Looking for new scents

 　　Iquitos is a difficult place to reach. The city is located deep in the tropical rainforest of the Amazonas in Peru and can only be reached by plane or water-there is no road to reach there. But for Saint-Piquart, remoteness is its charm. Piquart is a scent explorer for a French perfume company. He went to Iquitos to find new ingredients for making perfume. The root of a local plant has an unusual mix of fruit and leather, and Piquart wants to bring it into the perfume industry.

　　The world is searching for new scents, which is driven by business opportunities in the fragrance industry. The fragrance industry is huge, with annual sales of 70 billion U.S. dollars worldwide. Not only perfumes, but also everything from soaps and shampoos to aromatic candles and air fresheners.

　　But why are we willing to spend so much money on good smelling things?

　　Among the main senses, smell is the least we know, which makes it difficult for us to answer this question; but neuroscientists, psychologists and even artificial intelligence researchers are unveiling its mystery.

Preference for smell is acquired

　　From the love of perfume to the aversion to rancidity, all our scent preferences are acquired.

　　The first evidence is that young children do not have a particular preference for smell. They often do things that make adults sick, such as trying to eat poop on diapers. It is the lesson of the parents that let them know that it is a kind of "bad" smell.

　　Second, different cultures have different views on what smells are pleasant. For example, Americans think the smell of holly plants is good, partly because it is an important ingredient in root juice (a non-alcoholic beverage made from the root juice of holly plants); while Europeans smell it I am afraid that Holly will frown, because it will make them unpleasantly think of analgesic oil-something that you will put on when your limbs are aching.

　　In addition, studies have shown that the same smell smells good or bad, often depending on which associations are triggered. For example, many cheeses contain isovaleric acid, the same smell molecule as sweat socks. When told in advance that it was cheese, everyone liked it; but when it was told it was smelly sock, no one liked it. The smell has not changed, but the judgment has changed. Obviously, these judgments are all learned.

How do we perceive smell?

　　However, acquired acquisition is not the whole thing. As a feeling, smell naturally has its objectivity. This requires us to understand how smell works.

　　There are about 400 different smell receptors in our noses. Scientists still have controversy about how these receptors recognize odor molecules: most people believe that when the shape of the odor molecule is exactly complementary to the surface of a receptor (such as one convex and the other concave), the molecule can be recognized by the odor receptor. Come out; but it is also believed that at least some odor receptors work by sensing the vibration of odor molecules.

　　Regardless of the actual situation, each odor receptor can recognize a different set of odor molecules, and each odor molecule will only trigger one or a few different odor receptors. Therefore, recognizing an odor is actually activating one or several odor receptors, which is like playing a note on one or several keys of a piano.

　　What makes this picture complicated is that because everyone has different smell receptors, everyone is playing with their own unique "keys". No one can smell things in the same way, and even small genetic differences between individuals can affect our sense of smell. For example, there is an odor receptor called OR6A 2, and people with a variant of it prefer the smell of coriander than others.

　　Even smell receptors are not the whole thing. Studies have shown that nasal mucus contains an enzyme that can change odor molecules before they reach the receptors. No one knows how much these enzymes have on odor perception, but there are signs that they are important. For example, some people describe a particular smell as coming from a woody plant, while others describe it as a raspberry smell. Further research found that there was no difference in the odor receptors of the two groups, but in the second group, the odor molecules of woody plants were converted into molecules with the odor of raspberry by an enzyme.

Predict the smell of molecules

　　These individual differences have not prevented olfactory scientists from solving a big problem: can we predict in advance what a molecule will smell like based on its structure? If we can, we can create the odor molecules we need in the laboratory.

　　The first clue to answer this question appeared 10 years ago. Israeli scientist Noam Sobel and his colleagues asked 185 people to evaluate the pleasantness of 90 anonymous odors. Then, they surveyed and counted more than 1,500 physical properties of each odor molecule to find out the pleasantness of the odor. Depends on which attributes.

　　They found that, in general, good smells often come from large and loosely structured molecules, while small and compact molecules are unpleasant. Therefore, molecular size and compactness are a good indicator of odor pleasantness-although not absolute. Since then, several other studies have more or less confirmed this conclusion.

　　Why do large and loose molecules often correspond to "good smells", and small and tight molecules correspond to "unpleasant smells"? The answer may be related to how our hunter-gatherer ancestors assessed the freshness of food. We know that bacteria will eat the large molecules in food and break them down into smaller molecules; therefore, the smaller the molecules of the food component, the more decayed it is. vice versa. In this way, we have gradually evolved the ability to stay away from rotten and disease-causing foods by smell.

　　Artificial intelligence can do even better at predicting odors. In 2017, mankind held an international "olfactory prediction challenge" to see if there is a way to predict the smell of molecules based on their physical properties. The organizer provided participants with data on more than 400 physical properties of each odor molecule, and asked participants to design algorithms based on the physical properties of these odor molecules to rate their pleasure. An algorithm designed by an American scientist finally won. The accuracy of the algorithm in predicting pleasure reached 50%. Considering that the same person’s evaluation of the same molecule at different times is different, this accuracy rate is quite impressive.

Make the smell richer

　　But these efforts have a big disadvantage: they only focus on a single molecule. In practice, there is no situation where you will only encounter one kind of odor molecule. For example, in baked foods such as coffee or bread, hundreds of chemical substances give them a rich fragrance, which is unmatched by a single substance. Therefore, mixed odors need to be studied more.

　　When perfume masters create rich fragrances in their own formulas, they have some tricks of their own. A typical perfume usually has dozens of ingredients, and the dosage of each must be appropriate so that no single scent will dominate. Perfumers will also match various ingredients with varying volatilization rates to make subtle changes in the scent over time, leaving a lingering aftertaste.

　　Surprisingly, it doesn't matter whether the individual ingredients in a perfume smell pleasant or not. Perfumes usually contain small amounts of seemingly indecent-sounding ingredients, such as skatole that smells like stool or the anal secretions of a civet (a small mammal from Asia and Africa).

　　The demand for perfume has prompted perfume companies to look for new scent molecules to add to their formulations. This is why the scent explorer Saint-Piquarte mentioned at the beginning of this article was sent to Iquitos by the French perfume company. However, many companies prefer to let chemists synthesize certain key odorants in the laboratory. Tailoring odor molecules for odors is not far away. Who knows what weird smell we will smell in the future.

Tips

　　Skatole

　　found in feces, also found in some of the Flowers. It has a foul smell at high doses, but it smells of flowers at lower concentrations.

Further reading

　　The secret of the perfume laboratory

　　Not all new fragrances need to be found in remote areas. Some are right under our noses. For example, one ingredient of perfume comes from ordinary apples. In addition, scientists have successfully extracted a fragrance from a precious wild mushroom.

　　Once chemists find a molecule of interest, they will make some subtle changes to it to see how it affects the smell. For example: a chemist extracted a component of patchouli oil, fermented it with enzymes, and added a peppery flavor. The result was an ingredient called colchicum, which is currently one of the most popular new perfume ingredients. .

　　What conditions does a new type of odor molecule need to meet? First of all, this molecule smells no peculiar smell; secondly, it will not interfere with other ingredients or be rapidly degraded by organisms; and finally, it is safe for the human body. For example, many scent molecules extracted from citrus are not suitable for making aromatic candles because they form an unpleasant gasoline smell when burned.