Unwanted secondary fermentation

Q: Back in March I bottled a rosé that was quite acidic (think Muscadet acidic). There was no residual sugar, and the pH was 3.28. It had gone through coarse filtering, 1 micron, but not sterile filtering. Four months later I opened a bottle to find that it had gotten cloudy and fizzy. My wine had become a sparkling rosé. Could it have gone through malolactic fermentation or some other process?
— Jack Kerr

A: Thanks for writing in with your fizzing rosé mystery. I can see why it caught you by surprise — there’s nothing quite like popping open a bottle you meant to be still and finding it has turned into a sparkling version of itself. Let’s talk through what might have happened in your case.

You described a wine bottled in March, bright and very crisp, with no residual sugar and a pH of 3.28. You ran it through a one-micron filter before bottling but stopped short of a proper sterile filtration. Now, four months later, you’ve uncorked something that’s cloudy, spritzy, and decidedly different from what went into the bottle.

What you’re most likely seeing is a classic case of a secondary fermentation in the bottle. Even though your wine was dry, if it had any malic acid left in it, lactic acid bacteria could have survived the one-micron filtration and carried on their work after bottling. The process they carry out — malolactic fermentation — turns malic acid into lactic acid. It softens the wine, alters the mouthfeel, and just as importantly here, gives off carbon dioxide as a byproduct. That gas, trapped in the sealed bottle, is what made your wine sparkle when you opened it. The haze you noticed is another calling card for microbial activity.

The key detail here is your filtration. A one-micron cartridge will polish up a wine for the time being, removing a lot of particles and leaving it brilliantly clear to the eye, but it is not small enough to stop bacteria. True “sterile” filtration in the wine world only happens when the filter has pores of about 0.45 micron. At that level, you can reliably exclude most microbes — both bacteria and yeast. Without that finer step, there’s always a chance that a few tiny cells sneak through and set up shop in bottle.

Could yeast have been responsible instead? It’s much less likely. Since you had no residual sugar left, the yeast would have had nothing to feed on. At a pH of 3.28 — decidedly on the zippy side — and assuming your sulfur dioxide levels were in a normal, protective range, most spoilage yeasts would have been held at bay. That’s why I put my money on lactic acid bacteria here. They are uniquely comfortable working in a wine environment, and even though your acid level is high, they are robust enough to persist, especially when given the chance to slip through a one-micron screen.

Sometimes winemakers are tempted to stop malolactic fermentation (MLF) short or to bottle wine that hasn’t been sterile filtered, thinking that dryness and low pH will be protection enough. Usually things work out — but every once in a while, you get a surprise like this one. If you want to guarantee stability in the bottle, especially for high-acid wines that you’d like to keep crisp and fresh, sterile filtration at 0.45 micron is really the most reliable path. 

Sulfur dioxide management also plays a role. Keeping your free SO₂ at the right level for your pH will help discourage microbial growth, though it won’t make up for incomplete filtration. Higher-pH wines need higher levels of SO₂ to be effective, and you don’t want to add so much that you smell or taste it. Your pH is so low that I think bottling with a free SO₂ of 30–35 ppm is great . . . but at that dose it’s still not a substitute for a sterile filtration.

So where does that leave you? Mostly with a great learning experience and a story to tell at your next tasting. Your rosé wasn’t “ruined” (many people love a little sparkle, after all) but it wasn’t what you had in mind. Next time you want a wine to stay perfectly still, make sure you either finish MLF completely and also protect the wine with SO₂, or keep it from ever starting in the first place with sterile filtration.