Wines & Vines

Recent research on SO2

Recent research on SO2 – sulphur dioxide

K.C. Fugelsang

Recent research on [SO.sub.2]

Sulfur dioxide is used in winemaking to inhibit browning and growth of undesirable microorganisms including wild yeast, and acetics and lactic bacteria. Because of potential health problems that have arisen – largely from misuse of the product in other food processing applications – the use of sulfur dioxide in wine has recently come under review. As a result of increasing pressure from government, and concerned groups from the public sector, current bottlings must prominently display on the bottle the presence of total sulfites in excess of 10 mg/L. Thus, winemakers are faced with the problem hopefully of finding a viable substitute for the antioxidative and antiseptic action of sulfite in winemaking. The problem has been approached from the viewpoint both of lowering final concentration in wine and eliminating its use altogether.

This paper reviews those areas in wine production where [SO.sub.2] is known to play important role(s) and summarizes results of studies done at C.S.U.-Fresno as well as at Christian Brothers, Mont LaSalle Vineyards in Reedley, Calif. The stimulus for this study lies in the fact that most reports on winemaking using limited [SO.sub.2] come from viticultural areas where inherent grape chemistry is conducive to successful implementation of this technique. By contrast, the chemistry of grapes grown in the warm areas of the Central San Joaquin Valley is frequently less than optimal. Further, the logistics of large scale grape growing, transport, and vinification greatly increase the potential for problems. Thus, if use of [SO.sub.2] were suddenly restricted, the large San Joaquin Valley wineries would, expectedly, have the most difficulty implementing the restriction and still producing acceptable wines.

The majority of sulfur dioxide found in wine is the result of winemaking practices. However, efforts to minimize its use in processing should begin in the vineyard. Initially, harvest and transport of good quality, sound fruit is critical. In this regard, timing of harvest and transport is also critical. In California, vineyards may be distant from the winery and fruit transport time may play a major role in final wine quality. Because the majority of pre-fermentation oxidation is enzyme-catalyzed (via the action of polyphenoloxidases), processing of cool fruit will slow the rate of oxidation as compared with fruit held at higher ([greater than] 90 [degrees] F). This would suggest that evening harvest and transport may become necessary in warmer regions. Further, the practice of adding sulfite to fruit in gondolas at the vineyard as well as additional levels in mechanical harvesting must also be monitored carefully in any attempt to reduce its presence in finished wine.

Given modern processing equipment, handling sound fruit should require little sulfur dioxide. Historically, 50-100 mg/L was used at crush even on sound fruit to prevent the growth of undesirable microbes and impede enzymatic oxidation. Work by White and Ough (1975) indicates that as little as 35 mg/L [SO.sub.2] will inhibit oxidase enzymes originating from the grape. This value, however, varies with variety and cultural practices. In the case of oxidases (laccase) of molds, [SO.sub.2] would be required at much higher concentrations and, in fact, may not be a good inhibitor of oxidative activity Over the past 10 years, increasing experimental evidence and plant-scale production suggests that these levels are not needed and, in fact, excellent wines may be produced from must receiving no sulfur dioxide additions (Long and Lindblom, 1986). A recent survey of 10 California and three eastern winemakers and consultants indicates that, on the average, less than 50 mg/L (and generally, near 30 mg/L) is used in pre-fermentation processing. In cases where bunch and sour rot represents an annual problem, use levels may be as high as 75 mg/L (Zoecklein, 1988).

At the winery, use of inert gassing (carbon dioxides or nitrogen) to purge oxygen from lines and tanks is useful in slowing the oxidation rate. Must-chilling and low-temperature skin contact may also be effectively used to reduce the activities of oxidative enzymes and yet extract the important varietal character in white wine processing. In white wine fermentations, pre-fermentation clarification of must with bentonite also reduces the concentration of oxidative enzymes.

In recent years, ultrafiltration has been suggested as a processing technique for clarification and stabilization of wine and juice. By utilization of membranes with nominal cutoff between 10 and 50,000 daltons, one may be able to achieve three effects simultaneously: (1) clarification; (2) removal of polymeric flavonoids and oxidized and polymerized nonflavonoid phenolics and (3) proteins including enzymes involved in juice oxidation, as well as removal of microorganisms.

Ascorbic acid and its isomer erythorbic acid have been suggested as alternatives to or adjuncts with sulfur dioxide. Grapes have nearly 100 mg/L ascorbic acid. With the exposure to oxygen present in must, the ascorbic acid level falls quickly to very low levels as a result of coupled oxidation involving oxidized phenols. If ascorbic acid is effective in bringing about reduction of oxidized substrates (i.e., quinones to the corresponding dihydroxyphenols), can it be used in processing? Over the years, ascorbic and erythorbic acids have been used and are still being used. In some instances the desired effect (reduction of oxidized substrates) has been achieved. In other cases, oxidation, beyond that already present, was observed. Thus, the problem with ascorbic and erythorbic acids is predictability of results.

Given the historic precedence for routine use levels of 50-100 mg/L, winemakers in the last 15 years began to generally question whether this amount was needed in routine processing of sound fruit (Long and Lindblom 1986; Muller-Spath, et. al. 1978). Proponents of sulfite reduction or elimination in processing point to the following problem areas associated with its use:

(1) Production of aldehydes increased during fermentations when sulfur dioxide was used. Since acetaldehyde quickly binds with free sulfur dioxide, the potential for more [SO.sub.2] is increased to achieve the same level of protection.

(2) Higher initial levels of [SO.sub.2] lead to difficulties in initiating and completing malolactic fermentations.

(3) Levels of wine phenolics, especially those associated with bitterness, are lower in wines processed in the absence of prefermentation additions of [SO.sub.2]. Transitory exposure to oxygen introduced by processing results in oxidization and subsequent precipitation during fermentation. In fact, the color of wines produced without pre-fermentation addition of [SO.sub.2] and limited exposure to oxygen was equivalent to conventionally processed wines. Further, since the oxidizable substrates were removed, wines made from unsulfited juice did not brown further, whereas conventionally processed lots were prone to further deterioration in color.

Long (1986) reports that Simi Winery in Sonoma County began production trials using no [SO.sub.2] in pre-fermentation processing of Chardonnay. She describes the technique as “very successful in terms of producing the desired style of wine.”

Subsequent pilot-scale fermentations were completed with and without pre-fermentation addition of [SO.sub.2]. The wines produced indicate that initially the unprotected juice was visibly brown whereas the protected juice was characteristically green. After fermentation no differences were seen at A520 and the unprotected lot was lower in color than the protected at A420. Further, phenolic levels varied considerably. Wines made from protected juice were considerably higher in phenolic levels than the unprotected juice (values not provided). Follow-up sensory examinations using triangle difference tests indicated that judges were able to detect differences between each lot although there was no strong preference pattern between lots.

Ough and Crowell (1987) examined chemical and sensory changes in wines made from 10 cultivars with and without the use of sulfur dioxide. Each wine was monitored with regard to changes in color, free and total [SO.sub.2], as well as from a sensory point of view. Experimental design in this study called for three lots. One was aerated for 15 minutes (major oxidation), the second lot was sparged with nitrogen for 15 minutes and then blanketed with nitrogen gas. The third lot was not sparged with either nitrogen or air. It received an [SO.sub.2] addition of 50 mg/L after crush. Results indicate that wines made without sulfur dioxide received the worst sensory scores. This was followed by wines produced by initial must aeration and subsequent use of nitrogen blanketing and/or sulfur dioxide addition. Overall, wines receiving additions to must and continued additions to the wines (e.g., conventional methodology) yielded the highest scores. The conclusion was that “sulfur dioxide was essential in making better wines.” The best white wines were made using [SO.sub.2] in the juice and wine.

This report centers on the effects of limiting (eliminating) [SO.sub.2] using Colombard grown in Fresno County. Also summarized is the work of Julia Scott Querin of Christian Brothers’ Mont LaSalle Vineyards in Reedley. Querin used Muscat of Alexandria.

Table : TABLE 1 Pre-fermentation Analysis of Clarified Colombard Juice (Composite).

[degrees]B 20

pH 3.67

TA 6.65 (g/L)

D.O. 4.6 (ppm)

Lot I. Total S[O.sub.2] adjusted to 95 mg/L at clarification. Juice clouded with C[O.sub.2] at pressing and during cold clarification. 420/520 = 0.10/0.021 (4.76)

Lot II. No S[O.sub.2] added at any stage pre-fermentation processing. No inert gassing. 420/520 = 0.142/0.019 (7.47)

Lot III. Total S[O.sub.2] adjusted to 45 mg/L at clarification Juice protected as per Lot I. 420/520 = 0.127/0.024 (5.29)



Tank 1. Total S[O.sub.2] adjusted to 95 mg/L

prior to cold clarification.

Must/juice maintained under

C[O.sub.2] blanket throughout pre-fermentation


Tank 2. No S[O.sub.2] used pre-fermentation.

No special precautions taken to

prevent oxidation.

Tank 3. Total S[O.sub.2] adjusted to 45

mg/L prior to cold clarification.

Inert gassing (Nitrogen) utilized

throughout pre-fermentation


Sublot 1. S[O.sub.2] added at first racking (60 mg/L) (2 weeks post-ferm.); Sublot 2. S[O.sub.2] added at, filtration (4 weeks) at 60 mg/L (4 weeks); Sublot 3. No S[O.sub.2] added post-fermentation.

The experimental design of the project called for crushing and dejuicing 3.5 tons of Colombard harvested from C.S.U.F. vineyards. The juice was collected into three each 300-gallon fermenters. Additionally, a “light press” was done on the sweet pomace and this was apportioned equally into the three fermenters. Ambient temperature ranged from 93-104 [degrees] F. No attempt was made to cool the must as the dejuicing phase was carried out. Fermentation was carried out using the champagne strain of Saccharomyces cerevisiae. Upon completion of fermentation, each primary lot was subdivided into three sublots as seen in Table 1. All post-fermentation processing was done under inert gassing using nitrogen.

It is reported that nonflavanoid phenolics serve as major substrates for browning during pre-fermentation oxidation. Further, they are important sources of bitterness in wine. Utilization of [SO.sub.2] pre-fermentation is expected to extract and protect these substrates from oxidative activity of grape polyphenol oxidases. In the case of juice processed with no [SO.sub.2], oxidation and subsequent browning would be expected to occur. Over the course of fermentation, the brown compounds precipitate and subsequent wines should have color similar to that of conventionally sulfited lots.

Extraction of nonflavanoid phenolics in each sublot was examined by HPLC. A typical HPLC chromatogram shows the presence of intermediate derivative trans-2-S-gluthanthiol caffeonyl tartaric acid. Comparison of values published in the literature is very close. Increased concentration of trans-caffeonyl tartaric acid may correspond to oxidative changes resulting from no post-fermentation additions of [SO.sub.2]. This wine also received the lowest scores in sensory evaluation. The speculation is that the presence of 95 mg/L pre-fermentation protected the nonflavanoid from oxidation and thus it was retained in the final wines. In the case of sublots 1 and 2, [SO.sub.2] was added early enough post fermentation to prevent its oxidation. In the case of SL-3 oxidation may have occurred resulting in a greater absorptivity.

Sensory Examination.

Each lot of wine was compared sensorily. Results presented were derived from bench-top examination of samples. Nine judges were asked to score and rank wines, considering only those parameters normally associated with oxidation or lack of it. These were: color, aroma, bouquet, volatile acidity, flavor and astringency. From a maximum score of 12, result were as follows:

Tk1,SL1 9.1; Tk2, SL2 8.5; Tk1,SL2 8.0.

Tk3, SL1 7.6; Tk3, SL2 7.0; Tk3, SL3 6.6.

Tk1, SL3 6.3(1); Tk2, SL3 5.9(2); Tk2, SL1 5.3(3). (1)Expected, this is a sample not receiving post-fermentation sulfite additions; (2)Expected, worse case scenario; (3)Unexpected, may represent problems associated with small lot production.

It appears promising to note that one of the Tank 2 lots not receiving pre-fermentation additions of [SO.sub.2] scored second. Note first and third places went to protected lots.

All lots from intermediate (45 mg/L) level of pre-fermentation [SO.sub.2] score more or less as a group and the lot receiving no post-fermentation addition of [SO.sub.2] received the lowest ranking WRT oxidative changes.

In the Christian Brothers study, two lots of dry muscat representing approximately 1,200 gallons each were produced from Muscat of Alexandria. Post-fermentation additions were made at racking to yield a final total [SO.sub.2] of 90 mg/L and free of 20 mg/L. The Christian Brothers panel could not determine an appreciable preference for either sample.

From sensory results it is pleasing to see that one of the lots produced without pre-fermentation additions of sulfite scored among the top three (95 mg/L pre-ferm). Consistent with this, all lots receiving 45mg/L pre-ferm were grouped as an intermediate group, and the worst case samples ranked at the bottom. Consistent with other observations from sulfited and unsulfited juices, the color of young wine is similar to controls.


Cattaruzza A. Peri, C., and M. Rossi 1987 Ultrafiltration and Deep-bed Filtration of a Red wine: Comparative Experiments Am J. Enol and Vitic. 32 (2)`139-143.

Long, Z., and B. Lindblom 1986. A Report on Zelma Long’s Work at Simi Juice Oxidation Experiments. Wine & Vines: November, 44-49.

Muller-Spath,H., Moschtert, N. and G. Schafer 1978. Observations in Winemaking: the Present State of Art. Seitz Tech. Comm Reprinted from Die Weinwertschaft 36: 1978.

Ough, C.S., and E.A. Crowell (1987). Use of Sulfur Dioxide in Winemaking J. Food Sci. 52 (2):386-389.

White, B.B. and C.S. Ough. 1975. Oxygen Uptake Studies on Grape Juice. Am J. Enol and Vitic. 24:184.

COPYRIGHT 1989 Hiaring Company

COPYRIGHT 2004 Gale Group