Sparkling wine is a bit of a strange substance to work with in a laboratory as it contains those pesky bubbles. Of course the bubbles are why we call it sparkling wine. But what exactly do we mean by ‘sparkling wine’?
The definition actually varies from country to country and Champagne, the most famous sparkling wine of all, is strongly defended against imitation.
In Australia the definition of a sparkling wine comes from the Australian Foods Standards Code, in particular Standard 4.5.1. (1). I won’t reproduce this in full but the main points are:
From the above definition it can be seen that the creation of the carbon dioxide in the wine must be derived from the “fermentation of contained sugars”. Sugars can be added, but if you want to call the product “sparkling wine” you can’t add carbon dioxide by bubbling in gas from a cylinder, as soft drink manufacturers do.
The main steps in making sparkling wine are:
1.Making the base wine by primary fermentation. This is normally a white or pink still wine. Of course in Australia we also produce our famous Sparkling Shiraz, from a full bodied red wine.
2.Producing the carbon dioxide from the secondary yeast fermentation in a sealed container, usually a bottle, but special tanks can also be used. This secondary fermentation also produces extra alcohol.
3.Removing the solids such as dead yeast cells from the wine and adjusting some final parameters such as sugar content, alcohol and sulfur dioxide.
Which all sounds fairly easy, but of course the reality is that this can be a very involved process. There are three main production pathways that are followed in Australia; the Traditional Method (also called Methode Champenoise), Transfer and Charmat methods. Each technique produces different tasting products and command different price points in the marketplace.
The following is from the Wine Australia website (2) and is written by Ed Carr, Group Sparkling Winemaker, Hardy Wine Company.
“We use the Charmat method for the commercial wines as it delivers bright fresh fruit characters with less yeast contact. We even do some sparkling reds this way."
“Transfer method is always seen as a lesser process but it does have the advantage of bottle fermentation and yeast contact can be controlled. It delivers more quality at a lower cost."
“Even with all the work we’ve done, traditional method has the edge on quality and suits smaller volumes.”
So where does this lead to in the lab and what has to be done differently with sparkling wine compared to still wine?
Claire McDonald is the Laboratory and Quality Assurance Manager at Domaine Chandon in the Yarra Valley. Domaine Chandon is a subsidiary of the long established, highly experienced and huge Champagne house Moet and Chandon. As Claire says, “the main problem with testing sparkling wine is that it is full of bubbles and these get in the way of normal testing”. As all the wine made by Domaine Chandon is by the Traditional method, we will restrict ourselves to the testing required for this process. The other production techniques do use a similar regime.
At the base wine stage, one of the differences is that sparkling base doesn’t necessarily have to be dry. The final sugar content of still wine is important for the style being produced. But as sparkling wine has another dose of sugar added, the sugar content at the end of the first fermentation isn’t that critical. Of course the amount of sugar in the base wine does have to be known accurately prior to the addition of the tirage liqueur so that the total sugar content is known.
Once the base wine has been blended, it has to be checked for heat and cold stability prior to secondary fermentation. One precaution that Claire takes for the cold stability test is that extra alcohol is added to the sample of base wine to emulate the concentration in the final wine after the secondary ferment. As the final wine increases in alcohol by approximately 1%, this is roughly the amount added to the cold stability sample.
The main purpose of the secondary ferment is to add carbon dioxide. As the Australian standard above states, this has to be a minimum of 5g/L. A typical quantity of sugar required for the secondary fermentation is 24g/L to give greater than the specified minimum of carbon dioxide.
The amount of carbon dioxide generated is not usually measured in grams per litre, even though this is how it is specified in Food Standard 4.5.1. Instead the most convenient and common way to check the level of carbon dioxide in a bottle is simply to measure the pressure in the bottle using a special type of pressure gauge called an aphrometer. Typically a pressure of 6 to 7 bar is required. Work done by Domaine Chandon equates this to about 12 to 13 g/L of carbon dioxide in the finished wine, which is well above the specified level.
As sparkling wine made in the bottle can use either crown seal or cork during the whole process, there are 2 corresponding types of aphrometers. Crown seals are normally used during the secondary fermentation stage and so the main type of aphrometer used is the type that pierces the crown seal. The type that pierces a cork is also used for the finished wine but they are sometimes harder to use and are not as common.
According to Claire there is a new device on the market that is an in situ pressure gauge that is placed inside the bottle at tirage. Of course the actual bottle with the gauge in it cannot be sold but is used as a representative specimen of the whole batch.
There are other benchtop devices that are used to measure the carbon dioxide in a bottle; however these are not as commonly deployed as aphrometers.
One trick to remember when measuring the sugar content of sparkling wines is that sucrose (cane sugar) is normally added at both the tirage and dosage stages. Sucrose is not detected when using the normal tests for the reducing sugars glucose and fructose.
One molecule of sucrose is composed of one molecule of fructose and one of glucose bonded together. This bond gets cleaved by enzymes or acids to form fructose and glucose. When testing sparkling wine for total sugar, incorrect results can be obtained if all the sucrose is not converted to glucose and fructose. Prior to the test a small amount of the correct enzyme or strong hydrochloric acid can be added to the test sample to ensure all the sucrose has been hydrolysed.
The other main tests done on a sparkling wine are no different to a still wine. However, it is difficult to take an accurate aliquot by pipette of wine that has bubbles in it. In our labs we allow the sparkling wine to go relatively flat before we are able to do the test. By allowing the carbon dioxide to escape, some of the measurements may have a slight error. For example, the analysis of sulfur dioxide normally requires that the wine sample is exposed as little as possible to air prior to analysis to minimize any loss of sulfur dioxide. It is not possible to pipette the exact volume of sparkling wine required unless some of the carbon dioxide is released by leaving the wine open to the atmosphere for a short period of time. However it is virtually impossible to test the wine with gas bubbles any other way.
Claire McDonald also does the same thing, stating the slight error that may be introduced is unavoidable. Of course great care must be taken to not let the wine sit for too long during degassing and thus allowing other volatiles to escape that may impact upon other test results. This is a compromise that is not necessary with still wine.
One way we have found of minimizing the degassing time is to use a drop of antifoam solution, as this can also help to stop any bubble formation during the testing process.
Sparkling wine is one of the great wine styles, and one that finds favour with many people. Having carbon dioxide in the wine under pressure does make it a bit more difficult to deal with in the lab compared to still wine. A number of aspects of this difference have been discussed to shed some light on how to get around this peculiar property of sparkling wines.
1. Australian New Zealand Food Safety Authority, Food Standards Code Part 4.5, see http://www.foodstandards.gov.au/thecode/foodstandardscode.cfm
2. Australian Wine and Brandy Corporation, Wine Australia website,
Article from the Australian and New Zealand Grapegrower and Winemaker, Issue 537, 2008
Page Number(s): 84-88
Author: Greg Howell
Copyright© 2013 Winechek Laboratories. No part of this publication may be reproduced in any form or by any means without the permission in writing from the copyright owner.