The Art of Scent

The study of scent is an incredible field that delves into the world of one our five senses: smell. Researchers are approaching smell in ways none of us have considered. They see the air as an accumulation of many chemicals. Some of these chemicals we would recognize, such as oxygen, carbon dioxide, and other atmospheric gases.

However, researchers are looking for very specific odorous chemicals or odorous chemical compounds that have been volatilized. This means that the chemicals have become a vapor. These specific volatilized odorous chemicals float around in the air which we then suck up into our nose when we breathe. In our nose these chemicals dissolve in our mucus.


Essential oils are examples of products that contain odorous chemicals. Image provided by


After dissolving, they make their way to the 350 olfactory receptors which get stimulated by those chemicals. These receptors then send electrical signals to the brain. The part of the brain that receives those electrical signals is called the olfactory bulb.

This bulb sorts out those signals to tell the brain. This tells what the smell of that volatilized odorous chemical is that stimulated those olfactory receptors. From the start of sucking air into our nose to the end where our brain translates the scent, this is the process of smell.

Something very interesting that researchers have found is that because the olfactory bulb is part of the limbic system, it ties into emotion. They have been trying to find out which scents cause which emotions in people but have found something psychologically intriguing to skew their results. It depends on the individual and the past experiences they have had with a smell.


The olfactory bulb. Image provided by


For example, the general population will find lavender relaxing. However, someone who has had a bad experience in the past with that smell present may experience negative emotions instead of being relaxed. It is this connection to the general populations’ emotional reaction to smells that has not only scientific researchers experimenting with scent, but the everyday individual. This artful science of scent mixing is called Aromachology.

Castro JB., Ramanathan A., and Chennubhotla C., are the three professors who submitted the article, “Analysis of odor perceptual space using non-negative matrix factorization”(1). What this article basically says is that there are certain “groups” of smells and that these certain groups are described in certain ways. They data-mined a large amount of people to see how scents were described and in what groups they would be placed.

For example, there is a Citrus group. In this group is the scent of lemon, orange, and other such fruits. However, in the words used to describe the scents in the Citrus group, some people said it smelled like “perfumery” or described it as “sweet”. When individuals start to use words like “sweet” or other emotionally attached words, it becomes easier to see what scents cause which emotional reaction.


What is your reaction to the smell of citrus? Image provided by


By data-mining it is easy to see how many people used different words to describe something. This makes it easier to pick out which scent causes what emotional reaction for the majority of the people tested.

These professors had concentrated down the scents we could smell into 144 categories based on similar description by the majority of people who smelled them. They then boiled those categories down to 10. After that, they created categories for majority descriptions.

What they found were distinct correlations between description and chemical type:


  • Cluster 1: Fruity/Citrus, lemon, Grapefruit, Orange
  • Cluster 2: Nail Polish Remover, mothballs, Alcoholic, Etherish, Anesthetic, Cleaning Fluid, Gasoline, Solvent, Turpentine(Pine Oil), Leather, Tar, Creosote, Disinfectant, Carbolic, Medicinal, Chemical, Ammonia, New Rubber, Kerosene, Paint, Varnish, Metallic
  • Cluster 3: Tea Leaves, Caraway, Minty, Peppermint, Camphor, Eucalyptus, Anise(Licorice), Cheesy, Cool, Cooling
  • Cluster 4: Floral, Rose, Violets, Lavender, Cologne, Musk, Perfumery, Aromatic, Soapy, Incense, Light
  • Cluster 5: Fruity/Other than citrus, Pineapple, Grape Juice, Strawberry, Pear, Cantaloupe, Honey, Dew Melon, Peach(Fruit), Banana, Cherry(Berry), Sweet, Raisins
  • Cluster 6: Laurel Leaves, Black Pepper, Green Pepper, Dill, Oak Wood, Cognac, Woody, Resinous, Cedarwood, Geranium Leaves, Celery, Fresh Green Vegetables, Crushed Weeds, Crushed grass, Herbal, Green, Cut grass, Raw Cucumber, Cardboard, Rope, Wet paper, Musty, Earthy, Moldy, Raw Potato, Mushroom, Beany, Bark, Birch bark, Cork, Dry, Powdery, Chalky
  • Cluster 7: Honey, Almond, Nutty(Walnut ETC), Spicy, Clove, Cinnamon, Chocolate, Vanilla, Maple Syrup, Caramel, malty, Molasses, Coconut, Hay, Bakery(Fresh Bread), Peanut Butter, Burnt, Smoky, Fresh Tobacco Smoke, Coffee, Stale Tobacco Smoke, burnt Paper, Burnt Milk, Burnt Candle, Oily, Fatty, Buttery, Fresh Butter, Popcorn, Fried Chicken, Warm
  • Cluster 8: Apple(Fruit), Seasoning(For Meat), grainy(As Grain), Yeasty, Sour Milk, Fermented(Rotten), Fruit, Beery, Wet Wool, Wet Dog, Dirty Linen, Stale, Mouse, Eggy(Fresh Eggs), Burnt Rubber, Bitter, Sharp, Pungent, Acid, Sour, Vinegar, Sauerkraut, Urine, Cat Urine, Fishy, Kippery(Smoked Fish), Smein
  • al, Sperm, Sooty, Meaty(Cooked, Good), Soupy, Cooked Vegetables, Rancid, Sweaty, Household Gas, Sulfidic, Garlic, Onion, Blood, Raw Meat, Animal, Sewer, Putrid, Foul, Decayed, Fecal(Like Manure), Cadaverous(Dead Animal), Sickening, Heavy



For example, in the top graph above, the light green circle labeled “Cluster 1” represents the types of descriptions used that fall into that cluster. In the list below the top graph you can see that words such as fruity, citrus, lemon, orange, and grapefruit are what make up “Cluster 1.” This means “Cluster” 1 is the citrus cluster. 

The graph below represents the chemicals that received descriptions of various words from the population of test subjects. Depending on which words were used to describe it, the chemicals are moved on the chart towards their descriptions.

The chemicals that are clustered in the area of “Cluster 1” from the above graph would be the chemicals that match the description of those words. For example, in “Cluster 4” on this graph is the chemical Citralva. This chemical is sometimes put into lemon scented hand-soaps and air-fresheners. Using this pattern we can see which chemicals fall into which categories and what words are used to describe them.



  • Cluster 1: Aldehyde C-16, Allyl Carpcate, iso-Amyl Acetate, Amyl Butyrate, Dimethyl Benzyl Carbinyl Butyrate, Ethyl Butyrate, Ethyl Popionate, Fructone, Methyl Anthranilate, Undecylenic Acid, gamma-Valerolactone
  • Cluster 2: Amyl Phenyl Acetate, Auralva, iso-Bornyl Acetate, Cashmeran, Dimethyl Phenyl Ethyl Carbinol, Hydroxy Citronellal, Indolene, beta-lonone, alpha-Irone, Lyral, methoxy-Naphthalene: 2-Methoxy, Naphthalene, Methyl Acetaldehyde Dimethyl Acetal, Musk Galaxolide, Musk T onalid, Phenyl Ethanol, Sandiff, Santalol
  • Cluster 3: dl-Camphor, l-Carvone, p-Cresyl Acetate, Eucalyptol, l-Menthol, Methyl Salicylate, Safrole
  • Cluster 4: Amyl Cinnamic Aldehyde Diethyl, Acetal, Citral, Citralva, Floralzone, Hexyl Cinnamic Aldehyde, Linalool, d-Limonene, Melonal, Myrataldehyde
  • Cluster 5: Anisole, 1-Butanol, m-Cresol, p-Cresol, p-Cresyl-iso-Butyrate, p-Cresyl Methyl Ether, Cyclohexanol, 2,5-Dimethyl, Pyrazine, Diola, Diphenyl Oxitle, 1-Heptanol, 1-Hexanol, 3-Hexanol, Iodoform, Methyl Furoate, para-Methyl, Quinoline, Nonyl Acetate, 1-Octanol, Phenyl Acetylene, Terpineol, Tetraquinone, Thymol, Toulene
  • Cluster 6: Adoxal, Andrane, iso-Butyl Quinoline Chlorothymol, Iso-Cyclocitral, Cyclotropal, Decahydro, naphthalene, Dibutyl Amine, Grisalva, Hexanal, Hydratropic Aldehyde Dimethyl Acetal, 2-Methyl-iso Bomeol, Methyl iso-nNicotinate, Nootkatone, 1-Octen-2-OL, iso-Phorone, alpha-Pinene, iso-Propyl Quinoline, Propyl Sulfide, gamma-Undecalactone
  • Cluster 7: Abhexone, Acetophenone, Aldehyde C-18, Anethole, benzaldehyde, Dihydro Pyrone, Caryophyliene(beta and gamma Isomers), Celeriax, Cinnamic Aldehyde, Coumarin, Cuminic Aldehyde, Eugenol, Furfural, trans-1-Hexenal, ortho-Tolualdehyde, Vanillin
  • Cluster 8: ortho-Acetyl Pyridine, Cyclotene, 2,4-trans-Decadienal, 2,3-Dimethyl Pyrazine, 2,5-Dimethyl Pyrrole, 2-Ethyl Pyrazine, Furfuryl Mercaptan, Guaiacol, Heptanal, Thienopyrimidine, Zingerone
  • Cluster 9: Butyl Sulfide, Cyclodithalfarol, 2-Cyclohexanedione, Diethyl Sulfide, Dimethyl Trisulfide, Hexyl Amine, Pyridine, Tetrahydro Thiophene, Thioglycolic Acid, Thiophene
  • Cluster 10: Amyl Valerate, Butanoic Acid, Hexanoic acid, Hexyl Amine, Indole, Maritima, Methyl Thiobutyrate, Pentanoic Acid, 4-Pentenoic Acid, Phenyl Acetic Acid, Propyl Butyrate, Skatole, Trimethyl Amine, iso-Valeraldehyde, iso-Valeric Acid


This information allows us to see the beginnings of a measuring system for scent. Soon, we may have a way to measure scent that is as accurate as measuring color. This will let us move forward to finding out exactly what types of scents cause what type of reaction.  





1. Castro, Jason B., et al. “Categorical Dimensions of Human Odor Descriptor Space Revealed by Non-Negative Matrix Factorization.” PLOS One, vol. 8, no. 9, ser. e73289, Sept. 2013, pp. 1–16. e73289, doi:10.1371/journal.pone.0073289.

Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Most Popular

Would you like to be a part of Designn?

Every subscriber gets the latest magazine, Designn news & causes. Plus did I mention the awesome freebies?

To Top



I would like to get some ideas

And don’t worry, we hate spam too! You can unsubscribe at anytime.