Sulphur dioxide, or SO2, which is produced naturally in the primary fermentation process, is the primary anti-oxidant in white wine. (Red wines have tannins which also help protect the wines from oxidation.) SO2 performs two particularly useful anti-oxidative functions. First, it interacts with hydrogen peroxide in the wine, which is naturally formed as water and oxygen interact with other chemicals in the wine, and prevents those molecules of peroxide that it forms bonds with from forming acetaldehyde (oxidized ethanol), which we perceive as oxidation. Second, it bonds with acetaldehyde once formed, so that we don't perceive those now-bound aldehydes as oxidation.

While a technical discourse on wine chemistry is beyond the scope of this note (and the capabilities of its author[1]), there are three categories of SO2: (1) bound SO2–which simply means that the SO2 has been chemically bound to another compound or compounds in the wine. Chemically bound SO2 is irrelevant as an anti-oxidant because it is no longer capable of combining with other compounds which would, left unchecked, cause the formation of acetaldehyde; (2) “free SO2”–which means the total amount of detectable unbound SO2 in solution in the wine; and (3) “molecular free SO2”–which refers to the true level of free SO2 at the molecular level in an equilibrium state, which varies dependent on the pH of the wine, temperature[2], and alcohol levels.

While the second and third categories are related (molecular free SO2 is a subset of free SO2), it is only the molecular free SO2 that really protects white wines from oxidation in the bottle. There are no simple measuring devices for molecular free SO2. However, there are widely available charts that allow winemakers to calculate the amount of free SO2 (measured in parts per million) which must be added to wine of a given ph to achieve the recommended minimum level of 0.8 mg/l of molecular free SO2.

Molecular free SO2 levels are a direct function of pH–that is the level of molecular free SO2 resulting from a specified quantity of sulphites added to the wine varies directly with the pH of the wine. Thus, to achieve the recommended minimum level of 0.8 mg/l of “molecular free SO2” in wine with a 3.1 pH requires only 16 ppm of free SO2. But for a wine of 3.3 pH, 26 ppm is required to reach the same 0.8 molecular level. It is also important to note that the 0.8 molecular free SO2 is an absolute minimum level. The limit on SO2 is the level at which it is widely detected. Like most other chemical compounds, the ability to detect a given level of SO2 varies within the population. According to Rotter, the range at which the population detects SO2 in wine range from 0.8 mg/l to 2.0 mg/l.

Both Steven Tanzer and Allen Meadows believe that lower usage of SO2 during the 1990’s is the primary cause of the observed premature oxidation of white burgundies. According to Steven Tanzer, “sulfur additions to the wines (including both free sulfur and total sulfur levels at the time of bottling) have trended steadily lower since the late '80s. * * * [N]umerous growers now believe that they have bottled past vintages without sufficient protection against oxidation. In the past couple of years some of these producers have actually increased their use of sulfites.” (IWC, Issue No. 116, Sept.-Oct. 2004).

Allen Meadows is in agreement about the lower level of SO2 and its effect upon the wines, and further attributes a cause for it. In Issue 16 of Burghound (October 2004), Mr. Meadows states that: “in fact during the 1993 to 1997 period, there was a particular fascination in some Burgundian circles with lowering SO2 levels, largely because of the widespread adoption of triage (sorting) tables during this period gave people a certain confidence that overall SO2 levels could be safely reduced because the fruit was cleaner.” In Burghound Issue 32 (4th Qtr 2008), Mr. Meadows notes the widespread recent trend to restore higher levels of SO2: “I would roughly estimate 25 to 30 ppm as typical in Burgundy today versus 18 to 25 ten years ago.”

I would add two additional potential causes of lower SO2 use to the one described by Mr. Meadows, namely the anti-sulphite movement and the rising importance of restaurants as a focal point in the distribution of fine wines in the U.S. The 1996 vintage coincided with the height of the anti-sulphite movement. You may remember the debate over labeling wines, salads and other prepared foods for sulphites that raged at the time. In addition to the typical “food allergy” concerns, some distributors and restaurateurs complained that people rejected or returned bottles of wine as “defective” when they could detect any level of SO2 upon opening the wine. Allegedly some American distributors and importers encouraged or pressured producers to lower their use of sulphites in the wines. John Gilman in the July-August 2007 edition of The View from the Cellar reports that “Certainly producers used lower quantities of SO2 in general with the 1996 vintage, as Christophe Roumier related that many 'vignerons were answering requests from their customers to lower sulfur levels in their white wines.' ” Jadot particularly bragged to Steven Tanzer about the ultra-low levels of SO2 in the 1996 Jadot wines. Girardin made a substantially similar comment to Tanzer touting the low levels of sulphur he uses when Tanzer was reviewing the 1999 vintage.

Some commentators have questioned whether using lower than normal levels of SO2 is consistent with the “almost random” frequency with which premature oxidation of white burgundies is encountered. Actually, the “seemingly random” nature of the failures is not at all surprising once one takes into account that the corks used to seal the wines have unexpectedly wide degrees of variation in their oxygen permeability even within the same batches of a single grade of cork.

A recent study in Australia, which was intended to evaluate various artificial closures and their effect on the aging of wine, found dramatic evidence of unexpectedly high oxygen permeability of corks even within the same batches and same quality grade. The research was conducted by Allen Hart, Research and Development Winemaker for Southcorp in Australia. See "The Role of Oxygen in the Aging of Bottled Wine".[3] Mr. Hart and his colleagues examined bottled samples of 1996 Penfolds Bin 389 Cab/Shiraz blend which were bottled with high grade corks (and also bottled with alternative closures). These wines were chemically and spectrally analyzed at 2.5, 4.5 and 6.5 years after the wines were bottled (which occurred 14 months after the vintage, in 1997). Mr. Hart and his colleagues specifically measured the amount of remaining SO2 in each bottle of wine at each testing point. While they found that the concentrations of remaining total SO2 in the wine were relatively consistent bottle by bottle within the two categories of alternative closures that were tested,

the natural cork had highly variable total SO2 results (eg individual bottles had levels of 22, 38, 41 mg/L in 2000; 12, 21, 27 mg/L in 2002; 13, 13, 18 mg/L in 2004) …. The evidence presented here confirms that there is the potential for considerable variability within cork populations to oxygen ingress and that this variability is a most likely contributor to 'random oxidation.' “

Other more static tests have found considerable variation in the oxygen permeability of corks. An earlier internal Southcorp study in Australia, which is cited in the study quoted from above, found a 1000-fold variation in the oxygen permeability of 35 randomly selected “Reference 2” corks. A similar study done on 12 corks by the Austrialian Wine Research Institute produced substantially identical results. But both tests, which were run on a MOCON machine, were criticized as unrealistic because the bottle is tested in an upright position with “dry” cork. On the other hand, a 2005 University of Bordeaux study, which measured the permeability of corks in wine bottles stored on their side, which contained a colorimetric solution which changes color in response to increasing oxygen exposure, found only a three-fold variation in the oxygen permeability of corks. However, this research was part of doctoral research of the author which financed by Amorim, a cork producer. [4]

Indeed, taking into account the wide degree of variation in oxygen permeability in corks, the general trend to use lower or minimum levels of molecular free SO2 could be expected to cause nightmarish bottle-by-bottle variation, which is exactly what is now being experienced with the 96 Jadot and Girardin whites, which were proudly trumpeted by their respective winemakers as having been made with ultra low levels of SO2.

In my view, given the documented variability in permeability of corks within a batch used to bottle a single wine, usage of appreciably lower levels of free SO2 is the most explanatory factor in the “higher than expected but seemingly random” premature oxidation being seen today in many of the white burgundies from the 1995-2002 vintages. Moreover, lower levels of molecular free SO2 would only compound any problems associated with excessive batonnage or peroxide washing of corks.

Happily there is some good news to report. Both Allen Meadows and Steven Tanzer have reported that the typical white burgundy producer, beginning with the 2004 or 2005 vintage, has increased the use of free SO2 at bottling. However, since the premature oxidation problem does not generally begin to reveal itself in earnest until approximately seven years after the vintage, we won't really know whether this is having a positive effect on the oxidation problem until 2011 or 2012.

[Text and opinions by Editor Don Cornwell. © Don Cornwell 2005-2008]

[1] For those interested in a technical explanation of the role and effects of SO2 in wine, see Ben Rotter's excellent article on SO2 in winemaking at

[2] According to the Rotter artice (see note 1): “As temperature increases, free SO2 increases and bound SO2 decreases. (SO2 bound to acetaldehyde remains constant.) This is because increased temperatures cause partial dissociation of the bound SO2 form, resulting in increased free SO2 and hence increased molecular SO2 concentrations.

For example, a wine containing 68 mg of free SO2 at 0 C (30 F) will contain 85 mg at 15 C (57 F) and 100 mg at 30C (84 F) [Peynaud, 1984]. Sudraud [1977a] showed 64 mg/l of free SO2 at 16 C increased to 120 at 48 C and to 200 at 80 C, as BSO2 was released.”


[4] P.Lopes, C. Saucier & Y. Glories, “Nondestructive Colorimetric Method to Determine the Oxygen Diffusion Rate Through Closures Used in Winemaking,” Journal of Agricultural and Food Chemistry, Vol. 53 2005

  • sulphur.txt
  • Last modified: 2018/11/23 03:45
  • by doncornwell