In short
Gas chromatography-olfactometry splits the column output between a detector and a sniffing port, so a trained person can note what each elution point smells like. Its main lesson is that peak size doesn't predict odor importance. A large peak may smell of little, and a compound below the detector's reach may define the character. You met this in linking sensory and analytical data. This lesson covers how it's actually measured.
What you will learn
- Explain why peak size doesn't predict odor importance
- Compare the main GC-O methods
- Read and question a published GC-O result
Before this lesson: Fingerprinting Natural Materials
Why it's useful to a perfumer
Rose oil is a good illustration. Its largest peaks are alcohols such as citronellol and geraniol, yet trace materials like beta-damascenone and rose oxide contribute far more than their percentage suggests. In reconstruction, a formula built from the largest peaks is usually disappointing, and this method helps to explain why. It also suggests where to look for missing notes.
The main methods
| Method | What the assessor does | Output | Weakness |
|---|---|---|---|
| Detection frequency | Several assessors note when they smell something | How many detected each point | Doesn't measure intensity well |
| AEDA | Sniffs a series of dilutions | Highest dilution still detected | Time-consuming; depends on volatility and the extract |
| CHARM | Sniffs dilutions, recording onset and end | A peak-like area of odor activity | Laborious |
| Time-intensity, such as OSME | Rates intensity continually | An odor intensity trace | Requires training and careful scaling |
Odor activity value
A simple companion calculation is the odor activity value (OAV), which is concentration divided by threshold.
| Compound | Concentration (mg/kg) | Threshold (mg/kg) | OAV |
|---|---|---|---|
| P | 5000 | 10 | 500 |
| Q | 1 | 0.001 | 1000 |
| R | 2000 | 100 | 20 |
Q is 5,000 times less concentrated than P, and it has the larger OAV. But be careful. Thresholds differ between people and media, and published values vary enormously. An OAV above 1 suggests something might be detected, not that it matters. And OAV assumes compounds act independently, which they don't always do in mixtures. Use it to direct your attention, and not as a measure of importance.
Limits
- Extraction bias. The method sees what the sample preparation delivers. See headspace methods.
- Separation alters perception. Sniffing components one at a time removes masking and blending. A whole is more than its GC-O list.
- Human factors. Attention drifts over a long run, assessors vary, and they can't report what they can't name.
- Literature is food-heavy. Much of the published work concerns foods and beverages. Applying it to a fine fragrance calls for care.
Independent learners
You won't run GC-O at home, as it requires a specialist instrument and supervision. Don't improvise any arrangement for sniffing the output of any instrument. What you can do is read papers critically, ask a laboratory or university about collaboration, and carry out the bench analogy below. It teaches the logic, by finding how far you can dilute before a material stops being detectable. It's not GC-O, so don't report it as such.
Practical exercise: A paper review and a bench analogy
Part 1: a paper review. Find one published GC-O study on a natural material. Note the sample preparation, method, number of assessors, top odorants, and what the authors did to confirm their identifications. Write what the result can't show.
Part 2: a bench analogy, on strips only.
- Choose two materials with very different strength, one mild and one strong, and read their SDS.
- Make serial dilutions: 10%, 1%, 0.1%, 0.01% and 0.001%, as far as is practical with your scale.
- Have someone code them, and test each against a solvent blank, weakest first, in a triangle format if you can.
- Note the last dilution you can reliably detect. Compare the two materials.
You have done this well if your review names limits the paper did not stress, your bench work has blanks, and you label it a dilution study rather than GC-O.
Common mistakes
- Reading a dilution factor as importance in the whole. Mixtures behave differently.
- Quoting thresholds as fixed. They vary with the person and the method.
- Reconstructing from peak size alone. The trace materials may be the point.
Key terms
- AEDA
- Aroma extract dilution analysis: a GC-O method that repeatedly dilutes a sample to find the highest dilution at which each odorant can still be detected.
- Odor activity value
- A compound's concentration divided by its odor threshold, in the same medium. It's a simple estimate of whether a compound is likely to matter.
Knowledge check
Written and reviewed by the RUŌOD Lab team. This is general education, not legal, regulatory, medical or safety advice. Example dilutions and exercises are for learning, are not validated products, and have not been safety-assessed. Completing a lesson confers no qualification. How we check these lessons.