Removal of volatile sulfur compounds

Removing unwanted ‘reductive’ volatile sulfur compounds from wine prior to packaging

What are volatile sulfur compounds?

A range of volatile sulfur compounds (VSCs) can be present in wine, with some contributing positive ‘fruity’ characters, and others responsible for unwanted ‘reductive’ aromas such as ‘rotten egg’, ‘natural gas’ or ‘onion’. If these ‘reductive’ characters are seen in finished wine, they are generally considered winemaking faults. They can form during fermentation and are typically also managed during fermentation; however, they can sometimes persist post-fermentation. If such characters are evident in wine prior to packaging, they need to be addressed before packaging commences.

Common examples of ‘reductive’ volatile sulfur compounds

Hydrogen sulfide
Most winemakers will be familiar with the aroma of hydrogen sulfide (H2S) or ‘rotten egg gas’. The detection threshold of H2S in wine is about 1 – 2 µg/L (parts per billion) and it has been reported that concentrations below this threshold might play a role in wine complexity. During alcoholic fermentation, yeast cells excrete H2S into the fermenting juice when placed under stress; for example, when the yeast starts to run out of nitrogen. Australian grape juices can be low in nitrogen and winemakers often supplement the juice with a soluble nitrogen source, such as diammonium phosphate.

Hydrogen sulfide can also be produced in excess by yeast during fermentation due to the presence of elemental sulfur on grape skins (from sulfur sprays), inadequate levels of free α-amino nitrogen (FAN), added SO2, a deficiency of B-complex vitamins (pantothenic acid or pyridoxine), unusually high levels of cysteine in juice or a high concentration of metal ions. The production of H2S is also yeast strain dependent.

Hydrogen sulfide may react with other wine compounds to produce larger molecular weight sulfur compounds; it is therefore best to remove H2S before it reacts further to produce other unwanted odour-active compounds.

Unwanted sulfhydryls
Sulfhydryls are compounds that contains a -SH group. These compounds are also known as ‘thiols’ or ‘mercaptans’, with aromas variously described as ‘cabbage’, ‘garlic’, ‘onion’ and ‘rubber’. Both methanethiol (methyl mercaptan, MeSH) and ethanethiol (ethyl mercaptan, EtSH) are formed directly as a result of yeast metabolism, and can also be produced in wine post-bottling by the breakdown of methyl thioacetate and ethyl thioacetate, respectively. The aroma of methanethiol (MeSH) is described as ‘rotten eggs’ and ‘cabbage’, and it has a sensory threshold of 0.02 – 2.0 µg/L. The aroma of ethanethiol (ethyl mercaptan) is described as ‘onion-like’ and ‘rubber-like’. The sensory threshold for ethanethiol is 1.1 µg/L.

Disulfides
Sulfhydryls such as EtSH and MeSH can be rapidly oxidised to produce symmetrical or asymmetrical disulfides. Examples of symmetrical disulfides are diethyldisulfide (DEDS) and dimethyldisulfide (DMDS), respectively. The aroma of DMDS is described as ‘onions’ and ‘cooked cabbage’ and its sensory threshold is 29 µg/L. The aroma of DEDS is described as ‘burnt rubber’ and ‘garlic’ and its sensory threshold DEDS is 4.3 µg/L.

Dimethyl sulfide
Dimethyl sulfide (DMS) is one of the major compounds found in aged wines and is formed during maturation. It is produced from the breakdown of a sulfur-containing amino acid called S-methyl methionine. At low concentrations it might contribute to the body of aged white wines and has a ‘vegetable’, ‘truffle’, or ‘blackcurrant’ character. At higher concentrations, the aroma of DMS is perceived as a fault and is described as ‘asparagus’, ‘cooked corn’, ‘cooked tomato’ or ‘molasses’. The sensory threshold for DMS is between 30 and 60 µg/L.

The table below contains aroma thresholds, common sensory descriptors and ranges for VSCs (Siebert et al. 2010).

Low MW sulfur compound Odour descriptor  Aroma threshold (µg/L) Concentration ranges found in wine (µg/L)
Hydrogen sulfide Rotten egg, sewage-like 1.1 -1.6 nd*-35
Methanethiol Rotten, cabbage, burnt rubber, putrefaction 1.8 -3.1 nd-8
Ethanethiol Onion, rubbery, burnt match, sulfidy, earthy 1.1 nd-1
Dimethyl sulfide Blackcurrant, cooked cabbage, asparagus, canned corn, molasses 25 11-760
Carbon disulfide Sweet, ethereal, slight green, rubber, sulfidy, chokingly repulsive >38 nd-45.1
Diethyl sulfide Garlic, rubbery 0.9 nd-0.5
Methyl thioacetate Sulfurous, cheesy, egg 50 nd-18
Dimethyl disulfide Vegetal, cabbage, intense onion-like 29 nd-1.5
Ethyl thioacetate Sulfurous, garlic, onion 10 nd
Diethyl disulfide Bad smelling, onion 4.3 nd

nd = not detected

Using copper to remove VSCs prior to bottling

The use of copper to treat volatile sulfide issues in wine is a common winemaking practice. Copper sulfate (CuSO4) fining can be used to remove sulfhydryls such as H2S, MeSH and EtSH from red and white wines. Copper sulfate reacts with these compounds to form copper salts, which were previously thought to be insoluble and easy to remove via filtration. However, recent studies by Kreitman et al (2016) and Clark et al. (2018) have shown that copper sulfide does not precipitate out after reacting as once was thought, but may act as a dissolved species, a large proportion of which may remain in the wine post-filtration.

If a wine containing sulfhydryls (mercaptans) is aerated to remove a suspected ‘reductive aroma’ fault, these compounds can be oxidised to disulfides, which do not react with copper and therefore cannot be removed by copper fining. Removal of disulfides requires the creation of reducing conditions, by the addition of ascorbic acid and SO2, in order to reduce these compounds back to the reactive species (MeSH and EtSH ), which may then be removed by treatment with copper.

The AWRI recommends that before any copper additions are made to treat ‘reductive’ characters that the current copper concentration in the wine be analysed. The diagnostic copper/cadmium test should also be performed, to establish which volatile sulfur compound(s) are present.

Careful laboratory trials should precede any CuSO4 additions to bulk wine, as an instability can result if the copper concentration in the wine exceeds approximately 0.5 mg/L (even lower in some wines). While there is no specified maximum limit for copper in wine in Australia, there are limits in many of Australia’s export markets (e.g. USA 0.5 mg/L), which should be taken into account when making copper additions.

Pre-packaging additions of copper should be conducted with both knowledge of the current level of copper in the wine and with enough time (around one week) for the wine to self-stabilise in tank prior to packaging. The AWRI recommends keeping copper concentrations below 0.5 mg/L. Any haze that develops can either be left behind when racking or removed by filtration during bottling.

Calculating the correct amount to add

The AWRI helpdesk has investigated many problems over the years from over-addition of CuSO4. Confusion can occur when a winemaker requests an addition of copper to a wine. Is the addition referring to the amount of copper (the exact copper ion concentration) or the amount of CuSO4 (of which copper forms only 25% by molecular mass)? Copper additions can be calculated using the AWRI copper sulfate addition calculator, which is accessible via the AWRI Winemaking Calculators app or from the calculators page on the AWRI website.

Avoiding copper instabilities

Another common winemaking mistake is the last-minute addition CuSO4, often on the day of packaging or even hours before. Late copper additions (even small additions of 0.1 – 0.2 mg/L) can increase the copper concentration in the wine to a level where instability occurs post-packaging. Increased residual copper concentrations have also been associated with an increased risk of developing H2S and MeSH in wines post-bottling (Ugliano et al. 2011, Viviers et al. 2013). Wines with concentrations of copper above 0.5 mg/L are strongly susceptible to instability (Rankine 1989). In the majority of copper-related investigations conducted by the helpdesk, the level of copper in the wine prior to the ‘late addition’ was unknown, and the late addition often saw the residual copper level reach or exceed 0.5 mg/L. In these cases, a haze often appears immediately post-packaging after the wine has been exposed to some oxygen during the packaging process. A copper haze typically consists of a copper-protein complex. Copper instabilities are one of the most common metal instabilities found in wine. The precipitated deposit or haze from a copper unstable wine is commonly referred to as copper casse, and can lead to an obvious haze as shown in the example below. More information about copper instabilities can be found in the Ask the AWRI article: Understanding copper hazes in wine.

An example of a copper casse

Copper removal options

If an over-addition of copper has occurred or high copper levels remain after fining to remove VSCs, then there are fining options available to winemakers. Polyvinylimidazole/polyvinylpyrrolidone (PVI/PVP) is a cross-linked polymer or absorbent resin which has great metal binding capability and approval for use in the Australian wine industry. Work by Mira et al. (2007) showed that PVI/PVP has the capacity to remove metals ions including copper, iron, lead, cadmium and aluminium from both red and white wine. It was further reported that higher fining rates removed higher levels of metals and that the highest copper removal occurred in red wines. At the applied rates (30 and 50 g/hL) the wines’ sensory characteristics were not significantly affected, although it was reported the concentrations of phenolics were reduced during fining.

If small amounts of copper need to be removed, then bentonite might be an option. Bentonite does not have the capacity to remove large amounts of copper; however, an addition of 0.1 g/L in either red or white wine will usually lower the copper concentration by about 0.1 – 0.2 mg/L. It should be noted that bentonite treatment can introduce other metals such as aluminium and iron, which may have other detrimental sensory impacts. Another option which is less desirable due to the toxicity of the lees produced, is potassium ferrocyanide (PFC) fining (often known as ‘blue fining’). Please contact the AWRI helpdesk for further information on removal of copper from wine.

Volatile sulfur compound remediation strategies during fermentation

Various techniques are used by winemakers to remove and remediate VSCs during fermentation. These include:

  • Diammonium phosphate (DAP) addition – Nitrogen supplementation can decrease the risk of slow or stuck fermentations, decrease the development of undesirable ‘reductive’ characters, and modify wine style by increasing the ‘fruity’/‘ester’ profile. When attempting to remove H2S towards the end of fermentation, it is preferable to add CuSO4 rather than DAP. Information on recommended YAN levels in grapes/musts/juices can be found on the Yeast Assimilable Nitrogen page.
  • Oxygen – introduction of oxygen during fermentation via sparging once 20% of the fermentable sugar has been consumed has been shown to reduce the potential sulfide formation in red wines. More information can be found in the Aeration of ferments page and the  Ways to introduce oxygen into an active red ferment fact sheet.
  • Copper – addition during fermentation or towards the end of fermentation.

References and further information

Frequently asked questions on copper

Managing ‘reductive’ aromas in wines (AWRI webinar, 5 November 2019)

The formation and remediation of stinky sulfur aromas in wines (AWRI webinar 31 March 2018)

Bekker, M., Espinase Nandorfy, D., Kulcsar, A., Faucon, A., Smith, P., Krstic, M. 2021. Choosing the best remediation strategy to remove ‘reductive’ aromas. Wine Vitic. J. 36(1): 42-45.

Bekker, M., Wilkes, E., Smith, P. 2018. How much do potential precursor compounds contribute to ‘reductive’ aromas in wines post-bottling? AWRI Tech Rev. 232: 5-9.

Coulter, A. 2022. Ask the AWRI: Stinky sulfur compounds in wineAust. N.Z. Grapegrower  Winemaker (704): 85-86.

Coulter, A. 2023. Ask the AWRI: Understanding copper hazes in wine. Aust. N.Z. Grapegrower Winemaker (714): 78-79.

Kreitman, G.Y., Danilewicz, J.C., Jeffery, D.W., Elias, R.J. 2016. Reaction mechanisms of metals with hydrogen sulfide and thiols in model wine. Part 1: Copper-catalyzed oxidation. J. Agric. Food Chem. 64(20): 4095−4104.

Kreitman, G.Y., Danilewicz, J.C., Jeffery, D.W., Elias, R.J. 2016. Reaction mechanisms of metals with hydrogen sulfide and thiols in model wine. Part 2: Iron- and copper-catalyzed oxidation. J. Agric. Food Chem. 64(20): 4105−4113.

Mira, H., Leite, P., Catarino, S., Ricardo-da-Silva, J.M., Curvelo-Garcia. A.S. 2007. Metal reduction in wine using PVI-PVP co-polymer and its effects on chemical and sensory characters. Vitis 46(3): 138–147.

Rankine, B.C. 1989. Making good wine: a manual of winemaking practice for Australia and New Zealand. Melbourne: Sun Books.

Siebert, T.E., Solomon, M.R., Pollnitz, A.P., Jeffery, D.W. 2010. Selective determination of volatile sulfur compounds in wine by gas chromatography with sulfur chemiluminescence detection. J. Agric. Food Chem. 58 (17) 9454–9462.

Ugliano, M., Kwiatkowski, M., Vidal, S., Capone, D., Siebert, T., Dieval, J.B., Aagaard, O., Waters, E.J. 2011. Evolution of 3-mercaptohexanol, hydrogen sulfide, and methyl mercaptan during bottle storage of Sauvignon blanc wines. Effect of glutathione, copper, oxygen exposure, and closure-derived oxygen. J. Agric. Food Chem. 59: 2564−2572.

Viviers, M.Z., Smith, M., Wilkes, E., Smith, P.A. 2013. Effects of five metals on the evolution of hydrogen sulfide, methanethiol and dimethyl sulfide during anaerobic storage of Chardonnay and Shiraz wine. J. Agric. Food Chem. 61: 12385–12396.