Why do the bubbles in Guinness Stout float down?

26-May-2000

Dear Cecil:

A subject perhaps worth extensive research: Why is it that poured Guinness Stout bubbles appear to float downward in the glass? --Manuel Martinez, Rancho Cordova, California

Cecil replies:

Spending a lot of time staring at beer glasses, are we, Manuel? But they say Crick and Watson, who discovered the structure of DNA, found daily inspiration at the pubs, and I've done some of my best work there myself. Although admittedly my breakthrough, instead of the double helix, was to discover that Stroh's spelled backwards was "shorts."

Now then. I'm not a big drinker of Guinness, which always seemed to me like something you'd pour on pancakes. But this is the Straight Dope, goddammit. Since I didn't care to be seen strolling into a gin mill at nine in the morning, I went down to the supermarket and bought myself a bottle of Guinness Stout. Went home, got a genuine tapered pub-style beer glass given to me by my brother-in-law's English wife. (I recognize that Guinness is made in Dublin, but England, Ireland, it's all the same, ain't it?) Poured said stout into said glass. Intensively scoped the bubbles. Noticed that they floated up. Thought, cripes, this Manuel has gotta lay off the sauce. But just to be sure, I went on-line and searched for "Guinness Stout" and "bubbles" and "down." Unfriggingbelievable. Not only do the bubbles float downward under certain conditions, but people have devoted, well, maybe not their lives, but a good two or three days to figuring out why. And not in vain, either. But I'm getting ahead of my story.

From my on-line reading I learned that the secret of Guinness's creamy mouthfeel, as the taste experts somewhat indelicately put it, is a mix of nitrogen and carbon dioxide rather than pure CO₂ as the bubblizing ingredient. Nitrogen bubbles are much smaller than CO₂ bubbles, a mere 50 microns in diameter, and produce a nice smooth head. But nitrogen doesn't produce bubbles as spontaneously as CO₂. At the corner tap they deal with this by using a special nozzle that aerates the stout with nitrogen as it's poured. In packaged goods that's not possible, so for a long time they were stuck with pure carbon dioxide. But science marches on. Using brain cells that might have cured cancer or solved the third-world debt crisis, but hey, we've all got our priorities, a couple engineering types invented a little plastic device, known in brewing circles as a "widget," that's placed in each can of Guinness Draught. (See patent info at www.ivo.se/guinness/patent.html <http://www.ivo.se/guinness/patent.html>.) When you pop the top and pour, the pressure in the can drops to ambient, stout squirts out of the widget, and nitrogen is liberated from solution and aerates the exiting beverage. Result: nitrogen microbubbles galore, same as if Murphy the barkeep did the pouring.

I hustled back to the supermarket, grabbed a four-pack of the tall black cans, and gave them a shake. Something inside rattled--a good sign. Took 'em home, stuck 'em in the fridge for the prescribed three hours, then popped one and poured. This was accompanied by a satisfying whoosh and--I tell you, it's a fascinating thing to watch--cascades of tiny bubbles, sliding down the inside of the glass.

All right, now for the part where I actually answer the question. Tiny or not, nitrogen bubbles ought to have more buoyancy than the surrounding liquid. Why don't they? I consulted Clive Fletcher, professor of computational fluid dynamics at the University of New South Wales, Sydney, Australia. Fletcher, a disciple of the Crick and Watson school of liquid research, was wondering one day why the bubbles went down. A chemist friend, one of these if-all-you've-got-is-a-hammer-everything-looks-like-a-nail kind of guys, tried to explain the phenomenon in chemical terms. Bah, thought Fletcher. Using computational fluid-dynamics software from Fluent Inc. that modeled the activity of the bubbles in the glass, Fletcher found that what occurs is similar to convection-- the bubbles rise in the center of the glass, where you can't see them because of the stout's opacity, then heel over and skitter down the sides, their buoyancy overcome by the viscous drag of the roiling brew. (See illustration at www.fluent.com/news/pressre l/guinness/tsld001.htm <http://www.fluent.com/news/pressrel/guinness/tsld001.htm>.) Simplicity itself, really.

To further extend the frontiers of knowledge, Professor Fletcher says he's thinking about trying to "persuade Moet & Chandon to fund some cross-industry research, with plenty of field trials." Sounds like a worthy goal to me. Naturally, professor, you'll need a second test site north of the equator, so as to control for the Coriolis effect. Let me know if I can be of help.

--CECIL ADAMS