Aroma analysis of vegemite 3. Data analysis and interpretation

This is Part 3 of my vegemite aroma analysis series. If you haven’t read Parts 1 and 2, you can do so here and here. In a wonderful complement to my vegemite posts, Vittorio Saggiomo of Labsolutely has also done NMR and a microscopy video of marmite.

As a reminder, here is the output from the GC of the vegemite aroma analysis. There are many detection systems that can be coupled to the end of a GC which provide different kinds of information about the compounds which are analysed. In this case, the detector I used was a mass spectrometer (MS, or mass spec for short).

chrom_zoom in

Mass spec is the most useful type of detector because it provides information about the structure of each molecule exiting the GC column, and often this information is good enough to deduce the identity of the chemicals in the separated mixture. At each point along the chromatogram (many times per second) the spectrometer collects a mass spectrum of the compounds present, which look something like this.


The mass spectrum is generated when the compound is ionised by electrons created within the mass spectrometer. Ionisation causes the molecules to fragment, and the mass and abundances of the fragments are measured by the spectrometer and provide key pieces of information about the chemical’s structure. In the spectrum above, each vertical line represents a molecular fragment and its relative abundance, with the number above it representing the molecular weight. Modern mass specs come with software that automatically searches huge libraries of compounds and finds likely matches to the spectrum collected. However, it is possible (and -[NERD ALERT]– fun!) to work it out by hand.

The matches collected by the software to the mass spectra of the vegemite aroma compounds, and their relative abundances are tabulated below.

% total compound Associated odour
1.7 limonene citrus, fruity, mint
0.2 benzaldehyde almond, burnt sugar
0.8 benzeneacetaldehyde cocoa, honey, spice, rose, lilac
0.1 phenylethyl alcohol floral, rose
0.7 octanoic acid fatty, pineapple, banana, sweat, cheese
12.5 octanoic acid, ethyl ester fruity, fatty, floral, green, menthol, anise
0.1 2-phenylethyl acetate floral, honey
0.2 2-phenyl-2-butenal cocoa, floral, musty
0.5 sulfurol sulfur, meaty, chicken broth
0.1 nonanoic acid, ethyl ester fruity, rose, wax, rum, wine
0.1 p-isopropenylphenol phenol, antiseptic
2.1 1-decanol fatty
1.6 n-decanoic acid fatty, rancid
16.6 ethyl trans-4-decenoate wax, leather, pear
2 n-decanoic acid fatty rancid
43.8 ethyl decanoate fruit, oil, sweet, wax
0.2 niacinamide odourless
0.2 decyl acetate floral, orange, rose
1.2 caryophyllene spice, wood, cloves
0.2 3-methylbutyl octanoate apple, coconut, grass, pineapple
0.2 farnesene apple
0.8 beta selinene woody, herbaceous, peppery
0.2 5-methyl-2-phenyl-2-hexenal cocoa, mocha
0.7 alpha selinene amber, orange, pepper
2.5 2,3-dihydrofarnesol floral, green
1 dodecanoic acid fatty, soapy
4.3 ethyl laurate floral, soapy, wax, peanut
0.3 isopentyl pentadecanoate floral, wine
0.1 1-(2,4-dimethylphenyl)-propanone no info
0.9 tetradecanoic acid flowery, woody
0.2 trans-nerolidyl formate wax, floral
0.1 farnesol acetate flowery, green rose
0.5 cis-9-hexadecenoic acid wax, old-person smell
0.5 n-hexadecanoic acid wax
0.8 ethyl 9-hexadecenoate wax
0.4 hexadecanoic acid ethyl ester rancid
0.1 octyl 2-phenylethyl ester oxalic acid citrus, fruity mint

Based solely on the aroma descriptors I was able to find in online odour chemical databases, I think that the chemical ‘sulfurol’ probably contributes significantly to the odour of vegemite.


Chemical structure of sulfurol (2-(4-Methyl-1,3-thiazol-5-yl)ethanol).

Other compounds of interest that were detected in the aroma analysis are;

  • Niacinamide: a derivative of one of the B-vitamins that vegemite is loaded with.
  • Caryophyllene: a compound common which contributes a peppery spiciness.
  • Hexadecenoic acid: notable for the fact that ‘old person smell’ is attributed to this compound!

The compounds revealed in the analysis are not an exhaustive list of all of the chemicals contributing to the aroma of vegemite. There were many more small peaks in the chromatogram that I did not search the mass spectrum of. It’s also likely that some of the chemicals that contribute to the aroma of vegemite, do not give a visible signal in the chromatogram, or were not picked up by the SPME fibre.

It’s interesting to note that many of the aromas in the table above are described as sweet, fruity or flowery, which are certainly not words you would use to describe the aroma of vegemite. There could be several explanations for this;

  • The odour thresholds for these chemicals may be quite high. That is, they may have to be present in large amounts in order for the odour to be detected.
  • When odours from different chemicals are mixed together, the whole may not be equal to the sum of the parts.
  • Some compounds smell different depending on their concentration, or may even vary from person to person.
  • The mass spec library searching program may not have been able to correctly identify some of the chemicals.

More specialised aroma analysis by GC can include the use of an olfactory detection port (ODP). Here, once the mixture is separated by the GC, the effluent is split in two with half going to a conventional detector (such as an MS) and the other half to the ODP or ‘sniffing port’ where an aroma analyst can smell what is exiting the column and assign odours to specific compounds.


Olfactory detection port from

I hope you’ve enjoyed this 3-part series on the chemical analysis of vegemite aroma. Thanks once again to Chris Slape for the inspiration, and if you have any questions, comments or ideas for future analyses 🙂 please leave them below.

Aroma analysis of vegemite 2. Separation of vegemite aroma chemicals by gas chromatography

This is Part 2 of my vegemite aroma analysis series. If you haven’t read Part 1, you can do so here.

Now that the vegemite aroma compounds have been extracted from the sample, and onto the SPME fibre, they will be separated using gas chromatography (GC). GC is a versatile analytical technique with many applications in areas like environmental science, forensics and petrochemicals. Separation of mixtures is a really useful way of finding out what is in them by splitting them up into their component parts. Lots of different kinds of samples of varying complexity can be analysed with GC, the main limitation being that they must have a boiling point below about 400°C. GC enables the analysis of many complex mixtures by separating them out into their individual component chemicals. The separation happens in a very long (usually 15 – 100 metres) and thin tube called a column. The column is housed in an oven which increases temperature over time and this facilitates one mode of separation. The temperature ramping up over time causes chemicals to travel through the column faster if they have a lower boiling point and slower if they have a higher boiling point. The other influencer of separation is a polymer coating applied to the inside of the column walls (very similar to the coatings on SPME fibres mentioned in Part 1). Different chemicals will interact differently with particular coatings and this also affects the separation. These two interactions give the analyst very useful information with respect to the chemical properties of the mixture components.

2013-05-31 14.42.13a

A gas chromatograph, with the oven door open. Green arrow = injection port, blue arrow = column, red arrow = outlet to detector.


Now, back to the vegemite! In order to get the aroma chemicals trapped on the SPME fibre into the GC, the fibre is inserted into the heated injection port. 

2013-06-11 13.08.21

The high temperature of the injector causes the aroma compounds to become gaseous and desorb from the SPME fibre. A flow of helium gas sweeps the molecules out of the injector and onto the GC column for separation. The output of the GC following separation of the vegemite aroma is below.

chrom_zoom out

There are maybe four main peaks in the chromatogram, representing the compounds which were most concentrated on the SPME fibre. But a zoomed in view of the chromatogram reveals many, many more peaks. Each peak represents at least one different chemical comprising the aroma of vegemite.

chrom_zoom in

In the next post, I will discuss how we go about discovering which chemicals these peaks in the chromatogram are, and what contribution they might have to the distinctive aroma of Australia’s favourite spread.

Aroma analysis of vegemite 1. Sample preparation and aroma sampling

Vegemite: that salty, yeasty, vitamin B-rich, Australian sandwich spread. Its taste and odour are distinctive, iconic and entrenched in the Australian cultural identity. But what are the chemical compounds that make up the unique smell of vegemite? This is a topic close to my heart (or nose), as I drive past the Kraft factory where vegemite is manufactured on my way to work, so the aroma often fills my nostrils and makes me crave a thick slice of sourdough toast slathered with butter and vegemite. I was shocked to find that this area of research into sensory analysis of vegemite is sorely lacking in the scientific literature, so I decided to do my own experiment* which I will take you through in a series of three blog posts.

2013-06-09 08.53.27


The aromas we smell from foods, fragrances or anything with an odour, comes from chemicals that are released from the object into the surrounding atmosphere and then into our noses. In order to escape from the object into the air (and noses), these aroma chemicals must be ‘volatile’, where volatile means the compounds do not require much energy to enter the gas phase and may do so spontaneously, or with a little heat or pressure.

In order to do the analysis of the aroma of vegemite I will be using two popular analytical chemistry techniques: solid phase microextraction (SPME) and gas chromatography-mass spectrometry (GC-MS). These will be discussed later on.

  1. Sample preparation

A sample of vegemite was taken directly from a refrigerated** 600 gram jar purchased from a supermarket and manufactured in Melbourne, Victoria. The sample was transferred by means of an ordinary household knife to a 20 mL glass headspace sampling vial with rubber septum. The weight of the vegemite sample transferred to the sampling vial was approximately 2 g. At the same time, a sample of vegemite from the same jar was applied to buttered toast in order to ensure the sample was of the acceptable quality.***

2013-05-27 16.40.14

Headspace sampling vials will generally be of a much larger volume than the sample. The sample only takes up a small part of the vial, with the remaining volume available for the formation of an atmosphere concentrated with volatile compounds emitted from the sample. The vial lid has a rubber septum which permits a sampling needle to pierce the top while the vial remains sealed, with no gases escaping.

2013-05-28 09.27.59

Headspace sampling vial showing rubber-lined cap and tool to crimp cap and seal the atmosphere inside.

  1. Aroma sampling by solid phase microextraction (SPME)

The SPME sampler looks like this…

2013-05-28 10.02.29

… and here it is taken apart…

2013-05-28 10.05.10

The parts of a SPME sampler (left to right): body, needle/fibre assembly, retaining screw, depth gauge.

The needle-like part of the SPME sampler houses what we call a ‘fibre’. The fibre is made from fused silica or stainless steel and coated with a thin layer of polymer. The fibres can be made with different polymer coatings which can help you analyse different types of volatile compounds. The type of polymer coating is indicated by the colour of the needle hub (in this case red is polydimethylsiloxane). When the SPME sampler is assembled, pressure applied to the spring allows the delicate fibre to be either exposed or protected by the needle.

To encourage maximum release of volatile chemicals from the vegemite sample, the vial was placed in a laboratory oven set to 65 °C for about half an hour and allowed to come to thermal equilibrium.  In order to sample the aroma, the SPME sampler needle housing the fibre is used to puncture the vial septum. Once the needle is pushed through into the vial, the delicate fibre can be exposed. When the fibre is exposed to an atmosphere of volatile chemicals, they become temporarily trapped on the polymer coating. The extraction of volatile compounds from the vial onto the fibre only needs a few minutes, in this case I did the extraction of the vegemite aroma for 15 minutes.

2013-05-31 15.55.06

The SPME sampler inserted into the vial containing vegemite in a laboratory oven.

2013-05-31 15.55.45a

Close up of SPME sampler inserted into the vegemite-containing vial. You can just see the thin fibre circled in red.

Once the extraction of the aroma chemicals is complete, the next stage of the experiment can begin. This will be the gas chromatography-mass spectrometry and will be discussed in the next post.




*Although all of the work and analysis was done by me, full credit for the idea to do this goes to fellow practitioner of Teh Scients Dr Chris Slape (@is_chris).

**I am of the opinion (and so is Kraft) that it is entirely unnecessary to keep one’s vegemite in the fridge. However, storing vegemite in the fridge provides me with the benefits of accessibility and domestic harmony, so there it lives.

***It was.