Qualia Research Diary: Scents

[Epistemic Status: Diary Entries]

“Fake it until you deep fake it.”

― Joscha Bach

“Break often – not like porcelain, but like waves.”

― Scherezade Siobhan

“Ideology has two meanings- actually, most social terms have two meanings, one for the traumatized and one for the non-traumatized.”

― Michael Vassar

“You know the old adage about monkeys typing into infinity, and the question about whether they would eventually produce Hamlet? I think that maybe we are those monkeys, and we’re producing countless Hamlets every single day.”

― Jacob Stephen

“Reality is very weird, no doubt. At the same time, it is easy to get wrong ‘what kind of weird’ reality is.”

― Matthew Barnett

“It is not true that suffering ennobles the character; happiness does that sometimes, but suffering, for the most part, makes men petty and vindictive.”

― W. Somerset Maugham

December 13th 2019

In a different timeline, I open a high-class experimental qualia-focused restaurant. There is only one kind of meal every month, and it is a challenge to finish it. Only 10% of people manage to do so. On March of 2022, the menu consists of:

  1. A soup. A liter of (tap) water with a single mint leaf in it. Do not be deceived, this is not “spa water”. The amount of mint in it is exactly right below the perceptual threshold for the most discerning of tasters. Hence, you are guaranteed to (a) not be able to taste anything at all, while (b) fully knowing you are indeed drinking aromatic molecules from the mint leaf. Also, they give you a spoon and a straw. If you use the straw, you are “drinking your soup” while if you use the spoon you are “eating your soup”. Up to you. It’s a conceptual piece after all. Once -and only once- you finish it, they serve you the second course…
  2. There are aromas and flavors out there in the state-space of qualia-triggering molecules that cancel each other out perfectly. The second course consists of a series of small hors d’oeuvres that are completely tasteless. If you can taste anything- e.g. a hint of garlic, or orange- it means the chef didn’t prepare it well. The flavors need to be perfectly balanced for them to be entirely tasteless. And once you are done, they bring you…
  3. This thing they left on your table is akin to a wire puzzle, or one of those Hanayama pieces. They tell you that your third course consists of a tiny cookie hidden inside it. Average solving time: 25 minutes. 50% of people can’t solve it.
  4. You are given a miniature 3D printed sugar statue reconstruction of someone who shares your name (as close as possible). Before eating it, you have to scream “There can be only one!” and consume your namesake in a single bite.
  5. Trace minerals. They bring you this large metallic bowl with a tiny little bit of powder at the bottom; certainly no more than 50 or 60 milligrams of material. It contains half of your daily recommended dose of iron, manganese, copper, iodine, zinc, cobalt, fluoride and selenium. You can now finally know what these actually taste like. It turns out that the characteristic taste of your grandma’s famous tapioca dish was zinc. Moving on…
  6. Negative food. You donate 300ml of blood.
  7. Distilled saliva. You spit in a bowl a number of times. You are then given a little shot of perfectly tasteless and clean water. The water is chemically pure. However, it is the water in the saliva of the spit of another customer.
  8. Double blind taste experiment. You are given a dish. The waiters do not know what it is. You do not know what it is. You have to write a 100-word report of what you think about this dish. This is actual science; the data is used by a research lab at some undisclosed university. There is something very Buddhist about this course – how much does your top-down model of what you are eating modify your perception of it? What if you do not assume an “essence” behind it – block that specific energy sink from robbing you of the experience of raw low-level sensation?
  9. Sound control. Did you know that food tastes different in an airplane? Many factors contribute to this, but a major one is the constant background noise you can hear inside the aircraft. Turns out tastes change with specific sounds. The Qualia restaurant spent $500,000 dollars researching this (and publishing a number of peer-reviewed papers in the process). The output of that research is that you can now make chocolate taste like vanilla, and strawberry taste like melon – if only you play the proper sound at the right volume. And finally…
  10. Stroboscopic taste – you put on a mouthpiece that entrains half of your tongue to a 30Hz electric seizure vibration while the other half is entrained to 17Hz. As you eat the Ice-cream of Victory (flavored with passionfruit, peanut, and anise) you realize that the flavors combine with the stroboscopic stimulation to create the hallucination of an entire meal replete with much more complex flavors. The beat patterns are tasty.

If you finish the entire thing (which usually takes about 5 hours total) they take a photograph of you and “keep it to themselves”. No, there is no “victory board”. They just want a picture of you.

5/5 | Would recommend.

January 6th 2020

Favorite essential oils at the moment: Freesia, Violet, and Pear. It turns out Freesia was a predominant note in Dior “Addict 2“, a perfume I fell in love with when I was a teen. Violet is “ethereal” in that it feels strangely anesthetizing (the ketamine of smells). Pear is lovely.

High Entropy Alloys (HEAs) and Scents:

  1. Some scent combinations “collapse categories” (e.g. too many flowers combined blend into “generic flowery”).
  2. Others make unstable multi-phase blends (e.g. too many categories – spicy, citrus, minty, woody all at once).
  3. Violet + Pear create a scent HEA.

An interesting blend with “emergent” characteristics: Freesia, Pear, Violet, Sunflower, Azalea, and Patchouli. Very high valence mixture that has a novel feeling that does not seem to come from the ingredients. #HighEntropyAlloy #HighEntropyScent

January 8th 2020

Careful with raising the “scent entropy” too high!

In sound and sight, it seems that there is an inverted U curve relationship between stimuli entropy and the entropy of the experiential response. White noise may be- objectively- the way to cram in as much information as possible into a waveform. But perceptually, white noise is more like its own (neutral valence, indifferent) tone. Likewise visually, if you crowd your images way too much you can’t actually understand its meaning and true complexity. Perceptual complexity response is maximized in the middle, where you achieve “peak useful entropy”.

More so, extremely entropic stimuli can be used to “mask” any input by adding a dose of white noise or visual static. That’s how you can degrade the valence of something when you don’t know what kind of unpleasant input you will get in advance. White noise drowns out both construction sounds and baby screams. It’s a “universal diluter”, so to speak.

And so it seems that this is the case with smells too. If you combine any 40 (42?) scented molecules that are as different as possible, you get as a result a generic smell with neutral valence that is not distinctive at all. If you make a different 40-scent mixture with completely different molecules, it also smells the same! They call it white noise scent, or “Laurax”*.

In other words, the “high-entropy alloys” of smell may only really pay off in the range of 5 to 15 different molecules, where (perhaps) we maximize the experiential “character” of the resulting fragrance.

Now, of course commercial perfumes in practice do have dozens if not hundreds of aromachemicals. But their absolute “scent entropy” is probably not that high. Why? First, the entropy is reduced by the fact that most perfumes do concentrate on a few core notes; the many other notes are usually small additions and tweaks. And second, the perfumes are usually made with relatively few categories of smells blended together (musky, citrus, and flower could be one, or green, ozonic, and non-citrus fruity another, and so on). Additionally, to get true white noise smell you need to also add negatively valenced scents, which are rarely used in actual perfumes. I do wonder, though, if the perfume industry has a sense of the “scent entropy” of their various perfumes, and if having a measure of it would perhaps improve their ability to hone in on blends that have unique emergent characters without relying entirely on heuristics and trial and error. Or how about a portable “scent-entropy-o-meter”? I bet it would find some very useful applications.

[Good article about it: The “white noise” of smells; *I first learned about Laurax here: The whiff of white could hide strong odours: Complex mixtures of many odours tend to smell the same.].

January 10th 2020

Of all the industries, I get the impression that the perfume industry is ahead of the curve when it comes to incorporating hedonistic utilitarian notes into its embedded ideology.

January 11th 2020

Cilantro tasting like soap to 10% of the population is just the tip of the iceberg.


January 13th 2020

What are your favorite perfumes?
(and if it’s not impossible to describe – why do you like them so much?)

I’ll start:

Addict 2 by Dior
Eros pur femme by Versace
Light Blue by Dolce & Gabbana

Oh god, what kind of person have I become?

January 14th 2020

Scent combinations with unusual emergent characters that are “more than the sum of their parts” I have discovered so far:

  1. Violet + Pear
  2. Rose + Orange
  3. Honeydew Melon + Pomegranate
  4. Freesia + Golden Hydrangea

In each of these cases, combining in roughly equivalent intensities (i.e. 50-50 ‘equipotent’ mixtures) seems to give rise to qualities that are not present in either of the two scents. This is relatively rare, IMO. If you combine, e.g. lilac and jasmine, you just get something that smells like “lilac and jasmine”. But the four combinations above seem- to me- to give rise to new exotic qualia varieties.

An accord is more about getting rid of the individually distinguishable component scents. The end result, however, is one of a “generic” scent within a given category (or subcategory). For example “white flower accord” or “citrus accord” are common. And although you can distinguish between two citrus accords, they don’t really have unique character – at least not more than e.g. various kinds of brown noise have a unique character. The combinations I’m mentioning are not just ways of creating a category blend so that other elements of the perfume can be more noticeable. Rather, they are on their own uniquely characteristic, much like other pure essential oils.

If you mix a wide enough variety of flowers you inevitably get a flower accord. To get a new qualia type emergent you need something else. (I should add I’m new to the field and have a lot to learn).

I’m developing a way of explaining what a scent is like at a glance with relatively few parameters. One of them is category entropy, meaning how close a given category in the scent is to the maximally blended version of it (i.e. a fully generic “flowery” scent has maximum category entropy).

Then another parameter is the “global entropy” which describes how close the scent is to total white noise scent.

So we start by saying e.g. perfume X is “50% of the way to white noise scent and its distribution of core categories is 30% woody, 30% floral, 20% fruity, and 20% citrus”, then we zoom in to each category and describe its category entropy and salient notes: “the floral entropy is 40%, and the 60% remaining is shared in equal measure between rose and azalea” (repeat for each category).

Additionally, another important thing to add is if there are “note to note interactions”, which in my (limited) experience happens with some pairs. Maybe 10% of them, but I don’t know for sure. But you could describe them with lines between individual notes in a diagram. To round it all out, you also would point out the note accords that work as “phases” in the overall scent (drawing inspiration from high entropy alloys – an alloy that does not make a single crystal structure is called “multiphasic”). E.g. mango + patchouli + cinnamon + jasmine tends to produce two phases, a mango + cinnamon phase that toggles in your attention with the jasmine + patchouli phase. Finally, we would also note “valence inversion” effects that happen when there are combos of scents that when placed together give rise to a flipped valence (also a rare effect, IME).

For a slightly higher level of resolution, we would break down each category into subcategories and then describe the entropy of each. E.g. a floral perfume could be 80% of the way to maximum floral entropy in the “white flower” subcategory but only 10% of the way to maximum entropy in the “powdery flower” category.

This would allow us, I think, to put our finger on many scents that are hard to describe otherwise. Indeed, a lot of sophisticated perfumes, IMO, are playing a lot with different shades of high entropy, so talking about them in terms of notes like jasmine or amber is very misleading. It’s like calling a certain kind of brown noise “closest to a guitar sound” because one lacks words for describing noise profiles.

January 23rd 2020

Scent Factorization:

So we know that we can get “white noise smell” by combining 42 scents of completely different kinds at the same time. This maxes out the “scent entropy” (aka. “Laurax”).
If you combine 42 different flower scents, however, you get a maximally generic “flowery scent”. I call this “category collapse”.

Now some scents have what I call “special effects”, which are category-neutral qualities. An example is the ‘bitterness’ of grapefruit, which although is often associated with fruits, can occur in entirely different categories too.

So I thought: what if we try to combine scents from as many categories as possible that all share the same special effects? I call this “scent factorization”. Namely, you try to get “special effect + Laurax” by canceling out everything but the special effect.

I believe this actually works. Example:

A factorization of “bitter-sweetness” can be obtained by mixing:

Grapefruit + Geranium + Bergamot + Pomegranate + Cedar-wood

In this case you will see that geranium is almost like “the grapefruit of flowers” in that it is flowery in nature but still shares the same “bitter” quality as grapefruit (albeit at a different frequency – yes scent frequencies are a thing, but that’s a story for another time). Likewise, cedar-wood is the most grapefruit-like wood I’ve smelled.

Another interesting factorization is that of “creaminess”:

Coconut + Fig + Vanilla + Almond + Sandalwood

In this case, again, you’ll see that sandalwood is the most “creamy” of all woods (as far as I have tried), fig is the most creamy of all fruits, and so on.

But this is just the start. What other scent factorizations could we try? I’d say we could aim to have the special effects of “ozonic”, “green”, “ethereal”, “powdery”, “acrid”, “cloying”, and so on factorized. Each deserves to become its own perfume in my up and coming new line of high end perfumes called “The State-Space of Scents” (for the consciousness connoisseur).

February 2nd 2020

The Qualia Review – Episode 1: Women’s Perfumes (Part 1):

The Qualia Review – Episode 1: Women’s Perfumes (Part 2)

The Qualia Review is a tongue-in-cheek program where you will get non-expert opinions about the quality of experiences by people who really care about consciousness:

In each episode, Andrés Gómez Emilsson (qualiacomputing.com) reviews a particular qualia variety (i.e. category of experience) with a co-host (in this episode Victor Ochikubo).

In this first episode we review women’s perfumes. In particular, we review (from worst to best):

La Panthére by Cartiere (EDT)
By Invitation by Michael Bublé (EDP)
Guilty by Gucci (EDT)
Brit Rhythm by Burberry (EDT)
Jolie Fleur Bleue by Tory Burch (EDP)
Rose Goldea by Bvlgari (EDP)
Daisy Love by Marc Jacobs (EDT)
Valentino by Valentino (EDP)
Amazing Grace Ballet Rose by Philosophy (EDT)
Light Blue by Dolce & Gabbana (EDT)
Eros by Versace (EDT)

You will notice that this is unlike any other review of perfumes. This is because the review here provided addresses the following three aspects of scents:

  1. A qualia-focused account (i.e. entropy, categories, special effects, etc.)
  2. What kind of person would enjoy wearing this perfume (mood-congruence, personality, etc.)
  3. The social signaling that the perfume entails (sexual signaling, genetic fitness signaling, etc.)

In particular, (1) describes scents in terms of:

  • A) The global entropy (e.g. 40% of the way to white noise scent)
  • B) The within-category entropy (e.g. 70% of the way into ‘generic flowery’)
  • C) The individual notes that can be detected within each category (e.g. non-generic jasmine note being 30% of the flowery category)
  • D) Lines connecting notes that have non-linear interactions (e.g. pear & violet, rose & orange, pomegranate & honeydew make unique blends that have phenomenal properties unlike those of the individual ingredients)
  • E) Lines connecting notes that form separate “phases” across categories (e.g. with a mixture of mango, sandalwood, rose, lemon, and cinnamon you get three phases rather than a global consistent smell – mango + cinnamon, and lemon + sandalwood, with rose staying its own distinct scent)
  • F) Lines connecting “valence inversion” effects (some notes simply don’t seem to go together even though they are pleasant individually)
  • G) Special effects (e.g. “powdery”, “ethereal”, “acrid”, “creamy”, etc.)

Thus, we share an entirely new angle on how to describe the ineffable. Namely, the hard-to-put-your-finger-on elusive subjective quality of scents can finally be grounded in terms we can all understand (with a modicum of shared background assumptions).

Hope you enjoy! Happy scent qualia!

~Infinite Bliss~

February 5th 2020

Three scents that are surprisingly similar to strawberry (based on my personal experience with essential oils):

  1. Fig
  2. Freesia
  3. Peony

In fact, following the “scent factorization” concept – if you make a mixture of these three scents the resulting oil smells almost exactly like strawberry cake. Strange!

February 9th 2020

I love this video! The idea that the information content in a perfume could possibly fit so much phenomenal detail is enticing, albeit perhaps a bit optimistic.

In the interest of honesty, out of the 15 or so women’s perfumes I’ve experienced deeply so far, La Panthere by Cartier is the worst by quite a long shot.

I don’t mean this to troll! I am serious. I still don’t quite know why I feel it as so unpleasant. I think it has to do with its very high entropy quotient, and the fact that it centers around gardenia, which is my least favorite flower. It feels predatory – and perhaps the perfumist did succeed at telling a story. Too bad I aim to reprogram the biosphere so that predation is a long-forgotten nightmare of our ancestral Darwinian environment of adaptedness. So long! We should aim to transform scent exploration from its current state of commercialism mixed in with weapons of sexual conquest, and push it into new frontiers… the exploration of the state-space of consciousness, valence research, perhaps even energy parameter modulation! The future of scent qualia research is wide open.

The Qualia Review – Episode 2: Men’s Perfumes

The Qualia Review is a tongue-in-cheek program where you will get non-expert opinions about the quality of experiences by people who really care about consciousness:

In each episode, Andrés Gómez Emilsson (qualiacomputing.com) reviews a particular qualia variety (i.e. category of experience) with a co-host (in this episode Victor Ochikubo).

In this second episode we review men’s perfumes. In particular, we review (by order of appearance):

CK2 by Calvin Klein (EDT)
Pasha de Cartier Edition Noir by Cartier (EDT)
Virtu by Vince Camuto (EDT)
21 Le Fou by Dolce & Gabbana (EDT)
Le Male by Jean Paul Gaultier (EDT)
Scuderia Ferrari Light Essence Bright by Ferrari (EDT)
Jimmy Choo Man Blue by Jimmy Choo (EDT)
1 Million by Paco Rabanne (EDT)
Terre D’Hermes by Hermes (EDT)
Invictus by Paco Rabanne (EDT)
Bleu De Chanel by Chanel (EDP)

In this episode we also discuss the way in which an enriched conception of art could helps us reformulate the artistic potential of perfumes. We make allusions to the 8 models of art discussed in a previous video:

Harmonic Society: 8 Models of Art of a Scientific Paradigm of Aesthetic Qualia

See also:

Harmonic Society

Hope you enjoy! Happy scent qualia!

~Infinite Bliss~

February 10th 2020

Top 5 Male Perfumes:

  1. Bleu de Chanel (EDP)
  2. Scuderia Ferrari Light Essence Bright (EDT)
  3. Le Male by JPG (EDT)
  4. Nautica Voyage (EDT)
  5. 21 Le Fou by D&G (EDT)

It’s very sad that there is a huge paywall for scent qualia. It’s your birthright to know what they smell like!

February 11th 202084357695_2785896804835792_4296261472725499904_o

~120 essential oils and ~40 perfumes (ordered by categories and general character).

This is the dataset my brain has been training over to interpret the state-space of scent qualia for the last month and a half. This is still amateur level – but I can nonetheless confidently say that I now understand scent qualia at least 50% better than I did last year.

I would still appreciate specific suggestions for essential oils or perfumes to get that are very unusual or characteristic. I continue to be surprised by the uniqueness of oils, fragrances, and mixtures I haven’t tried before.

Also: drastic income inequality is a massive tragedy, no doubt. But why are people not talking about qualia inequality? I wish everyone was as qualia-rich as I am right now. I’m happy to share some scents with people who feel qualia-deprived; just come to the Bay and give me a call. 🙂

Ps. Peony is an incredibly versatile low-entropy flower scent with a creamy strawberry-like effect. I kept reading about how this or that perfume has peony in it, but it really took me owning an essential oil of it to grok the type of qualia peony is all about. Someday there will be a monument built to celebrate the qualia variety disclosed by peony formulas. I’m pretty sure of this.

February 14th 2020

People say “a blind buy” when they talk of buying a perfume they haven’t smelled. Shouldn’t it be more appropriate to say an “anosmic buy”?

February 18th 2020

In order to survive the apocalypse, having a “blue” fragrance on hand will become very useful. I suggest “Nautica Voyage“.

You can thank me later!

February 21st 2020

Sense of Smell is Linked to Sexual Orientation, Study Reveals

Very interesting! Two followup questions: (1) does it replicate on a larger sample size? and (2) is the baserate of different sexual orientations of anosmic people statistically different than those of the general population?

Gay men showed a strong preference for the body odour of other gay men in the scientific test of how the natural scent of someone’s body can contribute to the choice of a partner.


Although previous studies have shown that body odour plays a role in making heterosexual men or women attractive to members of the opposite sex, this is the first study that has investigated its role in sexual orientation. Charles Wysocki of the Monell Chemical Senses Centre in Philadelphia, a non-profit research institute, said the findings underline the importance of natural odours in determining a sexual partner whatever the sexual orientation of the person involved.


“Our findings support the contention that gender preference has a biological component that is reflected in both the production of different body odours and in the perception of and response to body odours,” Dr Wysocki said.

February 25th 2020

Review of Shalimar Eau de Parfum by Guerlain for women:


February 27th 2020

Jasmine, Tuberose, and Gardenia: the Dark Triad of White Flowers. Beware! They are treacherous, envious, and guileful. DO NOT TRUST. They will ruin your perfume with their high-entropy indolic ‘broad spectrum scent noise’. Deranged, distracting, and disingenuous. #FlowerProblems

March 12th 2020

Why you should not insufflate ketamine: (1) it can irreversibly damage your bladder and cause very serious untreatable chronic pain, (2) it can damage your liver, also very painful, but above all (3) it will slowly degrade your ability to experience scents! Not worth it IMO!

Cocaine is well known for causing anosmia in regular users. I suspect we are going to see a wave of anosmic people as ketamine becomes more popular. Don’t be a victim. “Remember kids, don’t insufflate drugs – either eat them or inject them” would be my DARE go-to phrase.

March 16th 2020

Running out of hand sanitizer but you are fab and have a perfume collection? Use some cheap perfume instead! It’s usually 70+% alcohol.


March 22nd 2020

There’s An Unexpected Loss Of Smell And Taste In Coronavirus Patients

Factoring in the loss of precious qualia would make this epidemic even worse. This year I’ve finally begun appreciating the state-space of scents. I’m heartbroken to learn about this effect. So much qualia in potentia that might be lost!

March 23rd 2020

We should emphasize the possibly of life-long loss of smell in order to get more young adults onboard with strict social distancing measures. A 20-something person might not fear a fever, but they may fear “having less sexy sex and enjoying food less for the rest of their lives”.

March 26th 2020EUF6KWvUEAIAbO5

Sense of smell over the years. People under 40: please do yourself a favor and get some nice scents so you enjoy them while you are still sensitive to them. It’s always a tragedy not to use a qualia variety and then lose it. #qualia #scent #aging #valence #bliss #WeAreTheQualia

March 29th 2020EUQLBqVUEAA3Nwh

This is the future – in 2010 I was saying that in the long run humanity will need to adopt entirely new and seemingly extreme measures against contagious diseases.

Nasal filters (aka. “nose condoms”) were one of the ideas I was considering at the time. Reality is now catching up with fiction.

Why adopt extreme measures? Because we haven’t seen anything yet. The possibility of rational virus design and the political will to invest in innovative weapons means that sooner or later we will encounter things with a case fatality rate > 80% and R0 > 4. Nothing short of large-scale contact network engineering and the widespread use of tech like nasal filters can really work against those long-tail risks.

Perhaps in the future going out without nasal filters will be considered as reckless as today it’s considered having unprotected sex with a random stranger. #NasalFilter #TheNewMask #PM2point5

April 8th 2020


Bright Neroli

Summer 2020 Unisex Perfume Recommendations:

1. Bright Neroli – Ferrari (amazing sharpness and cute Sicilian dry-down)

2. Monserrat – Bruno Fazzolari (incredible grapefruit punch and bitter-sweet resonance)

3. Born – Adidas (a cheap but highly rewarding lavender rhubarb scent).

April 21st 2020

Haven’t posted about scents in a while; I’m still actively researching this fascinating qualia variety (better do so while I still have scent qualia, which may of course go away if/when I acquire COVID-19).

I’ve developed a lot of new vocabulary to talk about scents. In particular, I like to break down a scent in terms of entropy (how close to ‘white noise scent’ it is), category distribution (% woody, citric, fruity, etc.), category-specific entropy (e.g. 70% of the way to ‘generic flowery’), specific notes (e.g. 10% rose), and of course, “special effects” (such as “creamy”, “powdery”, “bitter”, etc.).

A recent “special effect” I’ve explored is the rather peculiar feeling that the scent is “flammable”. For example, gasoline has it, and so does ethanol. It is similar to the feeling you get when you inhale nitrous oxide. A kind of fascinating gas-like intoxicated state that produces spatiotemporal confusion and a sense of resonance. Of the scents I currently have access to, 100% pure Neroli essential oil strongly triggers this particular special effect. Neroli has that strange “flammable” quality, perhaps an octave or two in pitch higher relative to gasoline. It’s equally enthralling as the smell of gasoline (for those who like it) but much more dinner-party-friendly.

Anyway, with this “flammable” special effect in mind, I’ve been exploring what can be added to it in order to create beautiful scents. Last night I found a combination that made me really happy. It consists of equal (intensity-adjusted) parts of:

  1. Neroli oil
  2. Orange essential oil
  3. Lime essential oil
  4. Pear essential oil

It is sweet, sour, and gasoline-like in an unexpectedly euphoric way. I highly recommend this quale. I very much like its vibe. Meet me there.

April 28th 2020



First I tried essential oils. Then I tried perfumes. Now I’m entering a third phase in my “scent literacy” journey: pure molecules.

I have 50 pure perfume ingredients in an air-tight container now. And I have been trying out a couple each day in a systematic way in order to map out the state-space of scents.
One core insight so far:

Essential oils are extremely rough approximations for “building blocks” of scents. Perfume notes are often described in terms of fruits, woods, flowers, animalic sources, etc. But “apple” is not a natural unit of scent qualia. Although there is a general “apple vibe”, in reality that vibe can come from any of 20 or so different molecules. Additionally, many molecules that have an apple vibe do not even appear in biological apples (and vice versa). I’ve so far tried two apple-vibe molecules:

  1. Alpha Damascone: The smell of a dried out green apple, slightly past its prime, unsweetened and with trace amounts of beeswax wrapper stuck to its skin.
  2. 5-octen-1-ol: The smell of extremely mild refrigerated apple sauce, slightly waxy, reminiscent of sandalwood, and at a slightly higher “phenomenal frequency” than damascone.

In other words, I’m learning that pure molecules are indeed more “simple” than essential oils by far. They feel very specific and low-dimensional rather than voluptuous and scenic. But despite their relative simplicity, they are still not “categorically pure”. A single molecule can smell woody, fruity, and camphorous all at the same time. Part of the story is likely that a single molecule can have a broad spectrum of receptor affinities. But even if only one scent receptor were to be activated, perhaps the resulting experience would also not be uni-categorical.

The fascinating implication here is that scents that feel very uni-categorical (e.g. pear essential oil being unequivocally “fruity” with no hint of floral or woody) are more likely to be compositions of many molecules!

Each uni-categorical accord is made by mixing many molecules that all share the same “main vibe” but have different “secondary traits”. This way the accord lets the secondary traits “cancel out in white noise scent” while the main vibe is additively compounded into a broad-spectrum power-punch of a single category, like fruity (reminiscent of “scent factorization”, which I’ve described in previous posts).

May 2nd 2020

You don’t need to be phenomenally rich in order to be phenomenally rich!

I’m an advocate of high-dose behavioral enrichment (I talk about it at 22:16):

May 3rd 2020

The Perfect Scent excerpt:

Ellena will dip a touche into a molecule called isobutyl phenylacetate, which smells vaguely chemical and nothing else, and another into a synthetic molecule whose common chemical name is ethyl vanillin. (A rich gourmandy vanilla molecule, its IUPAC name is 3-methoxy-4-hydroxy benzaldehyde, and it is the heart of Shalimar.) He puts the touches together and hands them to you. Chocolate appears in the air. “My métier is to find shortcuts to express as strongly as possible a smell. For chocolate, nature uses 800 molecules, minimum. I use two.” He hands you four touches, vanillin + natural essences of cinnamon, orange, and lime—each of these has the full olfactory range of the original material—and you smell an utterly realistic Coca-Cola. “With me,” says Ellena, “one plus one equals three. When I add two things, you get much more than two things.”


He will hand you a touche that he has sprayed with a molecule called nonenol cis-6, which by itself smells of honeydew melon or fresh water from a stream. He’ll then hand you a second touche with a natural lemon on it, direct you to hold them together now, and suddenly before you appears an olfactory hologram of an absolutely mesmerizing lemon sorbet.


The explicit point was not to create a thing but an illusion of that thing, an olfactory alchemy. The point of Nil was not to create a green mango but the illusion of a green mango.




Junior perfumers discover that Vetiver Huile Essentielle from Haiti smells like a Third World dirt floor and Vetiver Bourbon from Isle de la Réunion smells like a Third World dirt floor with cigar butts. (They hope to do something wonderful with the cigar butts.) They learn, as Ellena knew from decades of work, how to create the illusion of the scent of freesia with two simple molecules, both synthetics: ionone beta + linalool. And orange blossom: linalool + anthranylate de methyl, which by itself smells like aspirin. The classic Guerlain perfumes often used a molecule called styrex, which smells of olive oil pooled on a table in a chemical factory. Add phenylethylic alcohol and you get lilac. Add the smell of corpse (indoles), you get a much richer lilac. And you can give your lilac, freesia, and orange blossom a variety of metallic edges: Add allyl amyl glycolate, you get a cold metal freesia. Add amyl salycilate, and you get a freesia with the smell of a metal kitchen sink dusted with Ajax powder. Aldehyde C-12 lauric adds an iron with a bit of starch still on it.

May 8th 2020

Excerpt from Luca Turin and Tania Sanchez’s 2008 perfume guide:

Sports Fragrances:


The last decade has seen the unfortunate flourishing of a dismal genre, the fragrances for men and women who do not like fragrance and suspect that none of their friends do either. The result has been a slew of apologetic, bloodless, gray, whippet-like, shivering little things that are probably impossible, and certainly pointless, to tell apart. All fragrances whose name involves the words energy, blue, sport, turbo, fresh, or acier in any order or combination belong to this genre. This is stuff for the generic guy wishing to meet a generic girl to have generic offspring. It has nothing to do with any other pleasure than that of merging with the crowd. My fondest hope is everyone will stop buying them and the genre will perish. Just say no.


Lastly, and by way of contrast, remember that perfume is foremost a luxury, among the cheapest, comparable to a taxi ride or a glass of bubbly in its power to lift the mood without causing subsidence the morning after. Wear it for yourself.


– Luca Turin in PERFUMES: THE A-Z GUIDE (2008)

May 13th 2020

The perfume Tommy Girl just registered as an outlier to my nose. It registers as high in valence as Bleu de Chanel and Bright Neroli by Ferrari. Extraordinary perfume. 10/10 #ScentQualia

May 27th 2020

The Rainbow God Experience

One of the most interesting lines of evidence pointing in the direction of the Symmetry Theory of Valence is how in the neighborhood of the peak of high-energy neural annealing events one can often glimpse states of consciousness with a characteristic “full-spectrum of qualia” property.

This may happen nearing the peak of a strong LSD trip, during intense Jhanic concentration, Fire Kasina practice, or even just spontaneously (though extremely rarely).

At the actual peak of the annealing process you are likely to arrive at a “moment of eternity“- itself extremely high-valence- where the symmetry is so complete that it becomes impossible to distinguish between self and other, before and after, or even left and right (this is a phenomenal property of peak valence states, and not proof of Open Individualism and non-duality per se, even though most people tend to interpret such experiences that way).

The “Rainbow God” phenomena lives at the edge of such peak valence states.

Timothy Leary in “The Psychedelic Experience” says that as you approach the highest bardo you are given the choice between “tasting sugar” and “being the sugar”.

The former is close to the peak of the annealing process, where there is enough asymmetry in the state for you to be able to encode information and distinguish between past and future, self and other, etc. and thus able to experience a projective world-simulation and the illusion of a self that “experiences it”. At the top of the annealing process, however, the extreme symmetry does not allow you to do that. The valence is almost certainly higher, though the degree of consciousness is arguably lower. You are “the sugar” rather than “tasting the sugar” (i.e. you are luminosity rather than a constructed world-simulation “experiencing luminosity”).

Stunningly, this edge between perfect symmetry and its surroundings in configuration space often shows extreme levels of qualia diversity. This is an empirical observation you can verify for yourself (or you can trust me, find others who have experienced it, or derive it from first principles).

What is it like? At this boundary between quasi-perfect symmetry and perfect symmetry you experience rainbows with all the phenomenal colors in the CIELAB color space (and perhaps some other colors that you only see in heaven, like blue-yellow and red-green, which require enough energy to overcome the lateral-inhibition opponent process going on in the cortex at all other times). You experience a sense of “all possible temporalities”. A sense of “all possible scents”. And a sense of all possible spatial relationships at once.
If you get any closer to the peak of annealing, the rainbows collapse into luminosity, the scents into a sense of presence, the temporalities into a sense of eternal now, and the possible feelings of space into a projective-less “view from nowhere”. The combination of all qualia values of each qualia variety somehow, incredibly, seem to add to zero. But not any kind of zero. A special “Zero” perhaps equivalent to “no information but awake”. (Cf. David Pearce’s Zero Ontology for a possible grounding of this state in fundamental physics.)

Yes, this is very much a real state of consciousness. It is profound, and extremely important.

I call it the “Rainbow God” state of mind. I do not know how to reliably induce it, but I do know that it is likely to have extremely deep computational, ethical, and experiential properties capable of advancing our understanding of the nature of the state-space of consciousness. I just figured you should know this exists.

June 2nd 2020

Andreas Keller • Olfaction and Experiential Authenticity:scent_presentation

Really excellent presentation about the biological and physical underpinnings of scent. It’s a bit on the long end (50 minutes) but you can get 80% of it by just watching the first 12 minutes. It’s really good! So much information…

For instance: did you know there are about 400,000 scented flower species in the world? I struggle to come up with more than 30 flowers off the top of my head (up from 5 just less than a year ago). The remaining 399,970? Who knows what they smell like. We don’t have words for these smells… is it “rose” or “jasmine” smell? Good luck using that kind of ontology describing the space of possible flower smells.

Also: it turns out that volatile molecules don’t diffuse very effectively. So that’s why you get “whiffs” of scents – for the most part, in the wild, air is a very non-homogeneous gas, with all kinds of pockets with specific linear combinations of aromachemicals. Hence why holding two essential oils side by side doesn’t give rise to a proper mixture between them. You need to literally mix the oils and then smell the mixed result if you want to actually know what the combination is like. Otherwise you will get a whiff of one, a whiff of the other, etc. with a Poisson-like distribution. This also reminds me that: we have an olfactory bulb in each nostril! So if you apply one scent in one nostril and another scent in the other nostril, you will get a kind of “bi-scent rivalry” [binosmic?] similar to what you get when you see one image with the left eye and one image with the right eye (i.e. “binocular rivalry”).

I do think that “digital smell” is possible (unlike the presenter). But it will require us to describe each molecule in terms of their ADSR patterns for each of the basic scent qualities (that is, to describe how the sweetness develops across time – its attack, decay, sustain, and release – and do the same for each core qualia scent dimension). Without taking into account the ADSR envelope for each molecule, the mixtures will be uneven.

The lowest-hanging fruit would be to use a non-negative least squares regression that minimizes the error for the envelope of each of the core qualia scent dimensions. Hence, the molecular spectrum is not enough – the non-negative least squares requires pattern-matching across the entire temporal envelope of each dimension. IF we do this – then digital smells might be possible after all (IMO!).

June 3rd 2020

There are a TON of questions whose real answer is: “Bleu De Chanel”. Think about it.

That’s how VAST the multiverse is.

“Bleu De Chanel” spans eons and eons of subjective time – the grapefruit/incense/amber vibe ringing on and on throughout eternity. That’s how large it ALL is.


You can get a powerfully believable Smirnoff Lime impression with as little as a few drops of citral and aldehyde C-12 in an ethanol + water mixture. Amazing what passes as a “fine drink” these days.

“At least add some linalool to make it worth it” – would be my recommendation.


Note to self: by virtue of their sharp smell, aldehydes are powerful high-frequency psychoactives.

June 6th 2020

Note to self: Smelling a bunch of aldehydes over and over for several days in a row causes bad headaches. Use them only occasionally from now on.

June 13th 2020

I asked a DMT being about the nature of scent qualia. Its response: “One hint: are you sure it’s only one kind of qualia?”

An insight came like a lightning bolt. Yes! Two types:

  1. Aromachemicals that are “character impact”
  2. Flavor-like vibes

Totally different state-spaces!

Luca Turin, the quantum neurobiologist who has done research on the vibration theory of olfaction (showing “we can smell functional groups”) told me that if perfumes are tomato soups, the money is in “making the best cream” rather than in the “tomatoes”. Character impact!

Examples of character impact molecules:

  1. Beta-ionone
  2. Iso-E-Super
  3. Ambroxan
  4. Hedione
  5. Helional

Examples of flavor-like vibe molecules:

  1. Alpha-damascone (beautiful!)
  2. Aldehyde C-12
  3. Citral
  4. Cis-3-hexanyl-benxoate (yuk!)
  5. Verdalia

June 20th 2020


Magenta: The Non-Spectral Color

An important point of confusion about qualia to which I offer a clarification:

The qualia you experience as a result of light coming into your eyes can be logically and empirically dissociated from physical light. Color qualia, just as much as visual texture qualia, can be triggered by auditory stimuli in people with synesthesia, or people tripping. More so, you don’t even need light to ‘see’ in your dreams. Visual qualia is ultimately not intrinsically tied to physical light. Phenomenal light, as it were, is a particular spatial qualia that we use to ‘illuminate’ our inner world simulations. Yet this illumination is not based on photons.

Hence the mystery of magenta: phenomenal colors don’t always map on to frequencies of light. Even leaving aside the issue of metamerism, magenta itself is a ‘non-spectral color’ because you need to combine at minimum two frequencies of light to trigger that color qualia in your visual field (namely, a combination of the upper and lower frequencies you can detect).

Why do we experience color qualia from light, then? This is not out of logical necessity, but rather, because it happens to have the appropriate information processing properties for the mapping to be evolutionarily advantageous. The state-space of color and visual texture happen to have useful isomorphisms to the structure of visual data. But there is nothing to suggest they are the best at representing ‘projective data-structures’.

In fact, I strongly suspect that once we master free-wheeling hallucinations and qualia control techniques, we will discover new applications of exotic qualia varieties for information processing purposes. Such as, for instance, using complex synesthetic representations of natural numbers that make it easy to ‘feel’ whether a 10-digit number is prime or not.

Anyhow, this all informs the kind of answer I might give to the question “what is it like to be a bat?”. In particular, it compels me to say that for all we know echolocation information is represented with scent qualia. We simply don’t know enough about the information-theoretic properties of state-spaces of qualia varieties to make an educated guess for what kind of qualia is best at representing echolocation information.

And more so, even if you were to train a human to use echolocation from birth, there is no guarantee that the qualia varieties and the associated state-spaces their brain would recruit for that task would have anything to do with bat echolocation qualia. So the problem has more moving parts than is usually assumed.

June 28th 2020

“Son, there is something I’ve been meaning to tell you for a long time, but only now I’m brave enough to do so: I just don’t think aromatic Fougères are a good fit for you. Based on my experience, I think Chypres would fit you better. Or even some woody citruses. Not Fougères.”

July 16th 2020

I love smelling dirty every once in a while.Photo on 7-16-20 at 3.59 PM

July 19th 2020

If you have a prejudice against the smell of single molecules because they are “too simple” and you need some “entourage effect” balanced blend “only nature can provide”… try smelling Agrumen Aldehyde Light. A single molecule that smells like a full perfume!

Soapy lime herbal!

July 22nd 2020

Freesia is 90% linalool and 3% beta-ionol. I suppose that’s why my 50%/50% mixtures weren’t quite Freesia-like.

July 24th 2020

Vimalakīrti then asked the bodhisattvas from the Host of Fragrances [world], “How does Accumulation of Fragrances Tathāgata explain the Dharma?”


Those bodhisattvas said, “In our land the Tathāgata* explains [the Dharma] without words. He simply uses the host of fragrances to make the gods and humans enter into the practice of the Vinaya. The bodhisattvas each sit beneath fragrant trees, smelling such wondrous fragrances, from which they attain the ‘samādhi of the repository of all virtues.’ Those who attain this samādhi all become replete in the merits of the bodhisattva.”


– Chapter X – The Buddha Accumulation Of Fragrances

[*Tathāgata is an honorable name for the Buddha of a realm.]

July 30th 2020

Emergent scents – when you combine two or more aromachemical cocktails and you get as a result a scent that is different than the sum of its parts.

I have in the past found a number of essential oil combinations that do this (pear + violet, pomegranate + honeydew, lemon + lavender). But I figured that it’s much better to try to identify clear cases of this phenomenon by combining pure molecules.

So this little “research program” I have going on is to find pairs of aromachemicals and then mix them in many different ratios and smell the results (usually dissolved in ethanol at a concentration of ~20%). So far, it seems that about ~25% of pairs of molecules I’ve tried result in emergent scents. Here are some specific examples (please feel free to try at home and verify!!):

1) Humulene + d-limonene: Humulene smells herbal and earthy, d-limonene smells like orange or mandarin. When the ratio is ~4:1 I get an emergent scent that I can only describe as “classic chewing gum flavor”, completely distinct and phenomenally richer than the ingredients alone.

2) Linalool + beta-ionone: linalool smells like a very gasoline-like volatile version of a flower scent, beta-ionone is the classic “violet scent” molecule. When combined in 9:1 ratio I get an emergent scent that is like that of a citrus version of freesia or peony.

3) Humulene + vanillin: vanillin is the smell of vanilla, which is watery at the onset (attack and decay) and creamy on the second half (sustain and release). When combined in 1:1 ratio you get a completely new scent that feels close to a dried out old tobacco Cuban cigar blended with coffee liqueur.

That last one is also relatively close to the classic combination of vanilla + vetiver. Luca Turin told me that the perfume called Habanita is precisely playing with a vanilla/vetiver combo, which at first sniff comes across as a completely new and unrecognizable (yet very pleasant) scent. He said that a wonderful metaphor for this phenomenon is like the song Loro by Gismonti, where in the second half the piano and the flute play in such a synchronized fashion that you get the impression that there’s a new instrument involved. I’ve been smelling vanilla/vetiver while listening to this song. It’s quite beautiful.


Humulene combined with d-limonene create an emergent “missing fundamental” type olfactory illusion of classical chewing gum flavor. It only works when Humulene is between 70% and 90% of the mixture (before adding ethyl alcohol). Cleanest example of “emergent scent” I’ve found.

Humulene is a simple scent of the category “earthy”, roughly similar to a vetiver essential oil but “one octave higher”. It also has a very mild musky undertone.

D-limonene is an orange/lemon-like scent. Extremely common in perfumery. Chances are something you ate today has it.

July 31st 2020

The simplest example I can think of to illustrate what an “emergent scent” is comes from the auditory illusion called “the missing fundamental”.

If you play 200 hertz together with 300 hertz and 400 hertz you will hallucinate an emergent 100 hertz tone.

The 100 Hz tone is not there! But it is quite real in your experience.

Of course if you are very acquainted with this auditory effect, you might notice the fundamental (100hz) is a bit fainter than expected, and infer it’s an illusion. But it is nonetheless very much present in your experience.

Likewise, when you smell Humulene + Vanillin at a 1:1 ratio you will get a third smell that emerges as a sort of gestalt that “bridges together” the two underlying notes.

You can probably infer the input scent is made up of two notes if you are really experienced with this kind of phenomenon. But the third note, the gestalt, does not disappear when you have “reduced” it to the two underlying notes. It’s still there. Thus, really, the whole is greater than the sum of its parts.


Hear the effect yourself: Missing fundamentals. Periodicity and Pitch

August 1st 2020

I like my coffee how I like my perfumes: with the fewest chemicals needed to cause the desired effect.

As an aside, learning about emergent effects in low-entropy perfume recipes makes me think that there could probably be a job for “scent simplification”. Namely, take something like cacao, with hundreds of molecules contributing to its characteristic scent. The question is: what is the minimum viable number of aromachemicals you can use to replicate it (within a Just Noticeable Difference unit)?

I suspect most natural scents that come from a complex entourage effect have relatively minimalistic reconstructions. A question that also emerges is: what is the most complex scent? I.e. what is the smell whose minimum reconstruction has the maximum number of molecular diversity?

[It’s important to distinguish between molecular entropy and phenomenal entropy. A solution of Agrumen Aldehyde Light and ethanol has low molecular entropy but pretty high phenomenal entropy, whereas a “lime accord” made of tens of molecules could be high in molecular entropy yet low in phenomenal entropy because it smells very cleanly like a ‘single note’]


A master perfumer like Ellena has memorized hundreds, if not thousands, of recipes for manufacturing smells. Many complex natural scents can be conjured with only a few ingredients. The scent of freesia, he explained, is created by combining two simple molecules: beta-ionone and linalool, both synthetics. (To give freesia a cold, metallic edge, a touch of allyl amyl glycolate is added.) The smell of orange blossom is made by combining linalool and methyl anthranilate, which smells like Concord grapes.


In my presence, Ellena once dipped a touche into a molecule called isobutyl phenal acetate, which has a purely chemical smell, and another touche into vanillin, a synthetic version of vanilla. He placed the two paper strips together, waved them, and chocolate appeared in the air. “My métier is to find shortcuts to express as strongly as possible a smell,” he explained. “For chocolate, nature uses eight hundred molecules. I use two.” He handed me four touches—vanillin plus the natural essences of cinnamon, orange, and lime. The combined smell was a precise simulation of Coca-Cola. “With me, one plus one equals three,” Ellena said. “When I add two things, you get much more than two things.”


The Scent of the Nile: Jean-Claude Ellena creates a new perfume.
– By Chandler Burr

August 5th 2020

Imagine you have been a musician for your village all your life. You play drums and acoustic guitar and you have never heard modern music. One day you are gifted an iPod and you listen for the first time to the crazy sounds of psychedelic trance. For the first time in your life you experience the wonders of reverb, flanging, distortions, and FM-synthesis. Surely this gives you a sense that your conception of music only tapped into a tiny fraction of what had always been possible.

An analogy could be made with smells: having tried essential oils one gets the impression of understanding what is possible in the realm of scents. But one day you discover Galaxolide, hedione, and eso E super. Like reverb and FM-synthesis in sound, these compounds are capable of giving surreal, unexpected, and space-warping properties to scents (much like reverb in sound, they are character impact molecules, meaning that they modify the presentation of other scents more than contributing a ‘flavor’ of their own).

Galaxolide in particular is something you have probably smelled, either in perfumes or detergents, but it really only becomes clear just how insane of a substance it is when you smell it raw. I associate it with “DMT Realm Aesthetics” – like a smell coming from another planet where hyperdimensional experiences are common everyday events, and the world of the arts uses exotic phenomenal time routinely. It has a vibe I can only describe as “having already always been here yet just arrived”. It’s probably what traveling in time feels like when you are in a transcendent Bardo between lifetimes.


Pellwall describes galaxolide thus: “Galaxolide is an isochroman musk, that has an odour profile that is liked by most people and is similar to a macrocyclic musk. It is strong, clean smelling and a good fixative. It combines well with other musks and is often used in combinations.”

In wikipedia, they describe the scent as: “a synthetic musk with a clean sweet musky floral woody odor”.

I think the musk-like quality accounts for maybe 60% of its effect. But I swear there is something much more special about it than just a clean musk. It has a kind of time-dilation effect, and it seems to my nose as a “musk but high-dimensional”. Perhaps it’s musk + the harmonics of musk. So while other musks are just a single note, galaxolide is like the feeling of a musky accordion.


I’ll write about my setup for doing this kind of research, but suffice to say that it’s super cheap if you know what you are doing. Each experiment (i.e. a little bottle with a few ml of a new combination in precise proportions) costs me about ~30 cents to make, all things considered (the cost of the materials, the ethanol, the pipettes, the bottle).

I highly recommend just getting a 2ml sample vial. It can cost as little as $2.16 (plus shipment) here: Galaxolide.

Other stellar molecules to try out to expand your conception of what’s possible:

Linalool, dihydro linalool, alpha-damascone, damascenone, helional, C-16 aldehyde (strawberry), agrumen aldehyde light, farnesene, nerolione, and alpha-ionone. All of that can cost you as little as $30. Not a bad price for expanding your “sense of what’s possible”.


I so wish I had a “DMT-smell accord” to use as a note in perfume compositions.

There is this one here meant to evoke the hallucinogenic state, but reportedly it has nothing to do with the actual scent of DMT, which I find very disappointing. I will try to find the way to emulate the scent of it – I suspect that linalyl acetate and coranol could be part of the compounds making up that accord. I’ll let you know if I manage to make anything vaguely resemblant of that scent.375x500.59536

August 14th 2020

Lemon Lavender World

One of the first essential oil combinations I fixated upon was that of lemon plus lavender. You could say it is the “speedball” equivalent of essential oil combos, for it relaxes and excites at the same time. I figured that trying to “understand” the “lemon-lavender world” would be a good exercise in the quest of mapping out the state-space of scents.


Lemon Lavender experiments

I currently have six different lemon essential oils from different brands and places, and seven lavender essential oils. To my surprise, the variability is very substantial. The lemon essential oils range from extremely sour and astringent to sweet and waxy. The lavenders I have also have many different qualities: some are very oily and flavorful, while others are particularly camphorous. Which of the qualities are “essential” for lemon and lavender is surely a matter of convention, though I also think they point to roughly objective attractors – the citrus sharpness of lemon rings high and has a cascading sourness that can be used for waking up the senses, whereas lavender has a narcotic entrancing reverb effect. My quest to understand, and ultimately create, lemon lavender smells was not defined in terms of merely reconstructing the standard natural smells, but as an attempt at understanding how these two qualities interact at the phenomenal level.

The diversity of lemon and lavender oils means that the space of possible combinations is even larger. Of the 42 possible combinations of one lavender oil and one lemon oil I have some are far more blissful and rich than others. I picked a few of my favorite ones to use as “model lemon-lavenders” to try to emulate.

Starting in the spirit that in order to deeply understand a scent I have to be able to construct it from scratch- so that I understand how each piece contributes to the whole- I set myself the goal of creating both lemon and lavender accords and then exploring their combinations. All starting from raw aromachemical ingredients, of course:

Making a Lemon Accord

I have always wanted to know what makes citrus fruits smell the way they do. Empirically, both isomers of limonene are a key piece of the puzzle. For instance, both lemon and mandarin oil have upwards of 80% limonene. Alas, if you smell limonene alone, you will notice it is somewhat one-dimensional in character. It IS pointing in the direction of “citrus” quite clearly, but on its own is indisputably too simple to evoke a real lemon scent.

I had a false start: aldehydes. Aldehyde C-8 through C-15 are all “extremely high-pitch scents”. They give a sharp edge to perfumes like Chanel No. 5 and the like. But they are very hard to use – partly because they are extremely potent. So for a couple of days I worked with combinations of citral and aldehydes that had, though a somewhat citric quality, mostly headache-inducing effects. I ended this series of experiments when I got a headache that lasted 24 hours (this goes to show how far I am willing to go to understand that sweet, sweet lemon qualia).

Taking a step back, I decided to explore a different angle. Valencene (note the great name) is very similar to limonene, except slightly lower in pitch. When mixed in equal proportions with limonene one gets a richer, more believable citrus scent – both molecules seem to say the same thing but in a slightly different voice, which results in a kind of chorus effect (unlike merely doubling the volume of a single voice). Alas, at this point the scent is still a bit flat, and not particularly lemon-like relative to near-enemy citrus fruits like the good old orange, mandarin, or grapefruit.

I recall being very puzzled by the scent of lime, as it seems like a kind of “super lemon” when it comes to its high-pitched sour and astringent character. And no matter how much I tried mixing citrus-like aromachemicals, I found it hard to get any hint of lime in the results. That is until I discovered that lime oil has a great deal of alpha- and beta-pinene. These are molecules that are primarily found in trees (in pines!) and smell very woody. As it turns out, to turn a citrus smell into an outright lime scent you need to add woody molecules. In retrospect, this was always hidden in the name: Lemon + Pine = Lime. After having this insight, I realized that even lemon requires a bit of alpha- and beta-pinene to distinguish it from orange scent.

After a lot of trial and error, the most convincing minimalistic lemon scent I identified is (numbers represent parts):

3 D-Limonene
3 Valencene
1 Citral
2 Linalool
1 Alpha-Pinene
1 Beta-Pinene
1 Nerolione (optional; for a rindy effect)

Making a Lavender Accord

This turned out to be more difficult than making a lemon accord. I think this is not only me: I also own two “fragrance oils” (those products that are advertised in the same context as essential oils, yet in the fine print reveal they are not at all natural, and instead are synthetic reconstructions) of lavender, and neither of the two smell anything like lavender. So I wouldn’t be the first to fail.



Linalool is a key ingredient of lavender, making up about 30% to 50% of most lavender essential oils. This is a very powerful aromachemical that seems to work as a gasoline-like fuel amplifier and modifier for any other scent (“there is no boring ten-carbon alcohol” – Luca Turin). It is also one of the things that makes lavender so narcotic and entrancing. On its own it is already quite interesting. But it is only one of the voices in lavender.

Then you have linalyl acetate, which makes up between 0% and 30% of lavender oil, depending on the species, place of origin, and time of harvest. Linalyl acetate has a “dry” quality, which I associate with “salt” (in fact if you just add this to the lemon accord above you get a smell I would describe as “salted margarita cocktail”). Alpha and beta pinene also play a role in lavender.

Interestingly, a lot of lavender oils also have up to 10% of camphor, which contributes to its narcotic get-well-soon cozy quality. Alas, it is hard to work with this material, and it always smells too synthetic to me. I found that instead I could double-down on beta-pinene, which is more camphorous than alpha-pinene (which is more earthy), and does the job quite nicely.

Finally, centifoleather, farnesene, and various alcohols like coranol can give “flavor” to the accord. In the end, I’ve settled on a minimalistic (but I think effective) arrangement. It does not quite hit the flavor of lavender, but I think does a good job at evoking its “character impact”:

4 Linalool
1 Alpha-Pinene
4 Beta-Pinene
2 Linalyl Acetate
3 Farnesene

Putting It All Together

Ultimately, adding these two accords (and their variations) together does not always produce the best results, as some aromachemicals are repeated and the proportions that give rise to the desired interactions can be scrambled by the combination. This, by the way, is a general reason why synthetic combinations span a much larger space of possible scents. In brief, because to make reconstructions with natural oils you are constrained by non-negative least squares methods, and many combinations may simply be inaccessible that way.

Anyhow – with the combination, I found that adding some character impact molecules like abroxan and helional was important to create a “bridge” between the two phenomenal characters. Alpha-ionol also seems to do something good here that is hard to put your finger on. But I think it’s that it adds the right kind of waxy rindy effect (which it has some of) in a way that does not make the mixture feel “dry” (which more classically citrus waxy smells like nerolione inevitably do). So the end result has some of these three molecules.

I am happy to say that the best lemon lavender reconstruction so far is about as good as the median natural lemon lavender mixture. It is not as good as the best lemon lavender oil mixture, though, but it is a start. I still expect to perfect it quite a bit before unleashing it into the world.

Ladies and gentleman, I present to you Lemon Lavender World:

1 Citral
3 D-Limonene
3 Valencene
4 Linalool
1 Alpha-Pinene
4 Beta-Pinene
2 Linalyl Acetate
2 Helional
3 Farnesene
1 Ambroxan
2 Alpha-Ionol
80 Ethanol


Previous experiments
Marking the interesting ones to keep
Reference bottles of pure chemicals (and one chypre accord)


  1. drbobkauai · August 16

    Nui mahalos! I am looking forward to trying all three.

  2. drbobkauai · August 15

    As always I just love and learn so much from your wonderful research, ideas, and posts! Big thanks for the work you are doing for all of us! I will keep spreading the word.

    Two quick questions:

    Could you recommend simple essential oil combos for men that would be – 1: energizing and 2: euphoric?

    More nui mahalos & alohas too!

    • algekalipso · August 15

      Thank you! I appreciate the very kind words 🙂

      > Could you recommend simple essential oil combos for men that would be – 1: energizing and 2: euphoric?

      Sure! Lemon lavender is certainly that, I think. Though if you don’t enjoy sour scents, that might not be ideal. I know that’s quite variable between people – for instance, I enjoy drinking pure undiluted lemon juice (every once in a while as I know it’s bad for one’s teeth). But other people may not appreciate that energizing goosbump reaction to strong sourness the same way I do.

      The vetiver + vanilla combo comes to mind as well. Vanilla alone is very relaxing, but with vetiver the mixture is quite invigorating.

      Finally, not a combo, but neroli oil in general might fit the description.


Leave a Reply