[rjones7486]

So I have been trying to work on a way to produce and temporarily store Hydrogen for a project I am working on. I've tossed around ideas of creating an HHO generator, but found that they don't really produce that efficiently given the amount of material and energy that is required to be put into a decently working system. So recently, I have begun working on producing hydrogen via a muriatic acid and aluminum reaction and have had some measure of success. There has just been one problem that has been puzzling me that I'm hoping that maybe someone could answer:
 My setup consists of 3 main containers. A generic glass jar used for storing bits and bobs, that I sealed using PVC thread stick and epoxy, a refrigerator container that I've also partially filled with 600ml of water and sealed, and a balloon fitted onto the threaded end of a compression barb fitting. I have a container of Muriatic acid at a concentration of 31.45% that I poured about a centimeter's worth of into the glass container (sorry that I don't have a more accurate measurement), and I then place a wadded ball of aluminum foil into the container and screw it tightly shut (I had put alot of effort into making sure the airtight seal could be maintained despite the container being opened and closed multiple times). After roughly 15 seconds, the solution will slowly begin to react.
 I think it may be worth noting at this point that I am doing this outdoors and the air temperature where I live has floated between 38-42.

So I have the reaction jar connected to the "bubbler"/refrigerator container, via a vinyl tube that is connected to a barb compression fitting on both containers. The barb fitting on the "bubbler" has a 4 inch brass tube connected to it that protrudes half a centimeter into the water. There is a second barb fitting on the top of the "bubbler" that has a small vinyl tube going from it to the barb connector of the balloon.

After the reaction began, bubbles started come out of the brass tube and the refrigerator container swelled from the air pressure, and the balloon began to fill with hydrogen (despite the mishap, I made sure that the balloon was kept filled and stored to confirm hydrogen had been made later). About 2 minutes into the reaction, the balloon was almost full, and everything was going normal (aside from the relatively chaotic nature of the acid breaking down the metal), and suddenly I heard a small pop and all of the liquid flushed out into the container I had holding everything in one place. After cleaning up the liquid (rice works wonders), and I inspected the container for signs of what had happened, and all that I found was that the glass had perfectly split, as if cut, along the line where the liquid had been during the reaction. There was no fragmentation or shattering as you would have expected from too much air pressure (and honestly had that been the case, I would have expected the seal on the barb fitting to have broken first), just a simple, laser-like precision of a break. Nothing I can think of or know of could account for this other than that the ambient air temperature was largely different than that of the liquid, which during the reaction, has made the glass very warm to the touch, but definitely not HOT. I am thinking that the temperature difference was most prominent in the glass where the surface of the fluid warmed the glass to a much higher temperature than what the ambient air had been maintaining the top half of the glass at. I don't think this container was designed to handle any for of temperature, but the temperatures seemed negligible. Am I on the right track in guessing what is wrong, or is it something completely different (I.E. the acid reacting with something in the glass) ? I am trying to resolve this issue so that I can continue to build this reactor.

[Borek]

Apart from temperature difference I can't think about anything else.

[skbuncks]

Did you inspect the lass container prior to use for any hairline fractures?

skb

[Stepan]

Two common safety rules -
#1 you do not use flat-bottom vessels for pressure or vacuum, it may either collapse or explode.
#2 you do not use non-laboratory glassware if you expect heating or cooling, common glass does not take temperature gradient.

In your case you had a jar which was hot on the bottom and cold on the top. In result it formed micro cracks along the hot-cold zone. Elevated pressure multiplied by the area of the flat bottom (F=P*A) created force which "separated" the bottom from the jar.