The goal of this study was to set up a method for the determination of reductive sulphur compounds in order to offer an additional sensitive and reliable tool for investigations. This will allow not only the usual positive aromas to be evaluated (esters, higher alcohols, terpenes and thiols), but also some of the negative ones in order to obtain a more complex and complete picture of the phenomena occurring from the winemaking stages up to the consumption of the wine.
The possible sources of reductive sulphur compounds (RSC) are numerous and varied and can occur at several stages during the winemaking process or storage of bottled wine.1 The relationships between the factors that affect the formation of RSC are complex and worth investigating. From an analytical point of view, a limited number of methods is equivalent to limited knowledge of the subject matter.
RSC in wine are divided into ‘light’ (boiling point, b.p. <90°C) and ‘heavy’ (b.p. >90°C) compounds, indicating the difficulty of using a relevant common sampling and enrichment technique.2,3 The additional difficulties to overcome are related to the required sensitivity of the analytical technique and to the instability and differences in volatility of RSC. Several analytical approaches have been employed to quantify sulphur (S) volatiles in wine. Based on the state-of-the-art instrumentation available to us through the collaboration with the Central Analytical Facility of Stellenbosch University, we optimised and implemented a method based on headspace sampling coupled with solid-phase microextraction (HS-SPME), combined with gas chromatography (GC) followed by tandem mass spectrometry (MS/MS), a more performant variant of the single MS. This setup offers the increased selectivity and sensitivity required for the RSC. This was coupled with a chemically sophisticated, but low-tech type of device for the direct measurement of H2S.
S and S-containing compounds react selectively with a variety of reagents. This chemical property can be used for the determination of S-containing compounds. For example, the affinity of S for Mercury (Hg) was used in the original method for the analysis of thiols in wines. Some of the reactions produce coloured compounds and these can be used to monitor and measure S compounds. Some reagents are selective to only one compound, like H2S.
A commercial device that can be used for this purpose is available for a couple of years.3 Detector tubes are thin glass tubes with calibration scales printed on them to directly read concentrations of the substances to be measured. Each tube contains detecting reagents that are sensitive to the target substance and quickly produce a distinct layer colour change. The amount of reagent in a tube is calibrated in such a way that there are different ranges for H2S, depending on the application. The reaction is irreversible and progressive, starting from level 0 and continuing until all the H2S present in the headspace has reacted. The tubes can be used until the reaction level reaches the top of the tube.
Tubes from two different manufacturers were tested, to determine their suitability for use in our type of application. The colour changes and the reaction to water vapours were different for the two types of tubes, as can be seen in Figure 1.
FIGURE 1. The types of tubes chosen and their colour change after exposure to H2S. Left: Colour change white to dark brown/black. Middle: Colour change white to pink, yellow colouration due to exposure to water vapours. Right: Different types of tubes after exposure to water vapours.
Things to be taken into account when using the tubes are: Temperature, atmospheric pressure (depending on how the measurements are done), relative humidity, the volume of sample tested and the volume of gas sampled. The tubes chosen were tested for repeatability and robustness in the lab working conditions.
The sampling can be done with a gas sampling pump, which is commercially available. After considering various possibilities, including using the pump, we have decided on an alternative setup, based on the use of an inert gas to displace the H2S present in the solution. This setup was chosen because it avoids issues related to the atmospheric pressure adjustment needed when using the gas sampling pump, and only the volume and the temperature of the liquid have to be controlled.
Measuring RSC by GC-MS/MS
The list of compounds measured is comprehensive. It includes 17 RSC from various classes and with a variety of volatilities: Alkylthiols (methanethiol and ethanethiol), sulphides (dimethyl sulphide and diethyl sulphide), disulphides (carbon disulphide, dimethyl disulphide and diethyl disulphide), thioacetates (methylthioacetate and S-ethylthioacetate), thiazoles (4-methyl thiazole and benzothiazole), trisulphides (dimethyl trisulphide), thiolaldehyde (methional) and thioalcohols (2-mercaptoethanol, 2-methylthio-1-ethanol, 3-methylthio-1-propanol/methionol and 4-methyltho-1-butanol). Since there are many classes of S compounds included in the method, each class uses its own deuterated internal standard for normalisation of measurement. These internal standards are commercially available, therefore not a limiting factor. After testing its performance parameters, the method was considered suitable for the analysis of RSC in both red and white wines.
Given the various volatility properties of these compounds, not all of them will contribute to wine aroma in normal circumstances. Nevertheless, it is important to be able to measure them, since they can be involved in many processes occurring in wine and might be key to elucidating metabolic pathways, for example.
Take home message
The main issues with the determination of the RSC derive from the compounds’ high volatility and affinity for various chemicals commonly present in the work environment. Using a combination of low-tech and high-tech equipment, a large variety of RSC can be measured. The simple, straightforward determination of H2S can be done with the use of the commercially available tubes, taking into account a series of experimental conditions to ensure accuracy and repeatability. Volatility and contamination issues can be encountered in the measurement of RSC by GC-MS/MS if adequate care is not taken. Nevertheless, the developed method has shown to be linear, sensitive and reproducible in the conditions tested.
The measurement of H2S using the tubes is easy to implement and does not require any high-tech setup. It can easily be transferred to the industry for the producers to measure H2S in situ if necessary, while taking into account the restrictions on the measuring conditions to ensure accuracy. The sensitivity can be adapted based on the sample and on the type of tube chosen.
At the other end of the technology spectrum, the RSC method requires state-of-the-art equipment. The method is already added to the portfolio of analyses to be performed on commercial and experimental wines. The method is set up at the Chemical Analytical Lab of the Department of Viticulture and Oenology (DVO), in collaboration with the Central Analytical Facility.
By combining the two methods, a wide range of compounds can be measured and thus increase our knowledge regarding the RSC in wine. These methods also open the possibility for new projects and experiments that evaluate the impact that various winemaking techniques and products have on the final composition of wines. They can be applied further to a variety of projects that cover fundamental knowledge, such as the elucidation of pathways related to microorganisms and their metabolites in wine, which will translate into applied knowledge.
Organic sulphur (S) compounds play a considerable role in the sensory characteristics of a wine, as they are frequently present among the character impact compounds.
The presence of reductive sulphur compounds (RSC) in wines can affect quality and typical off-flavours resemble onions, garlic, cooked cabbage, rubber and putrefaction. The difficulties to overcome when measuring these compounds are related to the required sensitivity of the analytical technique and to the instability, reactivity and differences in volatility of reductive S compounds. The need for reliable data can be addressed by developing new analysis methods that make use of available technology. Some methods are easy to implement because they make use of low-tech equipment, while others are highly specialised and require state-of-the-art instrumentation.
In this study the determination of RSC was divided into two measurement types:
- Measurement of H2S using a colorimetric reaction and
- determination of other RSC by GC-MS/MS.
The two methods were proven to give reproducible results and were implemented at the Department of Viticulture and Oenology, in collaboration with the Central Analytical Facility.
- Kinzurik, M.I., Herbst-Johnstone, M., Gardner, R.C. & Fedrizzi, B., 2015. Evolution of volatile sulfur compounds during wine fermentation. Journal of Agricultural and Food Chemistry 63(36): 8017 – 8024.
- Nguyen, D., Nicolau, L. & Kilmartin, P.A., 2007. Application of an automated headspace solid phase micro-extraction for the GC-MS detection and quantification of reductive sulfur compounds in wines. Gas Chromatography in Plant Science, Wine Technology, Toxicology and Some Specific Applications.
- Butzke, C.E. & Park, S.K., 2011. Impact of fermentation rate changes on potential hydrogen sulphide concentrations in wine. Journal of Microbiology and Biotechnology 21(5): 519 – 524.
– For more information, contact Astrid Buica at firstname.lastname@example.org.