Sciencemadness Discussion Board

REpairing Pyrex (borofloat) glass with soda lime glass?

DrMario - 5-3-2015 at 10:07

Let's say you have this nice and very big piece of laboratory glassware. It was custom-made a decade ago. Some a**hole hit it and caused a crack on a receptacle, which is now leaking.

Repairing Pyrex glass requires torches working at very high temperatures, unavailable at my institution. My idea is to use a normal bunsen torch to melt a soda lime glass tube and let it drop on the pre-heated part with the crack. The adhesion of soda lime glass on Pyrex glass should be excellent, and it should effectively seal the crack.

What do you ladies and gents think of this idea?

Let me add that we need this apparatus fixed ASAP, so we don't have the time to ship it to a master glassblower. Just packing it up correctly, would take quite a long time.

Dr.Bob - 5-3-2015 at 11:22

The thermal expansion of the two glasses is different, it will likely crack or break upon heating. If there is only a crack, then just heating it up enough to partly melt the glass should fuse it without the need to add more glass, I have had that done with many pyrex items in the past. But don't try to mix glasses. Soda will not adhere to cold Pyrex, as best as I remember, and if if did, it would not last once it cools.

And just go buy a MAPP glass torch, that would liekyl get hot enough to melt Pyrex, but without annealing, the item will never be very strong, so the torch is not nearly as key as the annealing oven. Even if you don't have a scientific glass blower nearby, you may have a art type glass blower with the right equipment to repair a simple crack, most have an annealing overn and know how to work glass.

Sulaiman - 5-3-2015 at 12:26

I was given a Leibig condenser by a friend at work because a chunk of glass near the water inlet was smashed off leaving a hole and several cracks.
Using a butane/propane mix plumbers torch the borosilicate cracks 'healed' nicely just leaving a triangular hole about 1cm across.
I used a piece of soda-lime glass which softened then melted nicely to seal the hole .... beautiful
slowly cooling down fine cracks accompanied by a horrible crackling sound, unrecoverable damage.
So I can confirm that;
A) I am an idiot, I knew of the problem but thought I'd try anyway
B) the different coefficients of thermal expansion are VERY important
C) a plumber's butane/propane mix torch is hot enough to melt borosilicate enough to 'heal' cracks.
Torch used
Gas used

not much of a loss as both of us use Quickfit and this was an old style that uses bungs & tubing but it may have had some use. and I didn't want to sacrifice any of my existing borosilicate glassware.

Lesson learned.

[Edited on 5-3-2015 by Sulaiman]

Dr.Bob - 5-3-2015 at 18:47

Also, you can find borosilicate pipettes and they might work to help fill a crack, but if it is just a crack, there should be no need to fill it. The biggest risk is that if you don't heat it slowly and cool it slowly, the stress will build up and if not annealed, it might eventually crack. But I have done small repairs without annealing, and they survived. What I do however is heat the area slowly and evenly, do the repair, then gently keep the area warm and slowly move the heat away so the area cools slowly. It is tough to do well, but even a little bit if gentle flame annealing can help.

If the crack is not exposed to solvents, you could even try a windshield crack repair kit, which uses cyanoacrylates, but you said it is leaking, so I presume that it sees solvent.

MyNameIsUnnecessarilyLong - 5-3-2015 at 23:55

I wonder if sodium silicate (waterglass) could work to stop leaks in glass.

careysub - 6-3-2015 at 17:53

Quote: Originally posted by MyNameIsUnnecessarilyLong  
I wonder if sodium silicate (waterglass) could work to stop leaks in glass.

I've wondered the same thing on this site, but got no replies. Still wondering. However, I am pretty sure it could only be used on something that did not experience great temperature variations.

(My case was the return tube of a Soxhlet extractor, which I would only use with low BP solvents.)

I have seen it recommended that in the absence of an annealing oven, immersing a glass piece in a container of dry perlite (thus slowing cooling) can help (I would lay it on a bed of perlite, and dump more perlite on top). Preheating the perlite in a regular kitchen oven might help too.

IrC - 7-3-2015 at 00:50

I think Kaowool is better than perlite at keeping glass from cooling rapidly. It is available in several forms from thick pads to fiberboard but the fluffy form as in the image is what I use. I first tried it after Fleaker sent me a large box full and I would never go back to anything else when I wanted to keep glass cooling as slowly as possible. I use a big piece to lay hot glass on and cover with a like piece. Completely blankets the glass, Kaowool is light so very delicate glass is quite safe. A large box with a lid and two large pieces of Kaowool works better than anything else I have tried.

Kaowool640.JPG - 40kB

DrMario - 7-3-2015 at 18:01

Quote: Originally posted by Dr.Bob  
Also, you can find borosilicate pipettes and they might work to help fill a crack

If the crack is not exposed to solvents, you could even try a windshield crack repair kit, which uses cyanoacrylates, but you said it is leaking, so I presume that it sees solvent.

I am not aware of borosilicate pipettes? Only soda-lime glass pipettes. There may be some graduated glass pipettes that are actually borofloat, but they are usually quite large.

This container only sees aqueous solutions, but at high temperatures (over 120C).

Texium - 7-3-2015 at 19:00

I've seen borosilicate glass stirring rods for sale. I bought some for about 50¢ apiece once. You could definitely use one of those to patch the crack, no need to ruin an expensive pipette.

DrMario - 8-3-2015 at 14:45

I just remembered that I have a few broken borofloat wafers (with some patterned metal on top, which I can easily etch away).
OK, next step is to find a MAPP torch.

Texium - 8-3-2015 at 14:51

Perhaps to help it anneal you could gently wave it through a cooler propane torch for a while, and try to let it set properly. It wouldn't be as good as a proper annealing oven, but it would be something.

DrMario - 9-3-2015 at 11:54

Roger that, zts16.

WGTR - 10-3-2015 at 05:41

My own 2¢: For about $50-100 of firebrick, a ramp-soak PID controller, and about $10 of Kanthal, one can anneal all the glass that one's heart desires.

I would like to agree that thermal expansion coefficients are quite important in glass work. I had a terrible time joining soda-lime tubing to soda-lime test tubes. The joints would crack without fail. Come to find out, even soda-lime glass among different venders can have different expansion rates. At least Boro is more user friendly in that the expansion coefficient is less. If one has some broken glassware from the same vender lying around (and not "weathered" too badly), that might be an ideal source of filler material if it is needed for the repair.

Lab glass repair is an art form, though, and takes practice. I wouldn't chance ruining an expensive piece of lab ware unless I had practiced on a lot of other pieces first.


bb wooster - 24-3-2015 at 16:06

Hi all, I haven't been on this forum before but got here due to this thread. I haven't done lab type glass work before but have used 3/4" borosilicate rods in some art pieces. I heated them using our oxy-acetylene torches and we didn't anneal them. They are about 15 years old and have never broken. Did I just get lucky or is it because they are thicker than the thin walls of lab type glass? I'm asking because I have a project that calls for some thicker (1 1/4" diameter) rods to be bent and although we have an old kiln now that could be used if shorter, these pieces are going to be close to 6' in length. I don't know of anyone anywhere near me that has a kiln big enough to accommodate them. Thanks for any advice knowledge you can pass along.