Thiols and beyond
The science of Sauvignon Blanc

[This is a written version of a presentation I made at Sauvignon 2016 in Blenheim, New Zealand]

It would be fair to consider Sauvignon Blanc as a model system for understanding wine flavour. Over recent years, the New Zealand wine industry has funded a significant research project that has looked at all the whole circle of the science of Sauvignon, from grape growing, to winemaking, to perception of wine. It has revealed a lot about the connections between terroir, viticulture, winemaking and wine flavour, and there's no other grape variety that we understand so much about. And one group of chemicals in particular has become newly trendy: thiols. Five years ago, no one had ever heard of them. Now, people are putting 'thiols' in their tasting notes. Does anyone really know what is meant by this term?

The simple story: what makes Marlborough Sauvignon so distinctive?
There's a simple story, and this is a good place to start in understanding the success of Kiwi Sauvignon. It's not a completely accurate story, and there are some missing details, but this is a good place to start.

The two impact compounds important in Marlborough Sauvignon are methoxypyrazines and thiols. Marlborough Sauvignon is particularly high in these two elements. Methoxypyazines are responsible for green flavours: think green pepper, tomato leaf, grassiness. And thiols are responsible for passionfruit, grapefruit and tropical fruit aromatics. So good Marlborough Sauvignon is about greenness allied to thiol aromatic interest. This is what sets it apart.

What is a thiol?
So what is a thiol? Also known as mercaptans, thiols are sulfur-containing organic compounds with a sulfur atom bound to a hydrogen atom, which creates the SH group ('sulfhydryl' or 'thiol' group). Sulfur containing volatile compounds are often very smelly, and there are a whole group of them that, in wine, contribute to fruit aromas (e.g. blackcurrants, grapefruit, passionfruit and guava), and they also play a role in the aromas of other foods and drinks, such as roasted coffee, popcorn, grilled meat and beer.

Sauvignon Blanc has three thiols that are important in varietal aroma:


Thiol Sensory characteristic Perception threshold in wine (ng/litre) Concentration found in NZ Sauvignon (ng/litre)


passionfruit skin/stalk

60 100-20 000





4 (below 100,

doesn't dominate

above 100,

impact compound)

4MMP Broom, cat's pee 0.8 2-50


Factors affecting thiol levels
So which factors affect thiol levels, and if they are desirable how can their concentrations be increased? And why are they so high in Marlborough Sauvignon Blanc?

Thiols aren't present in the grapes. Instead, they are formed from precursors present in grape must by the action of yeasts. And only a portion of the precursor pool in the grapes is transformed into thiols by the yeasts. It's not simply a question of getting the right yeast strain: if it were that simple, anyone could make high thiol wines. Yeasts do have some effect, but it is limited.

There are many factors that affect the levels of precursors in the grapes, and it isn't a simple story. There's the unique high UV environment that New Zealand's vineyards enjoy. There are site influences, and vintage influences – some sites typically produce higher thiol Sauvignons, and sometimes a site produces a high thiol Sauvignon one year and not another. There's a well known phenomenon where the first crop off a site often wins lots of medals because of high thiol levels. And there's the interesting observation that machine-picked fruit typically has around 10 times the level of thiols than handpicked fruit. What is it about machine harvesting that has this effect? Could it be the disruption of picking sets some enzymatic changes in place which then result in the rise of thiol precursors in the must? Or, a more novel idea: could machine harvesting be causing the release of semiochemicals (volatile alarm signals) that then signal to nearby vines upstream of the block being harvest, that then results in physiological changes, raising the levels of thiol precursors.

Producing Sauvignon with elevated thiols
So how, practically, can you make a Sauvignon with elevated thiol levels? We still don't know exactly which environmental factors affect thiol precursor levels. But the key is converting more of those precursors into the actual thiols during fermentation. There seem to be two important ways of doing this. First, it is important to protect the C6 green leaf volatiles – precursors of thiols – by getting SO2 in early. The second is to work with elevated levels of H2S (hydrogen sulfide) very early in ferments. One way of achieving this is to start a ferment that you deliberately allow to get stinky with lots of H2S, such as by working with full solids, stressing the yeasts, having low nitrogen levels and then piping this stinky H2S into the main ferments at an early stage. This toying with reduction is really interesting.

Preserving thiols
So, once you have these thiols, what can you do to preserve them? The key thing is temperature of storage, because the very aromatic passion fruit thiol 3MHA is transformed back to 3MH by acid hydrolysis. If Sauvignon is stored at 10 C then after two years you will have lost half your 3MHA. If you store it at 20 C then you lose half every two months. So shipping temperatures are important, and you might want to store wine cool and have several bottling runs.

The complex story
This is the simple story. But there's a more complex story, too. Humans are not measuring devices; green is more than just methoxypyrazines; and thiols aren't quite as important as we thought they were.

Wine flavour chemistry
First, we need to think a bit about wine flavour chemistry. The wine is a whole, and we need to consider the flavour molecules in wine together. It's possible to split wine aroma chemicals into three main groups: global wine aroma, contributory compounds and impact compounds.

Global wine aroma
This is the set of molecules found in almost all wines which gives the base wine aroma. Wine smells of wine!

Ethanol (don’t underestimate this one!)

Higher alcohols (e.g. butyric, isoamylic, hexylic, phenylethylic)

One present in grapes (β-damascenone); rest produced by metabolism of yeasts

20 aromatic chemicals in all wines make global wine odour

Acids (acetic, butyric, hexanoic, octanoic, isovaerianic)

Ethyl esters from fatty acids


Contributory compounds
These contributory compounds are a group of 16 compounds present in most wines, but at low levels. They frequently occur below the detection threshold, but they have an effect that is synergistic with other aroma compounds.

Impact compounds
Then we have the impact compounds. Not all varieties have these, but many do. They are aroma molecules with a strong imprint on the wine. These are some examples: 

Polyfunctional thiols (mercaptans) - 4MMP, which has a box tree aroma (4.2 ng/litre detection threshold), 3MHA, which has a tropical fruit scent (60 ng/litre) and 3MH

Rotundone: gives pepperiness to Syrah, at incredibly tiny concentrations

Methoxypyrazines: 2-methoxy-3-isobutylpyrazine (MIBP) is the key one

Monoterpenes, such as linalool, floral, citric aromas

Rose-cis oxide: characteristic of Gewürztraminer

The non-volatile wine matrix
These are wine constituents that don’t have any aromatic characteristic of their own influence strongly the way that the various aromatic molecules present in wine are perceived. Work by Vicente Ferreira has looked at this, with surprising results. If you take white wine aromatics and add them to the non-volatile wine matrix of a red wine, the wine will smell like a red wine, and vice versa. The matrix is having a powerful effect on the smell of the wine, even though it doesn't smell itself.

The significance? It's important to consider the wine as a whole. While we sometimes need to break wine down into its components to understand it, we need to be aware that all these aroma compounds act together to create wine flavour. This is where reconstitution experiments are interesting.

Reconstitution experiments
Work by Frank Benkwitz, Laura Nicolau and colleagues has examined Sauvignon aroma by looking to see which the important aroma compounds are, and then taking a deodorized wine base and adding these compounds at their original concentrations to create a model wine. The power of this approach is you get to look at how aroma compounds interact in the context of the wine. You can take out one compound at a time, or in groups, or in any combination you want. Some of the results are very interesting.

The terpenes linalool and α-terpineol had a huge impact when omitted, even though they are present at relatively low concentrations. If esters are removed, there is a small drop in intensity for most of the descriptors, and a large decrease in ‘passion fruit skin stalk’ and ‘sweet sweaty passion fruit’, both of which were previously thought to be associated with thiols. What these results indicate is that the thiol story is just a little bit too simple, and we need to think about interactions among the different flavour molecules.

One further twist. Green is more than just methoxypyrazine. Recent results indicate that the green pepper character comes from the green-smelling C6 compounds such as hexanal, as well as the methoxypyrazines. Also, the mix of thiols and other compounds can also add to greenness. Asparagus aroma has an influence from dimethyl sulfide.


What can we conclude? Thiols and methoxypyrazines are elevated in Marlborough Sauvignon, and they are important in helping produce such distinctive wines, with green allied to lovely fruity aromatics. But there's more to Sauvignon than just these two impact compounds. This is a list of some of the important compounds in Sauvignon aroma.

  • Thiols

  • Esters

  • Methoxypyrazines

  • Higher alcohols

  • Terpenes

See also:

Carbonic maceration
How wine is made

Find these wines with


Back to top