Even when people wear the same perfume, its scent can develop differently over time. A 2018 study investigated the background to this experience by examining compounds in the air above the skin. When eight people applied the same fragranced product to their forearms, the mixture of compounds recovered changed over time, and differences between individuals increased.
The paper’s main purpose was to evaluate an analytical method for repeatedly collecting volatile compounds without damaging the skin, to study fragrance persistence and individual differences. Rather than tracking changes in what people smelled, it first examined which compounds could be recovered from the skin.
Creating a small sampling space above the skin
The participants were eight healthy adults aged 20–30, four women and four men. On measurement days they did not apply perfume or cosmetics to their arms, and before testing they washed their hands and arms with water and a specified soap, then dried them.
A small glass device was taped to the inner forearm. Inside it, the researchers placed a thin fiber with a coating that captures volatile compounds. Rather than touching the skin directly, the fiber collected compounds from the air accumulating above it.
After examining sampling periods from five to 30 minutes, the researchers selected 15 minutes. The collected compounds were separated and analyzed by gas chromatography–mass spectrometry. Controls for the fiber, room air, and tape helped exclude external contaminants as far as possible. Measurements took place in the same location, at an average room temperature of 21°C and humidity of 62%.
Untreated skin already yields multiple compounds
Before the product was applied, 24 compounds were reported across the eight participants, with acids and aldehydes among the main groups. These included nonanal and decanal, for example. Fourteen compounds were detected in every participant.
Some compounds were relatively stable across repeated samples from the same person, while others varied more. To examine differences between people, researchers need to assess not only individual differences but also measurement variability within a person. This was another important part of evaluating the method.
Some substances classified as baseline skin compounds could, however, have originated from previous cosmetic use. Identification relied mainly on mass-spectral database matching and retention indices, and some identifications were tentative.
Following one microliter of the same product for two hours
Next, one microliter of an eau fraîche product marketed for women was applied to the inner forearm. The first 15-minute collection began 30 seconds after application. Further samples were collected from the same site after one and two hours. Participants could carry out normal office work in the meantime, but were instructed not to wipe the application site.
Across the post-application samples, 51 compounds were reported: 19 classified as skin-derived and 32 as product-derived. Signals for linalool, linalyl acetate, limonene, methyl dihydrojasmonate, and other compounds were large in the first collection; signals for many product-derived compounds then declined.
Not everything disappeared at the same rate. For example, β-pinene was no longer detected after one hour, whereas methyl dihydrojasmonate, galaxolide, linalyl acetate, and γ-decalactone were still detected in all participants after two hours. Different compounds therefore showed different patterns over time.
The same compound does not necessarily persist equally on everyone’s skin
There were also differences between participants. For example, α-ionone and neryl acetate were still detected after two hours in some participants but not others. Even with the same amount of the same product applied to the same body site, the composition of recovered compounds was not uniform as time passed.
The researchers also examined these differences through the rank order of compound signal sizes. Pairing the eight participants yielded 28 pairs. Using a rank correlation below 0.9 as the criterion for distinguishing profiles, 64% of pairs were distinguishable immediately after application, compared with all pairs after one and two hours.
This is not an identification accuracy showing that unknown people could be recognized by scent with 100% success. It is a comparison showing that, among these eight people, the similarity of compound profiles decreased over time and analytical differences became more pronounced.
The authors discuss skin properties, air movement, and concentrations on the skin as factors potentially affecting fragrance diffusion and persistence. They also measured skin hydration, but all participants were within the normal range. This did not demonstrate that hydration caused the individual differences observed; compounds’ affinity for water and their relationships with skin lipids remain possible explanations.
Distinguishing inferred fragrance notes from perceived scent
Using database odor descriptions, the researchers divided detected product-derived compounds into ten groups, including floral, citrus, and musk. Summing the analytical signals within these groups depicted a predominantly floral profile, with initially prominent herbal/terpenic components declining and the composition shifting over time toward one including musks and other groups.
However, this was not an assessment in which people smelled the skin and rated its fragrance notes. Each compound’s peak area is not its share of the perceived odor contribution; detection sensitivity and ease of collection also affect it. Establishing how many hours a fragrance smells floral to people requires sensory assessment alongside the analysis.
Immediately after application, signals for compounds originally recovered from the skin decreased, and some became undetectable. Signals for compounds such as nonanal and decanal subsequently increased again. The authors propose effects on sebum oxidation and microbial transformation as possibilities, but the study did not measure bacterial counts or odor ratings. The soundest interpretation is that applying the product also changed the profile of measured skin compounds.
Summary
A fiber-based method for sampling the air above the skin allowed repeated measurements of compound changes after application of the same fragranced product. Signals for many product-derived compounds declined over time, but some remained detectable in everyone after two hours, and patterns of persistence differed between people.
The value of this study is in presenting a tool for examining the experience that a fragrance develops differently on different people as time-dependent changes in individual compounds. It is a small starting point involving eight people, one product, and the forearm. Nevertheless, tracking which compounds remain detectable, and for how long, can help explain fragrance on skin when combined with assessments of what people actually smell.
The study was funded by Science Foundation Ireland. The authors declared no conflicts of interest and stated that the funder had no role in study design, data collection, analysis or interpretation, manuscript preparation, or the decision to publish.
Reference
Duffy, E., Albero, G., & Morrin, A. (2018). Headspace Solid-Phase Microextraction Gas Chromatography-Mass Spectrometry Analysis of Scent Profiles from Human Skin. Cosmetics, 5(4), 62. Original research article
