SEARCH

SEARCH BY CITATION

Keywords:

  • amelioration;
  • guaiacol;
  • reverse osmosis;
  • smoke taint;
  • wine

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Background and Aims:  Wines made from grapes harvested from vineyards exposed to bushfire smoke can exhibit objectionable ‘smoky’, ‘cold ash’, ‘medicinal’ and ‘ashy’ aroma and flavour characters. This study evaluated a combined reverse osmosis and solid phase adsorption process as a potential amelioration method for the treatment of smoke-tainted wines.

Methods and Results:  Smoke-tainted wines were treated using either pilot or commercial scale reverse osmosis systems and the chemical composition and sensory properties of wine compared before and after treatment. The concentrations of smoke-derived volatile phenols, including marker compounds, guaiacol and 4-methylguaiacol, decreased significantly with treatment. As a consequence, diminished smoke-related sensory attributes enabled treated wines to be readily differentiated from untreated wines. However, the taint was found to slowly return with time, likely because of hydrolysis of glycoconjugate precursors, which were not removed during the treatment process.

Conclusions:  Reverse osmosis and solid phase adsorption reduced the concentration of smoked-derived volatile phenols and improved the sensory attributes of smoke-tainted wines.

Significance of the Study:  This is the first study to evaluate the amelioration of smoke taint in wine using reverse osmosis and solid phase adsorption.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Aroma is an important aspect of wine quality and has therefore been the subject of considerable research. Indeed, several hundred volatile compounds have been identified in wine to date. These compounds contribute to the complexity and varietal character of wine and can originate from grapes, the action of yeast during fermentation, oak wood during barrel maturation or in the bottle with aging (Williams et al. 1981, Günata et al. 1985, Winterhalter et al. 1990, Pollnitz et al. 2004). However, not all volatile compounds make a desirable contribution to wine aroma. Some volatiles are indicative of winemaking faults; e.g. the ‘bruised apple’ and ‘vinegar’ characters associated with excessive concentrations of acetaldehyde and acetic acid due to oxidation and lactic bacteria spoilage (volatile acidity), respectively (Ribéreau-Gayon et al. 2000). In other cases, contamination by exogenous volatiles can lead to taints in wine; 2,4,6-trichloroanisole, for example, is considered to contribute to the ‘musty’ attribute associated with cork taint (Buser et al. 1982).

In recent years, the potential for smoke to taint grapes and wine has been a concern for some winemakers, following the occurrence of significant bushfires in the vicinity of wine grapegrowing regions. Kennison et al. (2007) demonstrated the presence of several smoke-derived volatile phenols, including guaiacol and 4-methylguaiacol, in wines made from smoke-affected grapes. These wines were found to exhibit objectionable ‘smoky’, ‘cold ash’, ‘medicinal’ and ‘ashy’ aroma and flavour characters (Kennison et al. 2007), with the intensity of smoke-related sensory attributes dependent on the timing and duration of grapevine smoke exposure (Kennison et al. 2009). Vineyard exposure to smoke cannot be readily predicted or prevented, but can have a significant financial impact on grape and wine production. As such, methods which reduce the concentration of smoke-derived volatile compounds in wine, thereby mitigating the effects of smoke exposure, would be of benefit to grape growers and winemakers. Ristic et al. (2011) investigated the effect of different winemaking techniques on the extent of smoke taint in wine and found the duration of skin contact, choice of yeast strain and addition of oak chips or tannins influenced smoke-related sensory properties. These techniques can be applied by winemakers when processing smoke-affected grapes, but do not address the issue of smoke taint in wine.

Reverse osmosis is a filtration process involving diffusion across a semi-permeable membrane against a concentration gradient (Paulsen et al. 1985), in which separation efficiency relies on both size exclusion and solution-diffusion mechanisms (Cuperus and Nijhuis 1993). Reverse osmosis is routinely used for water purification and desalination (Madaeni 1999), with an increasing number of applications being reported within food and beverage industries, e.g. the preparation of milk (Glover 1971) and fruit juice concentrates (Paulsen et al. 1985, Kane et al. 1995). Within the wine industry, reverse osmosis has been used to manipulate wine alcohol content, volatile acidity and acidification through concentration of grape must (Duitschaever et al. 1991) and wine (Bui et al. 1988, Massot et al. 2008). Reverse osmosis has also been coupled with solid phase adsorption to remove 4-ethylguaiacol and 4-ethylphenol from Brettanomyces-affected wine (Ugarte et al. 2005). Ugarte et al. (2005) demonstrated 4-ethylguaiacol and 4-ethylphenol could permeate the reverse osmosis membrane and be adsorbed by Amberlite XAD-16 HP resin, such that the ‘animal’, ‘medicinal’ and ‘stable’ aromas associated with Brettanomyces spoilage (Chatonnet et al. 1992) became less apparent with treatment. Given volatile phenols are implicated in both Brettanomyces spoilage and smoke taint, there is clearly potential for this treatment process to be used for the removal of smoke-derived volatile phenols from wine. This study was therefore undertaken to evaluate the capacity of a combined reverse osmosis and solid phase adsorption process to ameliorate smoke taint in wine.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Smoke-affected wines

Three Pinot Noir wines were used in this study. Two commercial wines, hereafter referred to as Pinot Noir 1 and Pinot Noir 3, were sourced from wineries located in the Gippsland (37°45′S, 147°43′E) and Yarra Valley (37°37′S, 145°24′E) wine regions, respectively. These wines were made from grapes exposed to smoke following a series of bushfires that occurred in North Eastern Victoria between 1 December 2006 and 7 February 2007; however, specific details regarding the timing, duration and density of grapevine smoke exposure are not known. Pinot Noir 1 was aged in French oak for 19 months, whereas Pinot Noir 3 was un-oaked. A third wine, hereafter referred to as Pinot Noir 2, was made from smoke-affected grapes sourced from field trials involving the application of smoke to vines, using experimental conditions described previously (Kennison et al. 2008, Dungey et al. 2011). Pinot Noir vines growing in a vineyard located in the Adelaide Hills wine region (34°30′S, 138°59′E) were exposed to straw-derived smoke at approximately 7 days post-veraison, i.e. at a juice total soluble solids (TSS) concentration of approximately 13°Brix, as determined using a digital handheld refractometer (PAL-1, Atago, Tokyo, Japan). The duration of smoke exposure was 60 min. Smoke-affected grapes (120 kg) were harvested when juice TSS reached 16 ± 1°Brix (i.e. an early harvest, due to fruit from this vineyard being allocated to a commercial sparkling wine base) and processed according to small-lot winemaking procedures (Holt et al. 2006). Bunches were de-stemmed and crushed, and tartaric acid added to adjust the pH to 3.5, prior to inoculation with Maurivin PDM yeast (200 ppm). Musts were fermented on skins for 7 days at 15°C, with the cap plunged at least twice per day. The wine was pressed at a TSS level of 1°Baume and transferred to stainless steel vessels and held at 20°C until the residual sugar approached 0 g/L. Wines were then racked from gross lees and cold stabilised (at 4°C for 3 weeks). Wine pH and free SO2 were adjusted to 3.4 and between 25 to 30 ppm, respectively, before filtration and bottling (under screw cap closures).

Pilot scale reverse osmosis and solid phase adsorption treatment of wine

Reverse osmosis and solid phase adsorption treatment of Pinot Noir 1 and Pinot Noir 2 was performed (in triplicate) using a pilot scale Micro AA unit (Memstar Pty. Ltd, Oakleigh, Victoria, Australia), in accordance with the manufacturer's operating instructions. The unit was equipped with a proprietary spiral wound nanofiltration membrane (‘1812’ full fit (‘sanitary’) form factor; nominal molecular weight cut-off of 150–200 amu; filtering area: 0.3 m2), a polystyrene based adsorbent resin (MemTaint, Memstar Pty. Ltd, Oakleigh, Victoria, Australia; 200 mL) in a packed column and a heat exchanger (to ensure wine temperature remained below 20°C). Wine was pumped from the feed tank (6 L) across the membrane under 1700 kPa pressure to generate permeate and retentate fractions at flow rates of approximately 50 mL/min and 600 mL/min, respectively. Retentate flow was circulated back to the feed tank, while permeate flowed through the packed column and then back to the feed tank (Figure 1). Samples (50 mL) were collected before treatment (i.e. untreated wine) and then after t = 0.5, 1, 2 and 3 h of treatment, for chemical analyses. On completion of treatment (i.e. at t = 3 h), the resulting wine was bottled and stored at 15°C, until sensory analyses.

image

Figure 1. Schematic diagram of combined reverse osmosis and solid phase adsorption system.

Download figure to PowerPoint

In a separate experiment, permeate and retentate samples (50 mL) were collected during the reverse osmosis treatment of Pinot Noir 1. Wine (6 L) was filtered as above, but with both retentate and permeate flow circulated back to the feed tank; i.e. without solid phase adsorption treatment of the permeate fraction. The system was allowed to circulate for 30 min before samples were collected for quantification of volatiles by gas chromatography-mass spectrometry (GC-MS) and glycoconjugates by high performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS). This experiment was not replicated.

Commercial scale reverse osmosis and solid phase adsorption treatment of wine

Reverse osmosis and solid phase adsorption treatment of Pinot Noir 3 was performed by Memstar Pty. Ltd. using a commercial scale R80940 unit (Memstar Pty. Ltd, Oakleigh, Victoria, Australia). The unit was equipped with nine proprietary spiral wound nanofiltration membranes (‘8040’ full fit (‘sanitary’) form factor; nominal molecular weight cut-off of 150–200 amu; filtering area: 270 m2), a polystyrene based adsorbent resin (MemTaint, Memstar Pty. Ltd, Oakleigh, Victoria, Victoria; 200 L) in a packed column and a heat exchanger (to ensure wine temperature remained below 20°C). Wine (5000 L) was pumped across the membranes under pressure of approximately 3000 kPa to generate permeate and retentate fractions. Retentate flow was circulated back to the feed tank, while permeate was passed through the packed column and then recombined with the retentate prior to being returned to the feed tank (Figure 1), as per pilot scale treatment. Wine was treated for 12 h during which time 16 400 L of permeate was generated (i.e. a permeate flow rate of approximately 1370 L/h). Untreated and treated wine was bottled (48 × 750 mL) and stored at 15°C, until required for chemical and sensory analyses. The commercial scale treatment was not repeated.

Wine analysis

Untreated and treated wines were analysed to determine pH and titratable acidity (TA), as tartaric acid equivalents to an endpoint of pH 8.2; residual sugar (glucose) using an enzymatic test kit (R-Biopharm AG, Darmstadt, Germany); alcohol content using an alcolyzer (Anton Paar, North Ryde, NSW Australia); and wine colour density and phenolics according to the methods of Iland et al. 2004. Analyses were performed within 1 week of treatment.

Quantification of volatiles by GC-MS

Guaiacol, 4-methylguaiacol, 4-ethylguaiacol, 4-ethyphenol, eugenol, vanillin, cis- and trans-oak lactone, and furfural were quantified by GC-MS using stable isotope dilution assay (SIDA) methods reported previously (Pollnitz 2000, Pollnitz et al. 2000, 2004). These publications describe the synthesis of internal standards used herein. Analyses were performed by the Australian Wine Research Institute's Commercial Services Laboratory (Adelaide, Australia) using an Agilent 6890 gas chromatograph (Agilent Technologies, Forest Hill, Vic, Australia), coupled to a 5973 mass selective detector. For Pinot Noir 1, both smoke- and oak-derived volatiles were quantified; whereas only the smoke-derived phenols, guaiacol, 4-methylguaiacol, 4-ethylguaiacol and 4-ethylphenol, were measured for Pinot Noir 2 and Pinot Noir 3. A deuterated internal standards solution comprising known concentrations of d3-guaiacol, d3-4-methylguaiacol, d4-4-ethylphenol, d3-vanillin, d4-cis-oak lactone, d4-trans-oak lactone and d4-furfural in ethanol (100 µL) was added to each sample (10 mL) in a screw cap vial. Organic solvent (n-pentane c. 2 mL) was added and the mixture shaken briefly. A portion of the organic layer (c. 1.5 mL) was then taken for instrumental analysis. GC-MS analyses were performed within 2 weeks of treatment. Analysis of untreated and treated Pinot Noir 3 wines was repeated at 6, 12 and 30 months post-bottling.

Quantification of guaiacol glycoconjugates by HPLC-MS/MS

Guaiacol glycoconjugates were quantified by HPLC-MS/MS using SIDA methods reported previously (Dungey et al. 2011, Wilkinson et al. 2011). Dungey et al. (2011) described the synthesis of the internal standard used herein. Analyses were performed using an Agilent 1200 HPLC system, coupled to a 4000 Q TRAP hybrid tandem mass spectrometer (Agilent Technologies, Forest Hill, Vic, Australia). Guaiacol glycoconjugate concentrations were determined for untreated and treated Pinot Noir 1 wine (n = 3) and for permeate and retentate fractions generated (as above) from Pinot Noir 1 (n = 1). d4-Guaiacol β-D-glucopyranoside (100 µg/mL in water, 10 µL) was added to each sample (1 mL) as an internal standard, the mixture shaken briefly and filtered through a 0.45 µm GHP membrane (Acrodisc®, PALL Life Sciences, Cheltenham, South Australia, Australia) prior to instrumental analysis. HPLC-MS/MS analyses were performed within 2 weeks of treatment.

Sensory analysis of wine

Descriptive sensory analysis of Pinot Noir 1 and Pinot Noir 2 wines.

Descriptive sensory analysis (DA) (Stone and Sidel 2004) was performed on Pinot Noir 1 and Pinot Noir 2, to determine any differences in the intensity of smoke-related attributes between untreated and treated wines. DA consisted of a series of training and formal evaluation sessions. Twelve panellists were recruited among staff and students from the University of Adelaide and the Australian Wine Research Institute (AWRI). Ten panellists had extensive experience in sensory analysis of smoke-tainted wines, while two less experienced panellists underwent additional training sessions prior to the panel training and formal evaluation sessions. Training sessions involved presentation of each wine replicate to the sensory panel to generate appropriate aroma and palate attributes. Six aroma attributes and eight palate attributes (Table 1) were rated during formal sessions, using an unstructured 15-cm line scale with indented anchor points of ‘low’ to ‘high’ placed at 10% and 90%, respectively. Evaluations were carried out using covered, three-digit coded International Organization for Standardization (ISO) tasting glasses. Each wine (30 mL) was presented at 22–24°C, in isolated and well-ventilated tasting booths under red-light illumination, to limit potential for bias based on wine colour. The presentation order was randomised and balanced across judges. Four wines were presented per session, with one session per day, run over three consecutive days. On each day, panellists assessed the different treatment wine replicates, as well as untreated wines. After assessing each sample, panellists rinsed with water and a 1 g/L citrus pectin solution (Sigma). Data was collected using FIZZ software (Version 2.40 E, Biosystemes, Couternon, France).

Table 1.  Sensory attributes used for descriptive analysis of wines before and after reverse osmosis and solid phase adsorption treatment.
AttributeDescription
Aroma 
 Fruit aroma intensityThe intensity of the overall fruit aroma, includes red fruit, red berry, dark berry, capsicum, green, tropical fruit, honey.
 SmokePerception of any type of smoke aroma, includes smoked meat/bacon, toasty, charry, cigar-box, estery.
 Cold ashBurnt aroma associated with ashes, includes ashtray, tarry, campfire.
 EarthyAny aroma associated with musty, dusty, wet-wood, barnyard, mushroom-like, dank, mouldy, stagnant, stale.
 Burnt rubberPerception of burnt rubber-like aromas.
 MedicinalAromatic characteristic of band-aids, disinfectant-like, including cleaning products, solvents, chemicals.
Palate 
 Fruit flavour intensityThe intensity of the overall fruit flavour; includes red fruit, red berry, dark berry, capsicum, green, tropical fruit, honey.
 SmokyPerception of smoke flavour, includes bacon and smoked meat.
 Ashy aftertasteLength of taste associated with residue of ashtray perceived in the mouth after expectorating, includes coal ash, ashtray, tarry, acrid, campfire.
 Woody aftertasteLength of taste associated with woody residue, includes wood, oak, pencil shavings.
 AcidityIntensity of sour/acid taste.
 MetallicThe ‘tinny’ flavour associated with metals.
 BitterIntensity of bitter taste, bitter aftertaste.
 DryingDrying, puckering mouth-feel after expectorating the wine.

Difference testing of wines

Difference tests were conducted for each wine (i.e. before and after treatment), using the duo-trio method described by Meilgaard et al. (2007) and a panel of 48 judges. Panellists were aged between 18 and 55, with similar numbers of males and females. Wines were presented to the panel using a balanced, randomised presentation order, comprising all possible configurations (i.e. RAAB, RABA, RBAB, RBBA, where A denotes untreated wine and B denotes treated wine) an equal number of times. Wines (20 mL) were served at 22–24°C, in covered ISO tasting glasses with randomly assigned three-digit codes. Panellists smelled and tasted the samples, and were asked to identify the sample which was the same as the reference.

Consumer acceptance testing of wines

Consumer acceptance tests were conducted using a hedonic rating method described by Meilgaard et al. (2007) using the same 48-judge panels described above. Untreated and treated wines were presented to the panel using a balanced, randomised presentation order. Wines (20 mL) were served at 22–24°C, in covered ISO tasting glasses with randomly assigned three-digit codes. Panellists were asked to rate their liking for each wine on a nine-point hedonic scale with anchors: ‘don't like the wine at all’ (1); ‘neither like nor dislike the wine’ (5); and ‘like the wine very much’ (9). The ratings for each wine were then measured in cm from the left hand anchor and subjected to statistical analysis.

Data analysis

Sensory descriptive data for each attribute was analysed using statistical software (JMP, Version 7, SAS Institute, Cary, NC, USA) and the analysis of variance (ANOVA), mixed model, testing for the effects of treatment, fermentation replicate, presentation replicate and judge (considered as a random effect). Mean comparisons were performed by least significant difference (LSD) multiple comparison tests at P < 0.05. Chemical data were analysed by one-way ANOVA using GenStat (9th Edition, VSN International Limited, Herts, UK). Mean comparisons were performed by LSD multiple comparison test at P < 0.05.

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Since vineyard exposure to smoke cannot be readily predicted or prevented, the current study intended to evaluate reverse osmosis and solid phase adsorption as a potential method of ameliorating smoke-tainted wine. Pinot Noir 1 and Pinot Noir 2 were chosen as examples of wines affected by bushfire and experimental smoke, respectively, and were treated (in triplicate) using pilot scale equipment. Pinot Noir 3, which was also affected by bushfire smoke, was chosen as an example of a wine treated on a commercial scale, but without replication. The initial composition of each wine was determined by quantitative GC-MS analysis with guaiacol, 4-methylguaiacol, 4-ethylguaiacol and 4-ethylphenol measured as markers of smoke taint (Table 2). Since guaiacol and 4-methylguaiacol are also oak-derived volatile compounds (Pollnitz 2000, Pollnitz et al. 2004), in the case of Pinot Noir 1, these compounds likely originated from grapevine exposure to smoke as well as oak barrel maturation. A number of additional oak-derived volatiles, i.e. eugenol, vanillin, cis- and trans-oak lactone, and furfural, were also measured in Pinot Noir 1, which received 19 months barrel maturation. Guaiacol and 4-methylguaiacol are typically the most abundant volatile phenols present in smoke-tainted wines (Kennison et al. 2007, 2008). The elevated levels of 4-ethylguaiacol and 4-ethylphenol observed in Pinot Noir 1 (Table 2) might therefore indicate some degree of Brettanomyces spoilage. Guaiacol and 4-methylguaiacol were detected in Pinot Noir 2 and Pinot Noir 3, albeit at lower levels than for Pinot Noir 1. Low levels of 4-ethylguaiacol and 4-ethylphenol were also measured in Pinot Noir 3 (Table 2). The capacity of reverse osmosis and solid phase adsorption to ameliorate smoke taint in wine was determined by monitoring the chemical composition and sensory properties of wines before and after treatment, in particular, the concentration of smoke-derived volatile phenols and smoke-related sensory attributes.

Table 2.  Volatile composition (guaiacol, 4-methylguaiacol, 4-ethylguaiacol, 4-ethylphenol, eugenol, vanillin, cis- and trans-oak lactone and furfural) of smoke-affected Pinot Noir wines before and after reverse osmosis and solid phase adsorption treatment, and permeate and retentate fractions derived from reverse osmosis treatment of Pinot Noir 1.
SampleConcentration (µg/L)
Guaiacol4-Methylguaiacol4-Ethylguaiacol4-EthylphenolEugenolVanillincis-oak lactonetrans-oaklactoneFurfural
  1. Values followed by a different letter are significantly different. †Pinot Noir affected by bushfire smoke and treated with pilot RO system (n = 3) and retentate and permeate fractions (n = 1);‡Pinot Noir affected by experimental smoke and treated with pilot RO system (n = 3);§Pinot Noir affected by bushfire smoke and treated with commercial RO system (n = 1). n.d., not detected. tr, trace.

Pinot Noir 1         
 Untreated49a36a294a391a16a237a137a111a154a
 Treatedt = 0.5 h35b28b225b277b12b176b114ab91ab130b
 Treatedt = 1 h27c23c190c214cn.d.148c102bc85ab117c
 Treatedt = 2 h18d15d134d126dn.d.104d79cd64bc109c
 Treatedt = 3 h13e11e94e78en.d.79e61d47c109c
 P<0.001<0.001<0.001<0.001<0.001<0.0010.0010.003<0.001
 Retentate44342773481622615197138
 Permeate3926193323111546847112
Pinot Noir 2         
 Untreated7a2n.d.n.d.
 Treatedt = 0.5 h5b2n.d.n.d.
 Treatedt = 1 h3c1n.d.n.d.
 Treatedt = 2 h2d1n.d.n.d.
 Treatedt = 3 h1en.d.n.d.n.d.
 P<0.001
Pinot Noir 3§         
 Untreated0 months post-bottling12514
 Untreated6 months post-bottling123n.d.n.d.
 Untreated12 months post-bottling154n.d.n.d.
 Untreated30 months post-bottling164n.d.n.d.
 Treated0 months post-bottling3tr.28
 Treated6 months post-bottling52n.d.n.d.
 Treated12 months post-bottling63n.d.n.d.
 Treated30 months post-bottling94n.d.10

Effect of reverse osmosis and solid phase adsorption treatment on wine composition

The volatile phenol content of each wine decreased significantly with treatment (Table 2). Analysis of Pinot Noir 1 and Pinot Noir 2 throughout the treatment process demonstrated the progressive loss of phenols. After 3 h of treatment, more than 67% of volatile phenols initially present in Pinot Noir 1 had been removed and guaiacol and 4-methylguaiacol were almost completely removed from Pinot Noir 2. A progressive loss of oak-derived volatiles was also observed during Pinot Noir 1 treatment. Eugenol, vanillin, cis- and trans-oak lactone and furfural concentrations were significantly reduced after treatment (Table 2), with potential implications for the contribution of oak-related sensory attributes to overall wine aroma. Some loss of desirable wine aroma was to be expected, given the membrane's molecular weight cut-off of 150–200 amu would allow permeation of a range of volatile wine constituents besides the volatile phenols. Indeed, GC-MS analysis of retentate and permeate fractions derived from reverse osmosis treatment of Pinot Noir 1 demonstrated passage of each volatile compound across the membrane (Table 2). The pilot scale system could be used to optimise the treatment process, to achieve the required reduction in smoke-derived volatile compounds with minimal loss of desirable wine components, prior to upscaling.

Commercial scale treatment of Pinot Noir 3 yielded similar results. Guaiacol and 4-methylguaiacol concentrations were reduced from 12 to 3 µg/L and from 5 to less than 1 µg/L (i.e. trace), respectively (Table 2). It is unclear why 4-ethylguaiacol and 4-ethylphenol increased slightly with treatment; however, these concentrations are well below reported detection threshold values of 70 and 1100 µg/L, respectively (Boidron et al. 1988), so these compounds are unlikely to significantly impact the aroma attributes of either untreated or treated Pinot Noir 3. Determination of volatile phenols in treated Pinot Noir 3 was repeated 6, 12 and 30 months post-bottling to investigate the cellaring potential of treated wines. A gradual increase in guaiacol and 4-methylguaiacol concentrations was observed for both untreated and treated wines (Table 2), suggesting smoke taint might be enhanced with bottle age. That said, it should be noted that the volatile phenol content of treated wine 30 months post-bottling was lower than that of untreated wine at the time of bottling. The evolution of guaiacol and 4-methylguaiacol with time is likely attributable to the hydrolysis of glycoconjugate precursor forms of these compounds present in the wine, and is consistent with earlier studies (Kennison et al. 2008).

Hayasaka et al. (2010) employed stable isotope tracer experiments to identify glucoside and disaccharide conjugates of guaiacol in grapevine leaves and berries. The glycoconjugation of guaiacol following grapevine exposure to smoke has subsequently been demonstrated (Dungey et al. 2011) and hydrolysis of these glycoconjugates can explain the increased phenol concentrations reported by Kennison et al. (2008) during fermentation of smoke-affected grapes and ageing of subsequent wines. In the current study, guaiacol glycoconjugate concentrations of 479 ± 11 µg/L and 550 ± 49 µg/L, respectively, were measured in untreated and treated Pinot Noir 1 wine. These concentrations were not statistically different, indicating glycoconjugate content was not affected by the treatment process. Analysis of retentate and permeate fractions of Pinot Noir 1 confirmed exclusion of the guaiacol glycoconjugate pool by the membrane. While a glycoconjugate concentration of 645 µg/L was measured in the retentate fraction, glycoconjugates were not detected in the permeate. This is not surprising, given glycoconjugate molecular weights range from 286 (glucoside) to 448 (glucose-glucose disaccharide) amu and the nominal membrane cut-off was 150–200 amu. These findings support the hypothesis suggested above, that the return of smoke taint in treated Pinot Noir 3 with time occurred as a consequence of the slow hydrolysis of glycoconjugates present in the bottle. This highlights a limitation of the treatment process; the removal of smoke-derived volatile compounds only and not their conjugate forms, which accordingly limits the cellaring potential of treated wines. For untreated wines, the intensity of smoke-related attributes would be further amplified with bottle age.

To ensure the treatment process did not significantly affect desirable wine attributes, a range of compositional measurements, i.e. pH, TA, sugar and alcohol contents, wine colour density and total phenolics, were made before and after treatment (Table 3). For pilot scale treatment, there was no significant effect on either pH or TA, but a small reduction in TA was observed during commercial scale treatment of Pinot Noir 3. It is unclear if this resulted from adsorption of organic acids by either the membrane or the adsorbent resin. Irrespective, the loss was minor and could be easily adjusted through acid additions, if deemed necessary. Neither residual sugar (i.e. glucose) nor alcohol content was affected by treatment and the process had no significant influence on total phenolics. However, wine colour density differed significantly between untreated and treated wines (Table 3). The molecular weights of red wine pigments (anthocyanins) far exceed the molecular weight cut-off of the reverse osmosis membrane used in the current study. As such, wine colour was preserved in the retentate, whereas permeate fractions were completely colourless (data not shown). While differences in colour density were statistically significant, wine hue was unaffected, and no meaningful difference in wine colour was apparent (Table 3).

Table 3.  Wine composition (pH, TA, glucose, alcohol, total phenolics, colour density and hue) of smoke-affected Pinot Noir wines before and after reverse osmosis and solid phase adsorption treatment (n = 3).
SamplepHTA (g/L)Glucose (g/L)Alcohol (%, v/v)Total phenolics (au)Colour density (au)Colour hue
  1. Values followed by a different letter are significantly different. †Pinot Noir affected by bushfire smoke and treated with pilot RO system (n = 3);‡Pinot Noir affected by experimental smoke and treated with pilot RO system (n = 3);§Pinot Noir affected by bushfire smoke and treated with commercial RO system (n = 1). n.s., not significant; TA, titratable acidity.

Pinot Noir 1       
 Untreated3.315.6<1.014.369.247.68a0.83
 Treatedt = 0.5 h3.345.5<1.014.470.517.82ab0.84
 Treatedt = 1 h3.265.6<1.014.471.987.98b0.84
 Treatedt = 2 h3.225.6<1.014.473.047.97b0.84
 Treatedt = 3 h3.215.5<1.014.475.398.09b0.84
 Pn.s.n.s.n.s.n.s.0.051n.s.
Pinot Noir 2       
 Untreated3.205.8<1.08.718.682.20a1.04
 Treatedt = 0.5 h3.245.6<1.08.516.821.91b1.05
 Treatedt = 1 h3.185.7<1.08.417.331.88b1.04
 Treatedt = 2 h3.175.8<1.08.517.141.79b1.02
 Treatedt = 3 h3.185.9<1.08.418.131.77b1.01
 Pn.s.n.s.n.s.n.s.<0.001n.s.
Pinot Noir 3§       
 Untreated3.389.0<1.014.139.135.190.80
 Treated3.387.8<1.013.734.795.000.77

Effect of reverse osmosis and solid phase adsorption treatment on wine sensory properties

Sensory analyses, comprising descriptive analysis, difference tests and acceptability ratings were undertaken to determine the impact of treatment on wine sensory properties. DA indicated significant differences were perceived between the aroma profiles of untreated and treated Pinot Noir 1 and Pinot Noir 2 wines (Figure 2). Pinot Noir 1 wines were characterised by intense ‘fruit’ aroma and flavour, ‘woody aftertaste’ and ‘drying’ finish. ‘Fruit’ aroma and ‘woody aftertaste’ diminished slightly with treatment (P < 0.05), but importantly, the intensity of ‘smoke’ aroma, ‘smoky’ flavour and ‘ashy aftertaste’ were significantly reduced (P < 0.0001), which is likely attributable to the significant reduction in smoke-derived volatile phenols (Table 2). Untreated Pinot Noir 2 exhibited considerably less intense aroma and flavour attributes compared with Pinot Noir 1 (Figure 2). The reduced ‘fruit’ intensity and higher ‘acidity’ rating were because of the early harvest date of this fruit (i.e. at TSS of 16 ± 1°Brix), whereas the relatively short duration of grapevine exposure to smoke, i.e. only 60 min, resulted in lower intensity smoke-related sensory characters. However, following treatment, Pinot Noir 2 wine displayed significantly reduced ‘smoke’ and ‘cold ash’ aromas, ‘smoky’ flavour (P < 0.05) and ‘ashy aftertaste’ (P < 0.01; Figure 2), again corresponding with the removal of smoke-derived volatile phenols (Table 2).

image

Figure 2. Mean ratings for sensory attributes in smoke-affected Pinot Noir wines before and after reverse osmosis and solid phase adsorption treatment. Values are mean scores from one (untreated) or three (treated, t = 3 h) fermentation replicate wines that were presented to 12 judges in three replicate sensory sessions. *, **, *** indicate significance between untreated and treated wines at P < 0.05, P < 0.01 and P < 0.001, respectively.

Download figure to PowerPoint

Duo-trio tests confirmed a perceptible difference in wine before and after treatment, attributable to the modification of chemical composition (Table 3) and sensory properties (Figure 2). Untreated and treated Pinot Noir wines were readily differentiated (at the 99% confidence level), yielding 35, 33 and 33 (out of 48) correct responses for Pinot Noir 1, Pinot Noir 2 and Pinot Noir 3 wines, respectively. While there was no significant difference in acceptability ratings for the Pinot Noir 1 and Pinot Noir 2 wines, a significant difference was observed for Pinot Noir 3 wines, with treated wine (4.65) rated somewhat more favourably than untreated wine (3.74; Table 4). Interestingly, the acceptability ratings for Pinot Noir 1 and Pinot Noir 3 wines ranged from 1.0 to 9.0, and from 1.0 to 7.5 for Pinot Noir 2 wines (Table 4), indicating that the sensory panel comprised judges who clearly perceived the impact and acceptability of ‘smoke’ characters quite differently. In the case of Pinot Noir 1, smoke-related sensory attributes might have been masked by oak complexity arising from extraction of oak-derived volatiles during barrel maturation, as proposed by Ristic et al. (2011). The narrower range of acceptability ratings observed for Pinot Noir 2 probably reflected the aforementioned lack of fruit ripeness and intensity.

Table 4.  Acceptability ratings for smoke-affected Pinot Noir wines before and after reverse osmosis and solid phase adsorption treatment.
SampleAcceptability rating
MeanRange
  1. Values followed by a different letter are significantly different (P < 0.05). †Pinot Noir affected by bushfire smoke and treated with pilot RO system (n = 48);‡Pinot Noir affected by experimental smoke and treated with pilot RO system (n = 48);§Pinot Noir affected by bushfire smoke and treated with commercial RO system (n = 48). n.s., not significant.

Pinot Noir 1  
 Untreated3.72 ± 0.32 n.s.1.0–9.0
 Treated (t = 3 h)4.18 ± 0.28 n.s.1.0–9.0
Pinot Noir 2  
 Untreated3.99 ± 0.27 n.s.1.0–7.4
 Treated (t = 3 h)4.10 ± 0.28 n.s.0.6–7.5
Pinot Noir 3§  
 Untreated3.74 ± 0.34a1.0–9.0
 Treated4.65 ± 0.37b1.0–9.0

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Reverse osmosis and solid phase adsorption proved a capable and effective process for reducing the concentration of smoke-derived volatile phenols and perception of sensory attributes of smoke-tainted wines, such that the acceptability of wine treated on a commercial scale was significantly improved. Compared with the direct addition of adsorptive resins to wine, the reverse osmosis fraction of wine involved in the process employed in this study should improve the selectivity of compounds targeted for removal, with less impact on desirable wine attributes. Some limitations were identified: the inherent removal of desirable wine volatiles and potential for smoke taint to slowly return with time. Nevertheless, the process offers the wine industry a method for amelioration of smoke taint in wine.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

The authors gratefully acknowledge Memstar Pty. Ltd. for the provision of equipment and technical support; the wineries that provided grapes and wine; Gemma West from AWRI for technical support with winemaking; Randell Taylor from AWRI for GC-MS analysis; Yoji Hayasaka from AWRI and Kerry Pinchbeck from the University of Adelaide (UA) for HPLC-MS/MS analysis; and staff and students from UA and AWRI for participating in sensory trials. This research was supported by the Grape and Wine Research and Development Corporation (GWRDC). Anthea Fudge thanks the GWRDC for the provision of a research scholarship.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References
  • Boidron, J.N., Chatonnet, P. and Pons, M. (1988) Influence du bois sur certaines substances odorantes des vins. Connaissance de la Vigne et du Vin 22, 275294.
  • Bui, K., Dick, R., Moulin, G. and Galzy, P. (1988) Partial concentration of red wine by reverse osmosis. Journal of Food Science 53, 647648.
  • Buser, H.-R., Zanier, C. and Tanner, H. (1982) Identification of 2,4,6-trichloroanisole as a potent compound causing cork taint in wine. Journal of Agricultural and Food Chemistry 30, 359362.
  • Chatonnet, P., Dubourdieu, D., Boidron, J.-N. and Pons, M. (1992) The origin of ethylphenols in wines. Journal of the Science of Food and Agriculture 60, 165178.
  • Cuperus, F.P. and Nijhuis, H.H. (1993) Applications of membrane technology to food processing. Trends in Food Science and Technology 4, 277282.
  • Duitschaever, C.L., Alba, J., Buteau, C. and Allen, B. (1991) Riesling wines made from must concentrated by reverse osmosis. I. Experimental conditions and compositions of musts and wines. American Journal of Enology and Viticulture 42, 1925.
  • Dungey, K.A., Hayasaka, Y. and Wilkinson, K.L. (2011) Quantitative analysis of glycoconjugate precursors of guaiacol in smoke-affected grapes using liquid chromatography–tandem mass spectrometry based stable isotope dilution analysis. Food Chemistry 126, 801806.
  • Glover, F.A. (1971) Concentration of milk by ultrafiltration and reverse osmosis. Journal of Dairy Research 38, 373379.
  • Günata, Y.Z., Bayonove, C., Baumes, R. and Cordinnier, R. (1985) The aroma of grapes I. Extraction and determination of free and glycosidically bound fractions of some grape aroma components. Journal of Chromatography A 331, 8390.
  • Hayasaka, Y., Baldock, G.A., Pardon, K.H., Jeffrey, D.W. and Herderich, M.J. (2010) Investigation into the formation of guaiacol conjugates in berries and leaves of grapevine Vitis vinifera L. cv. Cabernet Sauvignon using stable isotope tracers combined with HPLC-MS and MS/MS analysis. Journal of Agricultural and Food Chemistry 58, 20762081.
  • Holt, H.E., Iland, P.G. and Ristic, R. (2006) A method for mini-lot fermentation for use in research and commercial viticultural and winemaking trials. The Australian and New Zealand Grapegrower and Winemaker 509a, 7481.
  • Iland, P.G., Bruer, N., Edwards, G., Weeks, S. and Wilkes, E. (2004) Chemical analysis of grapes and wine: techniques and concepts (Patrick Iland Wine Promotions: Adelaide).
  • Kane, L., Braddock, R.J., Sims, C.A. and Matthews, R.F. (1995) Lemon juice aroma concentration by reverse osmosis. Journal of Food Science 60, 190194.
  • Kennison, K.R., Wilkinson, K.L., Williams, H.G., Smith, J.H. and Gibberd, M.R. (2007) Smoke-derived taint in wine: effect of postharvest smoke exposure of grapes on the chemical composition and sensory characteristics of wine. Journal of Agricultural and Food Chemistry 55, 1089710901.
  • Kennison, K.R., Gibberd, M.R., Pollnitz, A.P. and Wilkinson, K.L. (2008) Smoke-derived taint in wine: the release of smoke-derived volatile phenols during fermentation of Merlot juice following grapevine exposure to smoke. Journal of Agricultural and Food Chemistry 56, 73797383.
  • Kennison, K.R., Wilkinson, K.L., Pollnitz, A.P., Williams, H.G. and Gibberd, M.R. (2009) Effect of timing and duration of grapevine exposure to smoke on the composition and sensory properties of wine. Australian Journal of Grape and Wine Research 15, 228237.
  • Madaeni, S.S. (1999) The application of membrane technology for water disinfection. Water Research 33, 301308.
  • Massot, A., Mietton-Peuchot, M., Peuchot, C. and Milisic, V. (2008) Nanofiltration and reverse osmosis in winemaking. Desalination 231, 283289.
  • Meilgaard, M., Civille, G.V. and Carr, B.T. (2007) Sensory evaluation techniques (CRC Press: Boca Raton).
  • Paulsen, D.J., Wilson, R.L. and Spatz, D.D. (1985) Osmosis and ultrafiltration applied to the processing of fruit juices. In: Reverse osmosis and ultrafiltration. Eds. S.Sourirajan and T.Matsura (ACS Symposium Series 281, American Chemical Society: Washington) pp. 325344.
  • Pollnitz, A.P. (2000) The analysis of volatile wine components derived from oak products during winemaking and storage. PhD Thesis, The University of Adelaide, Australia.
  • Pollnitz, A.P., Pardon, K.H. and Sefton, M.A. (2000) Quantitative analysis of 4-ethylphenol and 4-ethylguaiacol in red wine. Journal of Chromatography A 874, 101109.
  • Pollnitz, A.P., Pardon, K.H., Sykes, M. and Sefton, M.A. (2004) The effects of sample preparation and gas chromatograph injection techniques on the accuracy of measuring guaiacol, 4-methylguaiacol and other volatile oak compounds in oak extracts by stable isotope dilution analyses. Journal of Agricultural and Food Chemistry 52, 32443252.
  • Ribéreau-Gayon, P., Glories, Y., Maujean, A. and Dubourdieu, D. (2000) Chemical nature, origins and consequences of the main organoleptic defects. In: The handbook of enology: the chemistry of wine stabilization and treatments. Ed. P.Ribéreau-Gayon, Y.Glories, A.Maujean and D.Dubourdieu (John Wiley and Sons Ltd: Chichester) pp. 209253.
  • Ristic, R., Osidacz, P., Pinchbeck, K.A., Hayasaka, Y., Fudge, A.L. and Wilkinson, K.L. (2011) The effect of winemaking techniques on the intensity of smoke taint in wine. Australian Journal of Grape and Wine Research 17, S29S40.
  • Stone, H. and Sidel, J.L. (2004) Descriptive analysis. In: Sensory evaluation practices. 3rd edn. Ed. S.L.Taylor (Elsevier Academic Press: San Diego) pp. 201245.
  • Ugarte, P., Agosin, E., Bordeu, E. and Villalobos, J.I. (2005) Reduction of 4-ethylphenol and 4-ethylguaiacol concentration in red wines using reverse osmosis and adsorption. American Journal of Enology and Viticulture 56, 3036.
  • Wilkinson, K.L., Ristic, R., Pinchbeck, K.A., Fudge, A.L., Singh, D.P., Pitt, K.M., Downey, M.O., Baldock, G.A., Hayasaka, Y., Parker, M. and Herderich, M.J. (2011) Comparison of methods for the analysis of smoke related phenols and their conjugates in grapes and wine. Australian Journal of Grape and Wine Research 17, S22S28.
  • Williams, P.I., Strauss, C.R. and Wilson, B. (1981) Classification of the monoterpenoid composition of Muscat grapes. American Journal of Enology and Viticulture 32, 230235.
  • Winterhalter, P., Sefton, M.A. and Williams, P.J. (1990) Two-dimensional GC-DCCC analysis of the glycoconjugates of monoterpenes, norisoprenoids, and shikimate-derived metabolites from Riesling wine. Journal of Agricultural and Food Chemistry 38, 10411048.