By | December 3 2015
The Champenois have long recognized the potential problems that can result from exposing sparkling wine to air during the disgorgement process, but they did not know the precise volume of oxygen ingress until a groundbreaking paper was published by the Comité Interprofessionnel du Vin de Champagne (CIVC) in 2003. This demonstrated that the issue was not simply one of volume (of oxygen), but of the wild variation in that volume from bottle to bottle. Prior to disgorgement, there is no oxygen in a sparkling wine, the yeasts having consumed it all during the fermentation process. Even the dead yeast cells in the bidule next to the crown cap act as an antioxidant barrier to the minuscule ingress of oxygen through the closure. After disgorgement, dissolved oxygen is introduced via the dosage and topping-up, but the amount added is insignificant compared to the volume that enters directly into the headspace during the disgorgement and corking. Anyone who worries about differences in cork porosity (as I do) would be horrified by the much greater volume and variation of oxygen entering the headspace during the disgorgement and corking of sparkling wine.
Once this problem was recognized in Champagne, the industry realized that it had to devise an appropriate defense. In the same year that the CIVC’s paper was published, Bertrand Robillard demonstrated at Moët that foam level is the principal factor in determining the air intake for Champagne, just as it is for beer. Moët had been jointly researching physical and chemical properties of bubbles with Heineken since 1989 under a EUREKA agreement and knew that the brewing industry had developed and used jetting technology since the early 1960s. In 2005, Moët began trialing the beer industry’s continuous-stream form of jetting, but its adaptation for wine was not ideal. By 2009, LBM Industries in Reims had patented a nitrogen-based pulse-jetting technology that is only now beginning to filter on to the disgorgement lines of all types and sizes of Champagne producer. But not, of course, in Trento, where Ferrari read the same paper and engineered its own jetting solution. I hope it also patented it.
How jetting works
In theory, jetting should be able to eliminate all externally sourced oxygen, but in practice oxygen is reduced to less than 0.5mg/l, and because the equipment is often calibrated to 80-90 percent efficiency to prevent overfoaming (which causes "fallback," literally sucking air into the bottle), less than 1mg/l of oxygen is a more realistic figure. LBM also claims that jetting will reduce the SO2 component of the liqueur d’expédition by "about 30 percent." Although this is theoretically true, in practice too many producers already use too little SO2, thus any reduction in their SO2 will only exacerbate the problem.
Most chefs de caves in Champagne are enthusiastic about jetting, but some are cautious. A small number claim to have trialed the technology and believe that jetting leaves their Champagne too "closed," or they simply prefer the results of non-jetting for all or (more commonly) some of their cuvées. These criticisms might well be valid in specific instances, but they all ignore the most important findings of the CIVC’s 2003 paper. They might prefer a faster development for one or more cuvées, but they are not achieving that evenly. Instead of rejecting the unprecedented advantages of this amazing technology out of hand, I would ask those chefs de caves who think that jetting is not right for them to think how the technology could be adapted to make it right. In theory, it should not be too difficult to engineer a metered dose of oxygen into each pulse of sulfited water. So, instead of jetting guaranteeing less than 1mg/l of oxygen, it could be calibrated on-site to provide a homogenous 1mg, 2mg, 3mg, and so on, depending on the cuvée being corked. This would turn jetting into a far more flexible technology that would appeal to even more producers.
For award-winning content from the world’s most respected and intellectually satisfying wine magazine.