A Technical Guide
for Wine Producers

RECENT ARTICLES   |   WYNBOER HOME

Extractable or not? - that is the question

M. H. Gey van Pittius, P van Rensburg & I. S. Pretorius
Institute for Wine Biotechnology, Department of Viticulture & Oenology, University of Stellenbosch, Stellenbosch

Introduction

Analytical analysis is the primary method of obtaining the most important "bouquet" facts about the chemical properties of an alcoholic beverage. If one knows the types and amounts of chemical compounds present in a juice, wine or brandy, it is possible to characterise and compare them and even to follow all the winemaking processes from the grapes right through to the final product.

Problem

Wine, grape juice and brandy are very complex matrixes. Chemical compounds present within these matrixes cross-interfere with each other, which may lead to difficulty in quantifying single compounds. The compounds of interest have to be extracted from the matrixes before any chemical analyses for quantification can be performed. Before deciding on the extraction method to be used, one needs to know three things: (i) into which chemical group do these compounds of interest fall; (ii) what is the nature of the matrixes from which these compounds have to be extracted; and (iii) whether the aim of the study is to perform a qualitative or a quantitative analysis.

Status quo

The most common procedure for studying the volatile components of largely aqueous products, such as grape juices and wines, firstly involves the extraction of the sample by means of an organic solvent to obtain a concentrated essence, and secondly, it involves the removal of most of this solvent by means of concentration.

The extraction of volatile substances with diethyl ether, without concentrating the extracted phase, is a method that has been used in the South African wine industry for the past 30+ years. Both research bodies and a well-known company have published numerous studies using this method. Traditionally, the general method for wine extraction has been described by using a low boiling-point solvent. The extract is then dried, concentrated and finally injected into the chromatograph by means of the split injection technique (Grob and Mller, 1987).

Objections

Although the method described above has been widely accepted for use by the wine research community in South Africa, there are several very valid objections to this approach.

  1. The extraction times used in this method are very long and the solvents must be purified scrupulously (Ferreira et al., 1993). If the solvent is not clean it could create a background of impurities. Diethyl ether is not distilled by this method, and this leads to the injection of a very dirty sample.

  2. In general, the extractants can only be injected into the split mode, getting rid of the excess solvent. This means that to reach the sensitivity required, the substances that are being quantified often have to be concentrated thousands of times. These multiple concentration processes result in a significant loss of solutes due to co-evaporation, as well as a certain loss in accuracy (Grob and Müller, 1987).

  3. Discrimination processes can take place during the concentration step and in the injection of a very dirty sample. In other words, not only will the impurities from the unclean diethyl ether also be concentrated, but only some of the compounds will be concentrated and some of the highly volatile substances will be lost. Thus, not all the compounds claimed to be present will still be there.

  4. Cobb et al. (1978) proved in the late 1970s that the use of diethyl ether as an extractant does not result in a very high recovery of the following compounds: ethyl butanoate, 2-methyl-1-propanol, 3-methyl-1-butanol, 1-hexanol, benzaldehyde, acetophenone, benzyl formate and 2-phenethyl butanoate, in comparison with organic solvents such as Freon 11, 2-methylbutane and dichloromethane. Thus, the method that is used does not extract all the compounds it is claimed to. Small volatile compounds, like methanol and acetic acid, cannot be detected by some GC-columns unless a derivitisation step is incorporated in the method. Without this step, the very small volatile compounds will not undergo partitioning in the column, moving right through it undetected.

Conclusions

We believe that the method currently used by some wine researchers in South Africa to quantify the presence of volatile substances in grape juice products has a lot of limitations and could potentially lead to incorrect results. We base this on the fact that (i) it does not extract all the compounds that are claimed to be extracted by diethyl ether; (ii) some of the compounds are being masked by other compounds present within the matrix, so that it is impossible for them to be quantified; and (iii) diethyl ether is not distilled, making it a very dirty solvent to use as an extractant and causing it to create a background. This method has to be reviewed and modified to take the abovementioned problems into consideration.

Recommendations

The volatile fraction in wine has a wide range of different chemical and physiochemical properties. Thus, the isolation step must rather be non-selective and should provide a strong pre-concentration; in addition, a very high selectivity is required in the further chromatographic separation and quantification steps (Ferreira 1998).

The isolation and preconcentration steps mostly used in extraction methods all over the world include purge and trap, liquid-liquid extraction methods, lipophylic resin extraction, supercritical fluid extraction or demixing/micro-extraction. From the point of view of extract cleanliness, the preferred sample preparation methods would be purge and trap-based methods, but they do not provide the necessary sensitivity for the analysis of the flavourants with medium to high boiling points. The demixing/micro-extraction approach may be very appropriate for the GC/GC-MS analysis, since:

  1. it avoids using the imprecise solvent evaporation steps

  2. extraction is performed at laboratory temperature

  3. ultra-high purity solvents are used, and

  4. the final extract will be clean enough if the extraction conditions are chosen correctly.

Acknowledgements

We would like to thank B.V. Burger for sharing unpublished results.

References

1) Grob, K. and Müller, E., 1987. Sample reconcentration by column-external solvent evaporation for injection of large volumes into gas chromatographic capillary columns? Journal of Chromatography. 404, 297-315.

2) Ferreira, V.; Sharman, M.; Cacho, J.F.; Dennis, J., 1996. New and efficient microextraction/solid-phase extraction methods for the gas chromatography analysis of wine volatiles. Journal of Chromatography A. 731, 247-259.

3) Ferreira, V.; Rapp, A.; Cacho, J.F.; Hastrich, H.; Yavas, I., 1993. Fast and Quantitative Determination of wine flavor compounds using Microextraction with Freon 113. Journal of Agricultural Food Chemistry. 41, 1413-1420.

4) Ferreira, V.; Lòpez, R.; Escudero A.; Cacho, J.F., 1998. Quantitative determination of trace and ultratrace flavour active compounds in red wines through gas chromatography-ion trap mass spectrometric analysis of microextracts. Journal of Chromatography A. 806, 349-354.

Wynboer is incorporated in WineLand, magazine of the SA wine producers.

Subscribe to WineLand

Visit our sister sites:


South African wine farmers' representative organisation


Facts, figures, contact details and much more in the 2009/10 Directory

UP COPYRIGHT (C) 2000 WineLand