A two liters plastic sealed bottle of Coca-Cola Light was left inside
a garage with no source of warm, for nearly one month. Outside (and
probably inside) the temperature was - 20 Celsius ( - 4 Fahrenheit? )
degrees. What I found was some brownish crystals of ice (0 to 2 inches
diam., not more than 200 grams) and the bottom detached nearly perfect
on the small bottom ridge of the label place. After search, the main
body of the bottle was found 10 feet away, also the red cap in the
nearby. The plastic wall of the bottle is ruptured from the (missing)
bottom up to the neck. The plastic is no longer cylindrical, it looks
more like a flat shell. The thick neck and the cap are broke. Few
other plastic sealed bottles with sparkling mineral water were at the
same place and these are completely frozen but not even one ruptured.
Now, the question: the freezing could produce the ruptures and the
breaks, a possible unfreezing after this could throw away the bottle.
But why is the bottle so expanded?

---

Clarification of Question by michael_2006-ga on 02 Feb 2006 07:23 PST

For <pafalafa on 02 Feb 2006 05:49 PST>: 
No animal signs. That room is made of bricks and all the openings are
tightly closed. There is no food inside, to attract an animal.

Ice requires air.  As water freezes, it sucks in air and expands to a
volume larger than it was as a liquid.  That's why the bottle
exploded.  The air gap at the top of the bottle wasn't sufficient, so
it expanded and popped the plastic.  As for your other bottles, they
must have had access to more air - maybe there was a larger air gap,
or maybe a leak at the lid.

As an experiment, take 2 bottles - fill one completely with water, and
fill the other one only half full.  Put them both in the freezer, and
you'll see that only the full one has popped.  That's because there
was sufficient air for expansion in the half full bottle.

The stretching of the material might be plastic deformation from when
the bottle expanded, right before popping.

AzDoug, water sucks in air as it freezes? Go to the back of the class!

Michael2006, the carbonated diet soda will discharge carbon dioxide as
it freezes (gases are generally less soluble in liquid as temperature
drops, and are far less soluble in ice). As the gas has no place to
go, pressure will build up and it will eventually burst the bottle it
is in. The sparkling water did not do this most likely because it is
not as carbonated or possibly because the bottles were sturdier, or
smaller.

Joseph Priestly described a process for making carbonated water in
1772-- he lived next to a brewery-- and even played around freezing
it.

Water does not require air to freeze.  Ice is the same stuff as liquid
water, just in a solid phase.  If you don't believe me, fill a
zip-lock baggie full of water, seal it, with no air bubble, stick it
in the freezer, and see what happens (ok, I'll tell you--you get a bag
of ice).

As the soda freezes, it expands, pressurizing the gas (mostly CO2)
above it.  The bottle stretches some, but the plastic is stiffer when
it's cold.  When the pressure gets high enough (for a 2-liter soda
bottle at -20 C it's around 10 atmospheres) the bottle ruptures.  This
happens very quickly, and as it does, the compressed gas expands,
thowing bottle fragments.  If the bottle was full of soda all the way
to the lid, with no air, it would rupture during freezing, but in a
very boring way, with no bang or thrown fragments.  As soon as the
bottle gives way, the pressure is released, because the soda (liquid
or ice) is more or less incompressible.  I don't know why the coke
bottle failed and the mineral water ones didn't.

Markvmd, gases are generally less soluble in liquid as temperature
drops?  Might be time to join azdoug there in the back.

http://antoine.frostburg.edu/chem/senese/101/solutions/faq/temperature-gas-solubility.shtml

i like ice.

Rracecarr has pointed out the folly of cut-and-paste without proofing.
I pulled the parenthetical quote from a paper that referred to several
gases, including carbon dioxide, and said "Such gases are...", etc. I
missed the first word of the sentence and the generalization makes for
quite the error!

I stand by the corrected statement, however.

There are several reasons for the rupture. As already pointed out the
most likely is the ice expansion combined with the high pressure from
the gases. I suspect the coke has far higher CO2 level than the
mineral water bottles and hence a higher pressure develops.
However there could also be a factor called "Environmental Stress
Cracking". This happens in plastics which are under strain and exposed
to an Environmental stress raiser. This could be as simple as soap
solution or a more complex chemical such as Coke. Hence the coke could
have acted as stress raiser combined with the pressure from the gas
and freezing. This effectively weakens the plastic and can cause
premature failure. I would need to see samples of the material to
determine if this was the cause. The other deformation of the bottle
is probably due to the material exceeding its yeild point. When a
material excedes the yeild point there is permanent deformation of the
polymer chains and hence structure. Generally the yield point reduces
with temperature. Hence a material with a 10% yeild at 20C may have
only 5% at 0C etc.
Coke bottles are made from PET. The yeild for this material is
relatively low but they are quite pronounced. Once again examination
of the bottle fragments could reveal something more telling.
The bottle design and manufacturing conditions can also play a
significant role. You mentioned the ridge on the lower part of the
bottle. Its highly likely that this has acted as a stress raiser in
the design and hence a potential weak point compared to the other
bottles. The wall thickness and the production speed will also play a
part in the overall bottle strength. If you want to know more check
out www.bayworth.com They are material experts.

There weren't, by any chance, any animal footprints near the bottle
parts, were there?

I think markvmd is correct that CO2 solubility in ice is less than in
water.  Does this mean when you freeze soda, CO2 is expelled?  Or does
it get 'frozen in' making super-saturated ice?  Is there any
documentation?  If not, it's an easy experiment to do...

Rracecarr, freezing soda water (carbonated water) forces out much of
the CO2 but not all. The water freezes in an interesting manner-- it
forms something that looks like snow. As I mentioned, this was an
experiment performed by Joseph Priestly in his home next to the
brewery.

Here ya go:

http://www.elmhurst.edu/~chm/vchembook/122Adensityice.html

The density decreases, the volume expands, and POP!

(my bad - the change isn't due to air becoming infused in the ice, but
rather the molecular structure changes.)  pretty cool!