DO + HO = TPO (the new equation for successful bottling)

Published by Sally on May 18, 2011

A version of this first appeared in Drinks Business magazine, February 2011.

Forget closure OTR (oxygen transmission rate) for the moment. In closing up a wine bottle, TPO (total package oxygen) is where the TLAs (three letter acronyms) are at, and the bottling operation is the bigger oxygen issue by far.

TPO is the sum of dissolved oxygen (DO) in the wine plus oxygen in the air of the headspace (HO), and whilst many folk have got their heads around DO, the idea of TPO is still relatively new.

Warren Roget, technical manger at the AWRI (Australian Wine Research Institute) said “normal bottling processes entrain oxygen in the bottle. There is oxygen dissolved in the wine, and there is oxygen in the headspace. Our work shows that 60 to 70% of the total package oxygen is typically contained in the headspace.”  What is concerning, added Roget, is that “it’s typical industry practice to measure DO in wine, meaning the majority of the oxygen in the bottle is not being measured.”  In practice he added, this means QA specifications may state permissible DO levels at less than 1mg/l, but when “you look at TPO, they may be closer to 3mg /l because they’re not measuring oxygen in the headspace.”  Wine specifications should be moving to detail TPO instead of the less relevant DO.

Measuring TPO has only been possible for a few years with the development of non-destructive (i.e. not needing to open the bottle of wine) measurement kits, such as PreSens, OxySense and NomaSense. Standard laboratory equipment only measures DO.

Importantly, TPO is a snapshot measure immediately after the bottle has been packed. “Three months after bottling” said Roget “the TPO will virtually all be consumed by the wine. It is from this point forward that closure OTR becomes the important factor in regulating oxygen intake into the wine. However significant quality and shelf life impacts may already have been incurred.”

Management of headspace is therefore one of the most critical areas for control of oxygen ingress at bottling.  Sometimes remedies are simple, though incur costs: using inert gases, for example nitrogen, or carbon dioxide, to flush out oxygen in the filling tanks, in pipework, the empty bottle, the headspace, prior to the wine being transferred.

According to closure trials done on riesling at Geisenheim Research Centre, Professor Dr. Rainer Jung found that after nearly a year, cumulative OTR varied significantly across screwcap and synthetic, from “0.5 to 2.5 mg/l in total, which is not very much.  It is not enough to oxidise the wine.”

But a different picture emerged with headspace trials. Jung said “we measured 6mg/l oxygen in the headspace. It takes about 4mg/l SO2 to reduce 1mg/l oxygen, so if you have 6mg/l of oxygen, you need 24mg/l SO2.” The first AWRI closure trial identified oxidative characters developing in white wine at about 10mg/l free SO2, and whites are commonly bottled with 30 to 40mg/l free SO2, so, said Jung “if you don’t want to lose of 24 mg/l SO2, flush the headspace [with inert gas].”

Jung highlighted the snapshot significance of DO and TPO, saying wine can arrive at the bottling line with 1-2mg/l DO already in the wine, though this completely depends on what has happened to the wine before, and its style. For example a micro-oxygenised red versus a reductively made sauvignon blanc. And he said, of a reductively made wine, where oxygen has always been kept away “in the last step, you pump into the bottling tank, and get the same amount of oxygen uptake. This will directly react with wine components, so no DO is measured, but the aromatics and phenols can be oxidised.” So the wine is in specification, but its defining characters have been lost.

Even some reds won’t benefit from oxygen at bottling. Stéphane Vidal, global oenology director at Nomacorc, said “syrah is quite reductive. If you add oxygen at bottling, you are simply wasting sulphur dioxide” and shelf life. He added that improving bottling TPO by 2mg/l “could save one year of shelf life of the wine”, if the bottle is closed with their Nomacorc Classic+, for example, which has a 2mg/l OTR over the course of a year.

Headspace management is also a critical control point for traditional method sparkling wine. Michel Valade, responsible for viniculture at the CIVC in Champagne said “the quantity of oxygen that might enter the bottle at moment of disgorgement varies according to conditions of disgorgement.” CIVC studies showed that the amount of oxygen introduced at disgorgement varies from 1 to 10 mg/l, and averaging 2 to 4mg/l.  Valade said that during ageing [sur lattes] “only 1 mg/l per year of oxygen enters the wine, and is consumed by the wine, so 2 to 4mg/l is the equivalent of 2 to 4 years of oxygen transmission through the closure.”

Valade explained “at the moment of disgorgement some bottles let some mousse escape, in which case there will be no oxygen entering the bottle [as the effervescence expels headspace oxygen]. But if bottle is very quiet, or stays a bit longer on the line, up to 6 mg/l may enter the bottle.” This clearly creates big bottle variation which will directly affect the flavour profile. 

The ideal is to have zero oxygen entering all bottles at disgorgement.  And to this end, the CIVC are developing a technique already used by the brewing industry, and, according to Valade, by some of the big Champagne houses, though the CIVC are still completing trials. A tiny amount (20 μL, or 0.02 mL) of wine is injected into the bottle under pressure, after dosage, immediately before the bottle is closed. This provokes the wine to effervesce, which expels the headspace oxygen. 

“Very soon” Valade said “this technique will be widespread in Champagne.” It “is not very expensive, and can easily be installed on the disgorgement line.“ Added to which, he said, achieving a close to zero ingress of oxygen at bottling means producers can use less sulphur dioxide, to achieve a lower final sulphur dioxide measure.

The level of TPO that industry should be aiming for depends on individual wine style, though “generally the lower the better” said Roget. Vidal said their studies showed “on average, DO is 1-2ppm mg/l. And headspace is 1-4mg/l, giving a TPO of 2 to 6 mg/l.” So, he said “A TPO of 2 mg/l is therefore already a good situation – a bottling line that is working fine.“

For the majority of wine which is drunk within two years of bottling, this is all crucial, as high TPO at bottling quickly erodes shelf life. And for these wines, the wine should be ready for drinking at the point of bottling.  Jung said “During bottling and storage the lowest quantity of oxygen coming into the bottle would be the best way to keep the wine in a ‘ready to drink’ situation.”  Roget added “a high TPO can have a shelf life reduction equivalent to 10 years of oxygen transmission through a Saran-tin screwcap.” He said it’s a “completely different order of magnitude with TPO versus OTR.”

It’s clearly time to focus attention on the moment immediately preceding closure.

CASE STUDY – Reh Kendermann GmbH Weinkellerei

Reh Kendermann (RK) moved from measuring DO to measuring TPO in March 2010, with the purchase of NomaSense equipment.  Their winemaker Phillip Maurer explained that some years ago their contract customers wanted to know about oxygen management, so “we started to measure oxygen input at all the different situations in the cellar – racking, filtration, centrifugation, blending, the whole bottling – tank, filter, filler, bottling line.” 

Measurement, by sampling bulk wine was relatively easy, enabling RK to control the whole process. Flushing pipes, tanks, bottles etc., and blanketing wine with carbon dioxide minimised oxygen uptake. But this only measures DO.

With NomaSense, the TPO is measured and the “main goal is to have a TPO in bottle below 2mg /l.” Maurer added that such a TPO measure is now included as standard in the wine specification for Black Tower and the Kendermanns wines.


4 Responses to “DO + HO = TPO (the new equation for successful bottling)”

  1. Jim Peck Says:

    I ran these calculations fifteen years ago when at Gallo, comparing the mg of oxygen in a bottle resulting from DO and headspace vs OTR. It suprised me and absolutly shocked the winemakers. If you do a proper job of headspace purging, a wine can be bottled with lower SO2. Our Technology department at G3 did a lot of DOE work on our mobile bottling lines optimizing the purging of oxygen from the headspace.

  2. Sally Says:

    Thanks for your comment Jim. Did you come up with a definitive protocol of headspace purging at bottling (share?), and by how much were you able to reduce SO2 levels? Any inherent risks with mobile lines versus factory bottling lines? Sally

  3. Malcolm Says:

    Well said Jim. We have discussed this a year or two back. It just goes to show that slavishly following a 0.8ppm regime for SO2 at bottling is not the total story. TPO – including the O2 in the closure in the case of cylindrical closures and well as DO in wine and O2 in the headpsace must be considered. OTR comes later in the life of the wine.

    Cheers, Malcolm

  4. Homan Says:

    Many thanks. Very interesting publication. I believe it particularly useful for cool climate winemaking.


Jump to the top of this page