Quantitative Determination of Perfume Notes Using an Electronic Nose

A recognizable response pattern is generated by traditional electronic noses (eNoses) using a number of non-specific, non-identical chemical sensors. People developing artificial intelligence algorithms and neural networks have shown interest in eNoses for some time now, but due to physical instability and overlapping responses, physical sensors exhibit only limited performance. eNoses are unable to separate or quantify the chemistry of aromas.

zNose®, a new type of eNose, is based upon ultra-fast gas chromatography. It simulates nearly infinite specific virtual chemical sensors, and generates olfactory images according to aroma chemistry.

The zNose® can perform, in near real-time, analytical measurements of volatile organic odors and vapors with part-per-trillion sensitivity. It takes only a few seconds of time to perform separation and quantification of the individual chemicals within an odor. Electronically variable sensitivity is obtained by employing a patented solid-state mass-sensitive detector, universal non-polar selectivity, and picogram sensitivity.

An integrated vapor preconcentrator together with the electronically variable detector allows the measurement of vapor concentrations across more than six orders of magnitude. This article describes a portable zNose®, shown in Figure 1, which is a useful quality control tool for measuring the concentration of chemicals employed in 25 basic fleuressence aromas as well as a perfumery mixture or designer perfume.

A trained perfumery technician can easily quantify a ‘good’ perfume aroma in near real-time. After the ‘good’ chemical signature has been defined, objective and quantitative quality control testing can be done with zNose® analyzers incorporated into the perfume production process.

Portable zNose® technology incorporated into a handheld instrument

Figure 1. Portable zNose® technology incorporated into a handheld instrument

How the zNose™ Quantifies the Chemistry of Aromas

Figure 2 shows a simplified diagram of the zNose® system, consisting of two parts. Section one employs a capillary tube (GC column), helium gas, and a solid-state detector. Section two comprises of a heated inlet and pump, sampling ambient air.

A “loop” trap joins the two sections, and acts as an injector if placed in the helium section (inject position) and as a preconcentrator when placed in the air section (sample position). The operation consists of two steps. First, the ambient air (aroma) is sampled, and organic vapors are collected (preconcentrated) on the trap.

Once the sampling has been performed, the trap is switched into the helium section in which the collected organic compounds are injected into the helium gas. The organic compounds, having different velocities, pass through a capillary column, and due to this, the individual chemicals exit the column at characteristic times. A solid state detector detects and quantifies the chemicals when they exit the column.

Simplified diagram of the zNose® showing an air section on the right and a helium section on the left. A loop trap preconcentrates organics from ambient air in the sample position and injects them into the helium section when in the inject position.

Figure 2. Simplified diagram of the zNose® showing an air section on the right and a helium section on the left. A loop trap preconcentrates organics from ambient air in the sample position and injects them into the helium section when in the inject position.

The capturing of sensor data is controlled by an internal high speed gate array microprocessor. An RS-232 or USB connection is used to transfer the data to a user interface or computer.

Figure 3 shows the aroma chemistry, which can be displayed as a sensor spectrum or a polar olfactory image of odor intensity vs. retention time. A single n-alkane vapor standard is used to achieve calibration. A set of retention times of known chemicals indexed to the n-alkane response (Kovats indices) enables machine-independent measurement and compound identification.

Sensor response to n-alkane vapor standard, C6-C14 in this case, can be displayed as sensor output vs. time or its polar equivalent olfactory image

Figure 3. Sensor response to n-alkane vapor standard, C6-C14 in this case, can be displayed as sensor output vs. time or its polar equivalent olfactory image

Chemical Analysis (Chromatography)

The spectrum of column flux is provided by the time derivative of the sensor spectrum (Figure 3); the flux is commonly referred to as a chromatogram. An accurate measure of retention times is provided by the chromatogram response (Figure 4) of n-alkane vapors (C6 to C14).

Red bands depict graphically-defined regions, calibrate the system, and provide a reference time base against which subsequent chemical responses are indexed or compared. For instance, a retention time index of 1050 is provided for a response midway between C10 and C11.

Chromatogram of n-alkane vapors (C6 to C14)

Figure 4. Chromatogram of n-alkane vapors (C6 to C14)

Fleuressence Samples

Many primary odor perfume compound bases consisting of one or more aroma chemicals can behave as a set of basic fleuressence groups useful for perfume compounding. Fleuressence bases are prime olfactory notes, which are used to create complex perfume aromas.

Perfumers World (http://www.perfumersworld.com) offers a perfumery training kit with 25 fleuressence bases representing a group of prime notes. Table I lists the bases and their aroma note descriptor or names.

Perfumery training kit

Figure 5. Perfumery training kit

Table I. Fleuressence Bases

A Aldeyde N Narcotic
B Iceberg O Orchid
C Citrus P Phenolic
D Dairy Q Balsamic
E Edible R Rose
F Fruits S Spice
G Green T Tar/ Smoke
H Herb U Animalic
I Iris V Vanilla
J Jasmine W Wood
K Konifer X Musk
L Linalool Y Yeast/ Mossy
M Muguet Z Zolvent

Aroma Testing Methods

A known concentration of vapor is produced when a small amount of flueressence base is injected into a septa-sealed vial. The zNose® vapor analyzer can be used to sample the vapor.

The analyzer’s internal temperatures are set to 160 °C and 200 °C for the vapor inlet. A short one-second sample time (0.5 mL) is best when using undiluted base materials.

Testing materials are 40 mL septa sealed vials, a 1-10 µL syringe, fleuressence samples, and a zNose® vapor analyzer

Figure 6. Testing materials are 40 mL septa sealed vials, a 1-10 µL syringe, fleuressence samples, and a zNose® vapor analyzer

Step one is draw 2 µL from Fleuressence vial into a clean syringe

Figure 7. Step one is draw 2 µL from Fleuressence vial into a clean syringe

Step two is to inject 2 µL of fleuressence through septa and into 40 mL to create vapor sample

Figure 8. Step two is to inject 2 µL of fleuressence through septa and into 40 mL to create vapor sample

Step 3 is to attach vial to inlet of zNose®. A sample needle works well for volatile compounds. Above C12 compounds begin to condense onto the walls of the relatively cool sample needle.

Figure 9. Step 3 is to attach vial to inlet of zNose®. A sample needle works well for volatile compounds. Above C12 compounds begin to condense onto the walls of the relatively cool sample needle.

Removing septa cap and pressing vial against Teflon face of zNose® inlet enables direct sampling of high molecular weight compounds by the 200 °C inlet of the zNose

Figure 10. Removing septa cap and pressing vial against Teflon face of zNose® inlet enables direct sampling of high molecular weight compounds by the 200 °C inlet of the zNose

Experimental Results

The column (a db624) was temperature-programmed to rise from 40 °C to 160 °C at 10 °C/second, for all vapor samples that were tested. The data acquisition (chromatogram) time was 20 seconds. Unless stated otherwise, the detector temperature of 60 °C.

Aldehyde

Two primary compounds with indices of 1104 and 1332 and with concentration counts of 4,519 and 11,114, respectively, were produced from the vapors from the aldehyde fleuressence base. Other significant compounds had indices of 1016, 1234, 1438, 1516, and 1584, at much lower concentrations.

The Vaporprint® image shows aroma concentration (radial) vs. retention time (angle) with 0 and 20 seconds at the top of the Figure 11.

Chromatogram of Aldehyde fleuressence

Figure 11. Chromatogram of Aldehyde fleuressence

Iceberg

The iceberg fleuressence consisted of a single major compound with a concentration of 20,007 counts and with an index of 1225. A minor secondary compound with a concentration of 1,800 counts and an index of 1062 was also identified.

Chromatogram of Iceberg fleuressence

Figure 12. Chromatogram of Iceberg fleuressence

Citrus

The citrus fleuressence consisted of a single major compound (limonene) with a concentration of 17,929 counts and with an index of 1057. Minor secondary compounds with indices of 947, 1002, 1275, and 1405 were also identified.

Chromatogram of Citrus fleuressence

Figure 13. Chromatogram of Citrus fleuressence

Dairy

The dairy fleuressence consisted of two major compound peaks with concentrations of 3,823 and 2,965 counts, and with indices of 1452 and 1583, respectively. Significant minor compounds with indices of 603, 1058, 1279, 1393, and 1651 were also identified.

The rounded portion of the Vaporprint® image is due to the presence of high molecular weight compounds.

Chromatogram of Dairy fleuressence

Figure 14. Chromatogram of Dairy fleuressence

Edible

The edible fleuressence consisted of two major closely-spaced compound peaks with concentrations of 2,752 and 3,832 counts and with indices of 1097 and 1120, respectively. Significant minor compounds with indices of 805, 965, 1274 and 1398 were also identified.

Chromatogram of Edible fleuressence

Figure 15. Chromatogram of Edible fleuressence

Fruit

The fruit fleuressence consisted of three major compound peaks with concentrations of 34,982, 22,317 and 19,439 counts and with indices of 1111, 1329, and 1461, respectively. Significant minor compounds with indices of 830, 1054, 1219, 1274, 1542, 1651, 1804, and 1994 were also identified.

Chromatogram of Fruit fleuressence

Figure 16. Chromatogram of Fruit fleuressence

Green

The green fleuressence consisted of a major compound peak with a concentration of 6565 counts and with an index of 910. Significant minor compounds with indices of 1093, 1120, 1224, 1269, 1339, 1402, 1443, 1511, 1565, 1637, 1723, and 1858 were also identified.

Chromatogram of Green fleuressence

Figure 17. Chromatogram of Green fleuressence

Herb

The herb fleuressence consisted of a major compound peak with a concentration of 47,348 counts and with an index of 1139. Significant minor compounds with indices of 693, 910, 952, 1010, 1062, 1209, 1283, 1407, 1556, and 1587 were also identified.

Chromatogram of Herb fleuressence

Figure 18. Chromatogram of Herb fleuressence

Iris

The iris fleuressence consisted of major compound peaks with concentrations of 15,717 and 72,716 counts and with indices of 1484 and 1533, respectively. Significant minor compounds with indices of 830 and 1366 were also identified.

Chromatogram of Iris fleuressence

Figure 19. Chromatogram of Iris fleuressence

Jasmine

The jasmine fleuressence consisted of a major compound peak with a concentration of 34,912 counts and with an index of 1224. Significant minor compounds with indices of 910, 1028, 1115, 1316, 1366, 1434, and 1470 were also identified.

Chromatogram of Jasmine fleuressence

Figure 20. Chromatogram of Jasmine fleuressence

Konifer

The konifer fleuressence consisted of a major compound peak with a concentration of 171,679 counts and with an index of 1254. Significant minor compounds with indices of 1049, 1111, 1190, 1456, and 1560 were also identified.

Chromatogram of Konifer fleuressence

Figure 21. Chromatogram of Konifer fleuressence

Linalool

The linalool fleuressence consisted of a major compound peak with a concentration of 73,221 counts and with an index of 1140. No significant minor compounds were identified.

Chromatogram of Linalool fleuressence

Figure 22. Chromatogram of Linalool fleuressence

Muguet

The muguet fleuressence consisted of major compound peaks with peak concentrations of 5,673, 8,228, 1,778, and 8,757 counts and with indices of 1136, 185, 1268, and 1359, respectively. Significant minor compounds with indices of 1100, 1527, 1591, 1749, and 1942 were also identified.

Chromatogram of Muguet fleuressence

Figure 23. Chromatogram of Muguet fleuressence

Narcotic

The narcotic fleuressence consisted of major compound peaks with peak concentrations of 7667, 2377, 3845, and 14,581 counts and with indices of 1136, 1185, 1213, and 1428, respectively. Significant minor compounds with indices of 1057, 1100, 1359, 1606, 1754, and 2085 were also identified.

Chromatogram of Narcotic fleuressence

Figure 24. Chromatogram of Narcotic fleuressence

Orchid

The orchid fleuressence consisted of major compound peaks with peak concentrations of 2436, 17386, and 38,257 counts and with indices of 1213, 1479, and 1601, respectively. Significant minor compounds with indices of 775, 905, 1057, 1113, 1140, 1419, 1532, and 1764 were also identified.

Chromatogram of Orchid fleuressence

Figure 25. Chromatogram of Orchid fleuressence

Phenolic

The phenolic fleuressence consisted of major compound peaks with peak concentrations of 4,962 and 3,577 counts and with indices of 1185, and 1222, respectively. Significant minor compounds with indices of 1596 and 2050 were also identified.

Chromatogram of Phenolic fleuressence

Figure 26. Chromatogram of Phenolic fleuressence

Balsamic

The balsamic fleuressence consisted of major compound peaks with peak concentrations of 775, 2639, 2166, and 1497 counts and with indices of 1053, 1109, 1213, and 1414, respectively. Significant minor compounds with indices of 1001, 1577 and 2070 were also identified.

Chromatogram of Balsamic fleuressence

Figure 27. Chromatogram of Balsamic fleuressence

Rose

The rose fleuressence consisted of a single major compound with a peak concentration of 21,176 counts and an index of 1190. Significant minor compounds with indices of 1140, 1268, and 1547 were also identified.

Chromatogram of Rose fleuressence

Figure 28. Chromatogram of Rose fleuressence

Spice

The spice fleuressence consisted of a single major compound peak a peak concentration of 53,246 counts and with an index of 1423. Significant minor compounds with indices of 1140, 1355, 1518, 1672, and 1772 were also identified.

Chromatogram of Spice fleuressence

Figure 29. Chromatogram of Spice fleuressence

Tar and Smoke

The tar and smoke fleuressence consisted of major compound peaks with peak concentrations of 2701, 5174, 5167 and 45,465 counts and with indices of 1095, 1163, 1268 and 2139, respectively. Significant minor compounds with indices of 709, 790, 905, 974, 1029, 1057, 1364, 1449, 1493, 1532, 1562, and 1596 were also identified.

Chromatogram of Tar and Smoke fleuressence

Figure 30. Chromatogram of Tar and Smoke fleuressence

Animalic

The animalic fleuressence consisted of major compound peaks with peak concentrations of 16,484, 12,631, 12,300, 2,027, 3,871, 9,207 and 8,944 counts and with indices of 1140, 1433, 1483, 1517, 1557, 1774, and 2085, respectively. Significant minor compounds with indices of 919, 1006, and 1057 were also identified.

Chromatogram of Animalic fleuressence

Figure 31. Chromatogram of Animalic fleuressence

V. Vanillin

The vanillin fleuressence consisted of major compound peaks with peak concentrations of 9,548 and 3,891 counts and with indices of 1433 and 1577, respectively. Significant minor compounds with indices of 1527 and 2085 were also identified.

Chromatogram of Vanillin fleuressence

Figure 32. Chromatogram of Vanillin fleuressence

Wood

The wood fleuressence consisted of major compound peaks with peak concentrations of 38,447 and 34,171 counts and with indices of 1483 and 1774, respectively. Significant minor compounds with indices of 1053, 1145, 1581, and 1700 were also identified.

Chromatogram of Wood fleuressence

Figure 33. Chromatogram of Wood fleuressence

Musk

The musk fleuressence consisted of major compound peaks with peak concentrations of 5,572 and 6,570 counts and with indices of 1591 and 1685, respectively. Significant minor compounds with indices of 983, 1086, 1140, 1226, and 1305 were also identified.

Chromatogram of Musk fleuressence

Figure 34. Chromatogram of Musk fleuressence

Yeast-Mossy

The yeast-mossy fleuressence consisted of major compound peaks with peak concentrations of 19,575, 1,190 and 60,211 counts and with indices of 1209, 1234, and 2084, respectively. Significant minor compounds with indices of 924, 1036, 1081, 1097, 1429, 1678, and 1750 were also identified.

Chromatogram of Yeast-Mossy fleuressence

Figure 35. Chromatogram of Yeast-Mossy fleuressence

Zolvent

This solvent base mix generates a much lower concentration aroma than the fleuressence bases. Hence, the analysis was performed with a 10-second sample of headspace vapors from 3 mL of zolvent in a 40 mL vial, and a 20 °C detector.

Chromatogram of Zolvent fleuressence

Figure 36. Chromatogram of Zolvent fleuressence

Summary of Vaporprint® Olfactory Images

Summary of fleuressence olfactory images

Figure 37. Summary of fleuressence olfactory images

Channel Number 5 Perfume

This popular perfume was tested using the same method employed on the fleuressence bases e.g. 1-second sample, 2 µl in 40 mL vial, 10 °C/second column ramp rate, and 60 °C detector.

The perfume consisted of a major compound peak with a concentration of 4,578 counts and with an index of 1,135. Several significant minor compounds were also clearly evident but some did not easily separate after employing this fast method.

Chromatogram of Channel No. 5

Figure 38. Chromatogram of Channel No. 5

Slowing the analysis method by employing a 3 °C/second column ramp enabled the perfume compounds to be better defined and separated.

The sensitivity to volatile compounds was enhanced by reducing the detector temperature to 20 °C. This analysis demonstrates approximately 18 minor compound peaks other than the primary aroma compound (index 1130).

Chromatogram of Channel No. 5 aroma using slower method

Figure 39. Chromatogram of Channel No. 5 aroma using slower method

Expanded scale showing trace elements of Channel No. 5 perfume

Figure 40. Expanded scale showing trace elements of Channel No. 5 perfume

Conclusion

An ultra-high speed gas chromatograph called the zNose® was used to judge chemical profiling of aroma fleuressence samples in order to represent 25 basic olfactory notes.

The profiling has been shown to be quantitative and fast. The quality can be quantitatively measured in a quick and efficient manner, and the chemical signature of perfumes created by trained perfumery technicians can be compared. A convenient method of identification is the indexing of retention times for target compounds employing an n-alkane perfume standard. This method allows for instrument-independent chemical libraries and eliminates the requirement for several chemical standards.

Dynamic headspace analysis employing ultra-high speed gas chromatography can be combined with sensory data to evolve an objective method to classify perfumes by olfactory images or fleuressence notes.

The sensory data and chemical image can be exposed to pattern recognition employing principal component analysis (PCA), multivariate analysis, and partial least squares (PLS) methods in order to model human perception or determine perfume classifications. While evaluating quality, it may be useful to have a proper choice of samples and employ optimized variables as well as to preprocess chemical data, including scaling, transformation and normalization.

The zNose® provides perfumery experts the portability, speed, accuracy and precision required for cost-effective quality control measurements. These measurements can easily be validated by independent laboratory testing, as they are based upon well-known chromatographic methods.

A ‘good’ perfume aroma if determined by a trained perfumery technician can be measured in near real-time. After the ‘good’ chemical signature is defined, objective and quantitative quality control testing can be performed, with other zNose® analyzers incorporated into perfume production process often located in isolated geographical locations.

A happy perfumery technician

Figure 41. A happy perfumery technician

This information has been sourced, reviewed and adapted from materials provided by Electronic Sensor Technology.

For more information on this source, please visit Electronic Sensor Technology.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Electronic Sensor Technology. (2019, July 18). Quantitative Determination of Perfume Notes Using an Electronic Nose. AZoM. Retrieved on December 12, 2019 from https://www.azom.com/article.aspx?ArticleID=13814.

  • MLA

    Electronic Sensor Technology. "Quantitative Determination of Perfume Notes Using an Electronic Nose". AZoM. 12 December 2019. <https://www.azom.com/article.aspx?ArticleID=13814>.

  • Chicago

    Electronic Sensor Technology. "Quantitative Determination of Perfume Notes Using an Electronic Nose". AZoM. https://www.azom.com/article.aspx?ArticleID=13814. (accessed December 12, 2019).

  • Harvard

    Electronic Sensor Technology. 2019. Quantitative Determination of Perfume Notes Using an Electronic Nose. AZoM, viewed 12 December 2019, https://www.azom.com/article.aspx?ArticleID=13814.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit