the term fixative was in the utterly brilliant book on chemistry, Molecules (p. 141) by Peter W. Atkins.  Atkins text seemed a unique entry in the bibliography of Neurograstronomy so I took a chance and gave it a read. In regards to fixatives from Atkin’s entry for Civetone C17H30O:

“Musk is used in two ways in perfumery. It is used for its odor, as a component of heavy, musky, oriental perfumes. It is also used as a fixative, sometimes in concentrations so small that its own odor is masked. That is, it is added to more volatile fragrances to retard their evaporation, so that they are experienced as a symphony of odors rather than as a sequence in order of decreasing volatility.”

This reference mentions the idea of the fixative being nearly sub recognition threshold, changing the volatility of the substances, and a spatial concept with a reference to the symphony.

The first time I ever saw the term was in Robert Léauté’s 1989 James Guymon lecture on Alembic distillation:

“The fatty esters give fruitiness to the Cognac; the fatty acids give body and are like fixatives for many other aromatic components; amino-acids are involved in thermic break down reactions.”

In this reference the fatty acid esters are only like a fixative, but we probably can’t put too much emphasis on the intention of that one word.  The specific esters Léauté is talking about are the most basic and abundant esters, ethyl-acetate, which are also only sub recognition threshold.  I can’t imagine the ethyl-acetate in a spirit is significant enough to have any effect on the volatility of the distillate.

Information is not abundant in web searches and among the only things notable are a few patent filings.

From a 1939 patent filing:

… “Ethyl alcohol, however, has the drawback of having a low boiling point and a high vapor pressure, hence when alcohol extracts are used in baking or candy making at temperatures for exceeding the boiling point of alcohol the alcohol used as the menstruum and solvent in the extract volatizes and carries off some of the flavor and aroma-imparting substance thus necessitating the use of considerably more extract than otherwise would be necessary if a solvent having a high boiling point were used.   The esters of glycerine, such as monoacetin and diacetin have been employed as solvents for flavoring extracts and although they are good solvents for most of the aromatic substances used in preparing flavoring extracts, these esters are susceptible to hydrolysis in the presence of water with the liberation of acetic acid which imparts a sour odor to the the flavoring extracts and renders them useless for culinary purposes.   I have found that the ethyl esters of glycerine and particularly the mono-ethyl-ether are valuable solvents in the preparation of flavoring extracts because they are perfectly stable, are miscible with water, they do not hydrolize, they have relatively high boiling points and therefore act as fixatives for the aromatic substances present in the extract, and they show exceptionally good solvent power toward the various essential oils and aromatic chemicals usually employed in the preparation of flavoring solutions. Said ethyl ethers of glycerine are likewise good menstruums when used undiluted or diluted with water for all kinds of vegetable substances such as vanilla beans, tonka beans, Phoenugreek seeds, roasted coffee, roots and herbs, all commonly used by flavoring extract manufacturers.”

Wow! That starts to explain things I’ve seen on so many ingredient labels over the years.  In this case no reference to a threshold is made, there is no spatial reference to the perception of the result. The fixative here is only used for its effect on volatility.

Another patent filing from 1981:

“A fixative for perfume compositions, 1-(2,6,6-trimethylcyclohexyl)-hexane-3-ol, itself odorless to many people, brings about a “rounding-off” of the perfume while increasing its intensity. The compound is made by condensation of citral with pentane-2-on, followed by cyclization and hydration.”   ….”It has been found that an addition of 1-(2,6,6,-trimethyl-cyclohexyl)-hexane-3-ol effectively brings about a “rounding-off” and fixation of perfume compositions while simultaneously increasing the intensity of the fragrance. The quantity used amounts usually to 1 to 25% by weight relative to the total composition, and a very distinct fixing effect can be ascertained particularly from approximately 3%. In principle, more than 25% may also be employed; this depends, however, upon the nature of the total composition and may be easily determined by routine experiments.”   “DETAILED DESCRIPTION OF EMBODIMENTS   The 1-(2,6,6-trimethylcyclohexyl)-hexane-3-ol itself is not to be considered as an odoriferous substance in the conventional sense as this saturated secondary alcohol has been found to be odorless, or only very slightly scented, by a significant number of the test persons who were asked to judge this substance, whereas other persons perceived an intensely radiating wood-like, slightly animalic-urinary odor.   Anosmia, i.e. the incapability to perceive smells, which numerous people have with respect to this new alcohol, may be compared with the effect of various very expensive animalic fixatives, such as musk and amber, whose scent can also be noticed only faintly by certain people, or not at all.   The occurrence of anosmia with respect to a specific substance, particularly in the area of perfumery, appears to always point out to particularly good fixative properties of a substance and holds true at least for all of the fixatives which have heretofore been employed in the manufacture of perfume compositions. These afore-mentioned advantageous properties of 1-(2,6,6-trimethylcyclohexyl)-hexane-3-ol distinguish this substance from the homologous products which have been known for a long time, such as iso-tetrahydromethylionol and n-tetrahydromethylionol, which do not develop any fixative effect and with regard to which also no cases of anosmia have been ascertained.”

Okay I’m getting a sense that people can make fixatives but not well articulate the mechanism of how fixatives work.  They might go beyond physical chemistry and volatility deep into perception.  Here we have emphasis on the sub threshold properties of recognizing the fixative. There is also lots of spatial language describing the fixative effect and somehow an increase in intensity despite a lowering of volatility? Mysterious! The description of this fixative does not emphasize volatility like the previous one does.

Yet another patent application from 2008:

“Compositions are provided that contain fragrance compounds and fragrance fixatives selected from C4-C8 alkyl terephthalates.”   “To slow the evaporation of fragrance compounds, the fragrance compounds are sometimes combined with lower volatility components generally known as fixatives. These fixatives are substances which improve lasting qualities of odorous substances of a fragrance. There is a continuing need for the development of new fragrance fixatives.”   “The present invention is based on the discovery that the use of C4 to C8 alkyl terephthalates in perfumes or fragrances can provide a more enduring fragrance by slowing the evaporation of fragrance compounds.”

Here, there is no mention of a sub threshold property or any spatial language, but rather only a mention of volatility.  But maybe there should be? The sweet-smell of phthalates was part of the olfactory illusion I encountered when drilling red plastic which brought on the illusory aroma of cherries.

From a really wild perfume industry blog post about fixatives from 2009:

“In perfumery there is something called a “fixative” which is used to elongate the life of the aromas as a whole and increase the life and hold the aroma of other oils which would otherwise just fade out.   Some fixatives can be synthetic or natural which include animal and plant derived fixatives. Almost all fixatives are used in minute quantities (except some synthetic ones) because of their powerful effects.   The four main animalistic fixatives, all of but one, hold up as very sweet smelling fragrances are:

*Musk (from the Musk Deer)

*Civet Musk (from the Civet)

*Ambergris (from the Spermwhale)

*Castoreum (from the North American Beaver)”

Really interesting! In this case volatility is the big emphasis, but “minute quantities” might also imply a desire to be sub threshold. The author also uses a few interesting terms.  The four natural fixatives are described as “animalistic” rather than animal derived but then “sweet smelling”.  For starters sweet smelling is a cross-modal metaphor and animalistic, if it is not only describing the source, might imply the olfactory-umami category.

I remember the paper on whiskey, Origins of Flavour in Whiskies and a Revised Flavour Wheel: a Review, by Lee, Paterson and Piggott mentioning a something along the lines of a fixative.  The paper is pretty astounding in its own right and the authors never mention the term fixative specifically but they do describe cyclodextrin bound reference standards for training whiskey assessors:

“The advent of cyclodextrin bound reference standards has enabled communication of information on flavour character in training of assessors, as exploited in the brewing industry.”   “For assessor training, reference compounds are recommended (Table I): formulations with cyclodextrins achieve consistency and parallel conceptualisations.”

The cyclodextrins in this case seem to function as fixatives yet the term is never used. This FAQ [pdf] describes cyclodextrins use as an “encapsulation agent” for food additives and being “virtually odorless” and “slightly sweet tasting” which fits the bill for a fixative.  A bar regular who is a medical research told me that cyclodextrins were known to have a stimulating effect but so far I cannot find any references to the phenomenon. I suspect not finding anything is a matter of language.