How could a Coke can leak its contents without any apparent holes?

November 23, 2005 2:38 AM

How could a Coke can leak its contents without any apparent holes?

I came back to eight cans of Coke and Diet Coke that had been left on a low shelf in a cold room for about six months. They were now one month past their use-by date. It was immediately obvious that they had leaked: there was a trail of sticky brown syrup leading away from the cans.

However, when I examined the cans, there were no apparent holes or cuts in them. They seemed completely intact. A couple of the cans still had some liquid in them, but no amount of shaking or turning could cause even the smallest amount to drip out.

The only thing I noticed was a strong vinegary smell, and a kind of white deposit (even small crystals in one case), like a kind of corrosion, on the bottom of the cans. This makes me suspect some kind of chemical reaction, but that's about as far as my chemistry knowledge goes these days.

Any ideas how this could have happened?

posted by chrismear to science & nature (25 comments total)

As befits this kind of unexplainable phenomenon, here is a small blurry photo of the bottom of a can, taken quickly on a cameraphone.



Seems like the deposit was actually brown and shite.
posted by chrismear at 2:49 AM on November 23, 2005

...brown and white.

Goddamit.
posted by chrismear at 2:50 AM on November 23, 2005

In reference to Coke, I think you were right the first time.
posted by dash_slot- at 2:53 AM on November 23, 2005

Well, you know there are seams where two edges of the aluminum are stamped/folded together to form the can. It's not really all that unreasable to expect that every now and then the machine that stamps out those cans malfunctions or wasn't setup properly, leading to an imperfect seam. Combine that with an above average amount of heat or vibration during shipping and the seam would weaken further under the pressure. Then after being left to just sit there for an extended period the soda slowly worked its way out. I would guess that the residue you see now is simply what remains of the syrup after all the water has evaporated, not any kind of chemical reaction.

Honestly, after reading your question I'm more shocked that this kind of thing doesn't happen a lot more often than it does... the machines that make those cans must really have been nearly perfected, because we just don't have to worry about cans failing all that often.
posted by Rhomboid at 2:55 AM on November 23, 2005

cold room

Any chance it froze at some point?
posted by normy at 3:16 AM on November 23, 2005

Rhomboid: But the only seam on an aluminium drinks can is at the top edge, and the cans were stored upright.

normy: Yes, there is a chance that it got below freezing. That's what I initially thought. I would have thought there'd be a more obvious break in the can, though, if it had been forced open by expansion?
posted by chrismear at 3:27 AM on November 23, 2005

The white deposit is almost certainly aluminum oxide. Could there have been salt on the shelf? Aluminum corrodes heavily when exposed to salt. If it's a steel shelf, that could do it, too, because of galvanic action. Freezing would permanently distort the can.
posted by Kirth Gerson at 3:46 AM on November 23, 2005

a more obvious break in the can

I've had a can in the back of a car in winter split due to freezing. A small split of 1" or so, circumfrentially near the base. It wasn't very catastrophic and the can more or less kept its shape - although the split was easily visible. A right sticky mess when it all thawed, of course. I'm wondering if a partial freeze might cause a smaller rupture that wasn't easy to see. Doesn't smell right, though...
posted by normy at 4:06 AM on November 23, 2005

normy: Any chance it froze at some point?

I'd easily explain it that way. We had a couple of 24 packs in the trunk of our car during a trip through half the country one winter. Unfortunately, there were record lows, and when we go to our destination, there were some pretty cool effects: Some cans looked completely normal, except they were completely empty. Some cans looked like they were subjected to a movie-quality explosion, and were ripped appart to dramatic effect.

The best part about it? The soda had time to leak through the bottom of the trunk, and the rest of it froze immediately. There was no staining. None. Freaky, eh?
posted by thanotopsis at 5:19 AM on November 23, 2005

Rhomboid: But the only seam on an aluminium drinks can is at the top edge, and the cans were stored upright.

Lies. it's a cylinder, the bottom piece is attached in some similarly magical way too. That or it's a serious weak point because I've made rockets out of soda cans before and they had a tendency to blast the bottom off.

(Instructions for making self propelled rockets from nothing but a can of coke and my own private invention. Mwhahahaha. There's a surprising amount of gas dissolved in those things.)
posted by alexst at 5:19 AM on November 23, 2005

If there was a leak in the bottom, a lot of cola, or "Coke," if you will, would have shot out of from the pressure. It would weigh less. Try using a digital kitchen scale to weigh a sample of new cans and your suspect cans to see if they weigh less. My guess: The syrupy residue came from another can. Someone opened one and it sprayed. When the culprit sees you going to great lengths to solve the mystery, he or she may confess and beg for mercy. Mercy is for the weak, Chrismear.
posted by planetkyoto at 5:21 AM on November 23, 2005

alexst: Lies. it's a cylinder, the bottom piece is attached in some similarly magical way too. That or it's a serious weak point...

According to this page, they make a cup shape out of one piece of metal, and then add the cap.

planetkyoto: You wouldn't need a digital scale -- they are obviously empty just from holding them in your hand.
posted by chrismear at 5:30 AM on November 23, 2005

Kirth Gerson: Could there have been salt on the shelf? Aluminum corrodes heavily when exposed to salt. If it's a steel shelf, that could do it, too, because of galvanic action.

No salt as far as I know, and six of the cans were still in their plastic wrapper anyway. And the shelf wasn't metal. Good thinking though -- exactly the kind of lateral sciencey thinking I was hoping for.

I'm intrigued by thanotopsis' report of frozen leaked cans that looked fine. If that's possible, then it could explain my cans. I'll have to go back and check them out more carefully for breaks. I am a bit surprised that the room in question actually got down to freezing temperature, though.
posted by chrismear at 5:49 AM on November 23, 2005

The page that you linked to might explain it too. The insides of the can are coated with a lacquer to protect the can from corrosion and it's contents from interaction with the metal.

Coke is acidic (though nowhere near as acidic as most people think). If there was an imperfection in the internal coating maybe it could corrode it's way through until it weakend the can enough to leak. It'd only take a small opening to leak out the Coke because it is pressurized.
posted by substrate at 6:09 AM on November 23, 2005

how about putting an "empty" can under water and seeing if there are any bubbles as air leaks out?

the whole thing seems odd to me, though. no chance someone was playing a trick on you, i suppose?
posted by andrew cooke at 6:27 AM on November 23, 2005

another possibility - could they be fake? i understand (no idea where i got this from) that faking of popular branded goods is pretty common, and i imagine they have poorer quality, so perhaps are more likely to corrode from inside.
posted by andrew cooke at 6:29 AM on November 23, 2005

Um, couldn't the white be sugar crystals?

Shake the cans. If they are sealed they will be tight and hard on the sides, if there is a leak they will not, if there is a tiny leak and the coke is not flat, the can will hiss.
posted by Pollomacho at 7:00 AM on November 23, 2005

A pinhole leak is possible, nay probable. I had a can recently that came out of the 12-pack only half full. At first I thought it was a bottling plant error, but I noticed some sticky residue that implied a leak.

Placing the can on a paper towel overnight verified that there was an active leak, and intensive staring at the can finally located a nearly microscopic hole.
posted by aramaic at 8:18 AM on November 23, 2005

Nothing else could have possibly been spilled near the area? Such as someone tripping and dumping an open can not from this stash, and not bothering to clean it up?
posted by vanoakenfold at 8:47 AM on November 23, 2005

Were these loose cans or contained in a pack? Could something have been spilled on them at an earlier time, say at the factory and you only notices now because they were hidden inside the twelve pack?
posted by Pollomacho at 8:58 AM on November 23, 2005

There was a six-pack wrapped in plastic, plus two loose cans. But it's definitely not a case of something else being spilled on them -- these cans are actually empty or near-empty. At least two of them were definitely completely empty just by the weight of them; some of the others still had maybe a fifth or sixth of the liquid still in them.
posted by chrismear at 9:21 AM on November 23, 2005

Wow, I hadn't realized (or missed) that it was ALL the cans, hmm. Here's my theory then:

Sounds like it started to freeze and ruptured the seals at the top just a tiny bit, spewing some liquid/semi-liquid coke out. This fluid then slowly dried and fermented leaving a vinegary smell and brown tar, plus crystals of the sugar and phosphoric acid.
posted by Pollomacho at 12:18 PM on November 23, 2005

shake them up, submerge them in water. any bubbles?
posted by Joseph Gurl at 6:22 PM on November 23, 2005

From my past experiences working at a grocery store, stocking (among many other things) cans of soda, they leaked. I've seen it happen more often with store brands - cheaper cans - but name brands do it as well, especially if they're abused, either knocked around or subjected to radical temperature changes. I'd say they were dropped too hard, or kicked, or got near or to freezing at some point. Given that the residue is on the bottoms and not down the sides (I guess?), they probably leaked out the bottoms, and as the soda dried, it gummed up the holes. If you washed one of the not-quite-empty cans to remove the residue, you would likely find the hole(s).
posted by attercoppe at 8:08 PM on November 23, 2005

Right, I'm going to try the submerging in water and the scraping off the residue. And maybe take a better picture for posterity. However, I don't live in the same house as the cans (they have a place to themselves, you see), so it'll have to wait until the weekend.
posted by chrismear at 2:51 AM on November 24, 2005