How much oxygen gets into white wine must during grape processing?

¿Cuánto oxígeno penetra en el mosto de vino blanco durante el procesado de la uva?

Martin Day
The Australian Wine Research Institute (AWRI)
Glen Osmond, Australia

The processes of harvesting, crushing and pressings, as well as fermentation involved in making white wine all potentially expose grapes, must or wine to oxygen. Measuring dissolved oxygen (DO) in situ during these processes has shown some interesting figures. Using a technique which gives winemakers and scientists actual values of DO, AWRI researchers have identified that high DO values occur in must at crushing. Similar measurements made during pressing appear lower indicating that crushing is the stage that can pick up most oxygen. Similar O2 exposure during machine harvesting is highly likely, especially if berry damage is prevalent.

Why would a winemaker be interested in knowing about oxygen pick-up during grapes processing? A lot is known about dissolved oxygen during micro-oxygenation (MOX) (Gómez-Plaza & Cano-López 2011) and oxygen ingress during bottling (O'Brien et al. 2009) and bottle ageing (Skouroumounis et al. 2005) but little is known about oxygen during vinification . Two major groups of compounds in must are affected by exposure to oxygen: phenolics and aroma compounds. The browning that occurs during traditional oxidative handling of must is caused by quinones of caftaric acid and flavanols generated through the action of polyphenol oxidases and oxygen on these compounds (Cheynier et al. 1990). Quinones are highly reactive radical species which can create cascades of further oxidative damage. Nature's antioxidant in grapes is glutathione and the intrinsic concentration depends on the variety. This peptide reacts with the quinones of the hydroxycinnamic acids (caftaric and coutaric acids) to form grape reaction products (GRPs) which are then largely protected from further oxidation. In addition, aroma compounds are also potentially exposed to the oxidative damage wrought by the reactive quinones, although the extent and mechanisms are not widely known, especially whether aroma precursors are involved. However, the effect of pressing and oxidative damage on thiol aromas has been studied on Sauvignon Blanc wines (Patel et al. 2010). Protection from oxygen for certain grape varieties during pressing is considered essential by some and use of a range of techniques to stop this happening, such as the use of dry ice or other inert gases, is common . Membrane presses that can be protected by inert gas have now been available for a number of years from several manufacturers. Anecdotal evidence shows that wines produced this way are fresher and more vibrant (Osicka 2010).

But how much oxygen is involved and if we are protecting the press, what is the impact at the crusher? Are winemakers ‘locking the stable once the horse has bolted?' .


Measurements were made using Nomasense equipment at the Josef Chromy winery (JCW) in Tasmania, Australia and at the University of Adelaide 's Hickinbotham-Roseworthy Wine Science Laboratory (HRWSL) on both destemmer-crushers and membrane presses. Nomasense DO measurement works when light from a blue LED, shone down a fibre optic cable, excites an O 2 sensor spot to emit back a red-coloured fluorescent signal. When oxygen is present, the fluorescence signal is ‘quenched' in proportion to the amount of oxygen in the juice (Anon 2010b). In these experiments, the oxygen-sensitive dot was placed inside a sight glass so this can then be placed anywhere along a pumping line. The sight glass was orientated so that the dot was at the bottom to maximise contact with flowing liquid. The potential of altering the sample when taking a sample for analysis in the laboratory is therefore avoided by making readings in situ .

A Bucher-Vaslin ‘Delta' destemmer-crusher was used at JCW where the must is fed into an Enoveneta peristaltic pump by a short closed screw with the DO being measured at the output of the peristaltic pump. At HRWSL, two crushers were used: a Diemme destemmer-crusher and a Demoisy 7EP crusher-destemmer. The sight glass was placed just after the screw drives below the crusher rollers before the must pump.

More than a dozen batches of several varieties of white grapes and Pinot Noir were measured during the 2010 vintage from both hand-picked and machine-harvested fruit.

Results and discussion


The average DO value during crushing measured over 12 batches of grapes was 6 mg/L with a standard deviation of 2 mg/L. The DO values of three batches of Riesling crushed at the JCW was over 9 mg/L which is close to the saturation level in grape juice. Measurements were taken only as the must pump was operating (Figure 1) or, for smaller loads at the HRWSL, when a reasonably steady state had been achieved. The DO values measured on two small-scale crushers at the HRWSL appeared more consistent and were lower than at JCW. Although there are only a small number of mean values, the crusher size and design do appear to influence the amount of oxygen that can be taken up by the must. It must be noted, however, that the values quoted are those read directly from the DO meter which is calibrated with respect to oxygen solubility in water. Oxygen solubility is in fact measurably lower in grape must and to a lesser extent in wine therefore actual DO concentrations will be lower (Nevares 2011).


Figure 1: Mean DO values during crushing operations. Error bars = 1 sd . WSL: Hickinbotham Wine Science Laboratory. JCW: Josef Chromy Winery. mh: machine harvested. CHA: Chardonnay. RIE: Riesling. VIO: Viognier. SAB: Sauvignon blanc.[Click here to enlarge the image]

Actual real-time DO profiles during crushing with a Bucher-Vaslin Delta destemmer-crusher are depicted in Figure 2. The DO values oscillated around a mean value probably due to variations in supply of grapes into the hopper of the crusher. When the must pump stopped, the DO values dropped quickly showing that the oxygen dissolved in the must around the sensor was quickly consumed. Typically, the rate of localised oxygen consumption around the sensor spot was around 0.5 mg/L/min. In the bottom right graph of Figure 2, the fast increase in DO is due to water push-through. On the time frame taken to fill a press and with little additional oxygenation the must DO will be near zero when pressing begins. This can explain why DO values measured after pressing appear much lower. Because of this, DO readings after the press do not indicate to how much oxygen a must has been exposed. For grapes that can be very sensitive to oxidative spoilage, such as Sauvignon Blanc, inerted commercial crushing equipment is now available (Anon 2010a).

Figura 2

Figure 2: Real-time DO profiles during crushing at JCW. Clockwise from top left: machine harvested Pinot Noir, Sauvignon Blanc, Riesling; hand-picked Riesling. [Click here to enlarge the image]

Obviously, this situation of grape ‘damage' activating enzymatic oxidation can also occur during machine harvesting. Depending on the equipment used and the degree of mechanical shear involved, oxidation is likely to occur in an uncontrolled manner.


In both wineries, the sight glass used to measure press DO was placed after a buffer tank. For the smaller Willmes press at the HRWSL, a shallow 1,000 L press tray (with dimensions the same as the foot print of the press) below the drum was protected with dry ice. At the JCW, a smaller press tray (250 L) integrated in the Bucher XPert 250 fed into a larger buffer tank (1,000 L) not protected with inert gas. When the press pump operates to transfer the juice to tank, the DO values represents a dynamic reading but when the pump is off, the DO value can be described as static and allows the measurement of oxygen consumption of the juice at that particular composition in the sight glass.

A typical DO profile in a commercial-scale press is shown in Figure 3. The DO value starts at 3 mg/L although the values during crushing were around 9 mg/L. The initial decrease on measured dynamic DO is the oxygen sensor spot acclimatising to the low DO environment. Intuitively, as there is a considerable air intake into the press upon press bladder deflation, a subsequent increase in DO should be observed. However, in Figure 3, the first deflate and crumble occurs at 60 minutes (draining time included), before which there had been four rapid increases in DO. This can most probably be explained by splashing in the empty buffer tank. The triangles along the time axis indicate when this occurred, immediately after which the DO increased.

Figura 3

Figure 3: DO profile during pressing of crushed machine-harvested Riesling. (Bucher Xpert 250 with ORTAL program) [Click here to enlarge the image]

Again as during crushing, when the juice is static, the DO decreases due to enzymatic consumption of oxygen to oxidise the phenolic material present. Typically, the rate of decrease calculated from several press cycles ranges from 0.13 to 0.350 mg/L/min. As shown in the last third of the cycle in Figure 3, the rate of oxygen consumption (dotted line) decreases as the press cycle progresses. This behavior was seen for several press runs.

 Inerted pressing

 Several manufacturers supply standard tank presses that can be inerted using carbon dioxide or nitrogen. The press is filled with inert gas before filling and inert gas is used when the bladder is deflated. Recycling of this gas has been incorporated into the design by at least one manufacturer.

In the experiment at JCW, Chardonnay was whole-bunch pressed using a Willmes ‘Sigma' press operated using nitrogen supplied from a nine-bottle manifold. The press was not inerted before charging but all subsequent juice collection and bladder deflation occurred under protection of nitrogen.

Initially, the DO of the juice reached a maximum value of 2 mg/L during the dejuicing step. For the remainder of the press cycle, the DO never exceeded 0.18 mg/L. The emergent juice appeared a more vibrant yellow-green colour and clearer. Unfortunately for practical reasons, this was the only press run that could be measured with inert gas cover.

In-tank DO values post-pressing

The DO of the pressed juice was measured immediately post-pressing with a dipping probe containing the same sort of oxygen sensor used in the sight glass. Several of the tanks were remeasured on subsequent days during cold settling. The DO values and position of the probe are summarised in Table 1. Storage volumes of the tanks measured ranged from 1,000 L to 20,000 L .

It can be seen from these data that the DO in juice post-pressing is very low due to the rapid nature of enzymatic oxidation which will consume oxygen picked up during processing. The limited option for oxygen pick-up across the juice processing chain compared to the total oxidative potential of the must and juice means that the oxygen actually dissolved will be the limiting factor, explaining the very low levels observed after pressing.

Table 1. Dissolved oxygen immediately following pressing and at various later time points
Tabla 1



This limited survey of dissolved oxygen in must and juice has finally given actual numerical values for DO during juice processing. In doing so, the crushing step has been identified as a potential source of oxidative damage for aromatic grape varieties. It also indicated that handling after the press can also be source of oxygen uptake adding these areas as potential critical control points in the winemaking process. The major sensory implication of oxygen exposure in white wine must relate to the amount of GRP produced or the amount of remaining hydroxycinnamic acids – the major phenolic compounds in young white wine. Unfortunately, the sensory characteristics of GRP are unknown as yet but might hold the key to understanding differences in palate structure.

Ultimately the wine style is chosen by the winemaker and the processing decisions used. To achieve fresh, vibrant wines by reductive techniques, protection of the whole juice processing chain needs to be considered.



This article was first published in the AWRI Technical Review No.189 Dec 2010.



Anon: Diemme Kappa 15 Neutral [website page], accessed 28 October 2010 from, 2010a.

Anon: NomaSense Oxygen Analyzer Sets [website page], accessed 28 October 2010 from, 2010b.

Cheynier V., Rigaud J., Souquet J.M., Duprat F., Moutounet M.: Must browning in relation to the behavior of phenolic compounds during oxidation. Am J Enol Vitic 1990; 41: 346-349.

Gómez-Plaza E., Cano-López M.: (2011) A review on micro-oxygenation of red wines: Claims, benefits and the underlying chemistry Food Chem 2011; 125: 1131-1140.

Nevares I., del Alamo M., Day M., Waters E.: La medida del oxígeno disuelto en vino y mostos, ¿Concentración o presión parcial?: estado actual del conocimiento. XIII Congreso Latinoamericano de Viticultura y Enología. Santiago de Chile, 2011.

O’Brien V., Colby C., Nygaard M.: Managing oxygen ingress at bottling. Aust NZ Wine Ind J 2009; 24 (1): 24-30.

Osicka, S.: Press technology to reduce juice oxidation: practical results from an industry trial. En: Blair, R.J., Lee, T.H., Pretorius, I. (eds.). Proceedings of the 14th Australian Wine Industry Technical Conference, 4-7th July 2010, Adelaide, South Australia. Australian Wine Industry Technical Conference Inc., Adelaide, South Australia, 2010.

Patel P, Herbst-Johnstone M, Lee S.A., Gardner R.C., Weaver R., Nicolau L., Kilmartin P.A.: (2010) Influence of juice pressing conditions on polyphenols, antioxidants, and varietal aroma of Sauvignon blanc microferments. J Agric Food Chem 2010; 58 (12): 7280-7288.

Skouroumounis G.K., Kwiatkowski M.J., Francis I.L., Oakey H., Capone D.L., Duncan B., Sefton M.A., Waters E.J.: The impact of closure type and storage conditions on the composition, colour and flavour properties of a Riesling and a wooded Chardonnay wine during five years' storage. Aust J Grape Wine Res. 2005; 11 (3): 369-377.