BONDING GLASS AND ALUMINUM

I am sorry to bust your bubble on this but what you propose is technically not possible unless you let go of some of your stated intentions.

The expansion coefficient of aluminium and glass are so different, Al is 22.2 m/m.K x 10^-6 and glas is varying between 4 for pyrex and 9 for plate glass.

As you can see the two will come apart if you use a glue that follows one of the two components expansion.

I therefor suggest you use an epoxy type filler glue that is strong enough for your vacuum and good enough to hold at the tempetratures but flexible enough to bridge the expansion between the glass and the aluminium.

This would simplify your search a bit as the expansion rates are not often given for the glues.

Good luck and I hope you will call back here.

Thanks Case,

For bringing my attention to the muchoo different expansion rates.

I'll try to look for epoxys, I used to cure expoxys in an oven at 200-300 deg,

but I don't know if it can stand 500 F. And I really don't know if it'll work on glass/aluminum. But it's cheap to try.

all the best

Jens

This is not true. There are glasses which will bond to aluminum and bond good. I am not sure what end application are. If one can fire the part at 575C then there is company Call Global Applied Material have leaded glass as over glaze glass you can apply and use.

If you like to have upto 500C then it will be crystallizable glass but there is lot a way to make this happen. If I know more details I can recommend a glass composition or glass manufacturer who can have this for you.

575 glass is over glaze glass manufactured by Global Applied Materials. This company is in Chennai, India

Masyood

Thanks Masyood for the info.

This is misconception we all have. It is interface CTE which is important along with CTE mismatch between glass and aluminum. Leaded glass CTE can be changed drastically

by changing melting procedures. Like melting glass under reducing conditions will increase CTE and melting in excess air will decrease CTE. Increasing silica will decrease and there is ways and ways.

One need glass formulation which is making interface compatible and CTE less than 20% different the system will work

Masyood

For most of the glass it is correct but you can have glass with thermal expansion coefficient of 14PPM. The bond of glass to alumina is done using two method

1. Developed interface as transition inter phase glass chemsitry is based on lead-boron-Tellurium bismuth oxide base will give transition from metal to metal glass composite to glass and this bond will be non sensative to expansion mismatch and will have full reliability and usability to 700F

2. Other is anodize aluminum surface then bond to glass this is one of the conventional method use

3. Select glass compostion which will be seed type crystallizable glass and will also work. Top of my head I do not know the correct composition.

At Global Applied Material they have glass which bond very good on gold, platinum, copper and nickel. This is low to high temperature usable glass paste and they also have metal paste filled with glass which bonds very good to metal, ceramics and glass

If you need I will provide you the contact information . you can email me on Masyood@gmail.com

Masyood

I tend to agree. That's a very tall order. I think the hardest part is the 500F, and I suppose you want it to be crystal clear too? Maybe you could make up a mechanical way to hold it together not unlike that is used for stained glass construction. I'm a glass blower, and do flame work too. Perhaps you could melt the glass together, but I think the thermal expansion difference would make it imposible, even if you use borosilicate glass.

No Bricktop,

The color in not important, the bond is not in line with the laserbeam.

All the best

Jens

I have patent on the glass formulation as low as 275C and as high as 1700C Recently I have developed glass which will bond good on Invar and Kovar but will need inert atmosphere to process so it does not get oxide. This is interesting area and depend on what you want to achieve at the end of day. Glass is very forging and Glass ceramic you can make at low temperature and use at high temperature since it is no more glass but ceramic and is use able below melting point which is at least 1.5 to 2 time of glass transition temperature.

If you defind end use clearly I can guide you through

MAsyood

"Recently I have developed glass which will bond good on Invar and Kovar but will need inert atmosphere to process so it does not get oxide."

Do you still have to oxidise the Kovar to get the glass to wet to it?

Are you able to but other metals besides aluminum? I am looking for hermetic seal in glass to metal without having to oxidise the metal to begin with. glass must be clear of defects in final process.

Most metals are oxidized prior to sealing because molten glass wets metal oxides. This helps to achieve a strong bond. Expansion matching and oxidizing however, are not always essential. Housekeeper seals are made between thin metal foils ( often platinum) and glass without the aid of oxide. Thin metal foils yield at low stress so expansion matching is not required.

I don't know if your design allows it, but this seems ideal for a clamped gasketed joint. But, if you're bound and determined to glue it, call up your local rep for Loc-Tite. Is there anything magic about aluminum? Could another material work as well?

Thanks TVP for the locktite suggestion.

There's no special reason for using AL,

416 steel or CU is also ok, but AL is easier to work with. Softer and will hold the vacuum for a long enuff time.

all the best

Jens

Jens,

While I was in the USAF I replaced alot of aircraft windows (windscreens). The aircraft side was aluminum and the other side was varing types of "glass" from plastics to polycarbonate, thick to thin. What our procedure was to put a "parting agent" on the airframe, a different "parting agent" on the glass side and a "bonding agent" between the two of them. This procedure allowed for thermal expasion, "g" loading and unloading changes on both structures and wide temp.changes while the entire time over coming internal pressurization /de pressurization to compensate for altitude changes. I do not remember the "agents" but I am certain that one call to your nearest airport that repairs high performance aircraft and they will give you the current agents and where to buy them.

Thanks Mike,

I was in the AF myself for 6 + years in avionics. (a loooong time ago).

Do you remember what that parting compound looked like and how hard it was when

it was cured. I guess before you aplied the bonding material.

all the best

Jens

The agents were in plastic tubes that we put in a chalking gun like tool and pushed it out. We required a steadystate room temp. of 70-75*f for 24 hours before the process. I believe the parting agents went on then we brought all of the together within 4 hours. Cure time was 12 hours then cabin pressure test.

You may want to check silicone caulking compound. It should meet most of your requirements. It's mich better than you may think. Get specifications from Corning, GE, etc.

If there is any heating up of the glass box Silicone may throw off a gas.

As well as Loc-Tite, and Dow Corning you could also try Devcon and Permatex.

Hi,

Try this site. Good luck.

http://www.dymax.com/products/glass/index.php

There are many types of "glass" out there in the world. You can find fused silica that has virtually no CTE, to softer glass that moves all over the place. There are companies that specialize in glass to ceramic and glass to metal seals.

Many of these same companies will make sputtering targets. You may want to add that to your google search.

Another thought here from a mechanical/optical point of view, if your refraction coefficient is anything you want to work with, you want the glass to have a tightly controlled perpendicularity to the beam of light.

This cannot be guaranteed with a too thick a layer of bonding material between the glass and the aluminium housing.

For this reason alone I would be inclined to go with a thin layer of sealing compound and a clamp down structure.

Hi,

the caution remarks above are really important!

I glued aluminum 2017 (20mm thickness) plates 10 x 10 cm with one side 20x100mm

onto sheet (float) glass 8 mm thick: the Epoxi was Araldite 2020 a very fluid epoxi suited for glass and aluminum.

All worked well until the glass at almost no temperature rise gave initiation of cracks growing in many directions.

The epoxi is good the glass is too brittle the aluminum is too strong and the CTEs do not match!

This will never work!

Possibilities:

A. Why not clamp as suggested above?

Get a Viton O-Ring for above and below, no need to clamp if you evacuate.

Not too tight at elevated temperature but the best you can get with plastics or rubbers - depending on needed pressure.

Maximum Viton temperature is 150°C - too low for you requirements but what about cooling?

B. Glue a thick Viton O-ring or square-ring on both sides with a good epoxi that fits your demand. (Look may be at Polytec). The rubber will be flexible so that both parts can expand according to the CTE and don't stress each other.

C. Make flexible your epoxi by adding rubber spheres or hollow glass spheres and use a thick layer of epoxi (calculation necessary!).

D. Use parts from a vacuum supplier: they have everything ready to use. Tubes with flanges, orings and seals are ok, but windows are expensive as windows need flexible mounting parts.

E. Make your aluminum part very thin walled and flexible so that the expansion forces on the glass are small.

Choose yourself and report please about leks and success.

RHABE

Thanks everybody,

for your suggestions

I'll have to do some thinking and make a decision.

all the best

Jens

Sorry to be a bit late jumping in to this thread, I've been using up my 2007 holiday allowance.

We regularly have to join glass parts to steel/ceramic assemblies where we can't always match the expansions. We don't normally mount glass to aluminium but have the same problem with differential expansion. This is normally overcome by bonding the glass into a thin walled metal component which is then welded to the main body. We weld because we need a vacuum tight seal, you could use some other joining method. The expansion difference is taken up by flexing of the side wall in the metal part holding the glass.

Hi Nigh,

I think I understand what you are doing and how the stress is minimized.

You slide and glue the egde of the glass into a U shaped metal frame and weld the edge of the metal to whatever you wish to mount/seal it to.

(with a MIG or TIG). Am I correct ????

If you could send me a picture or small drawing of it, that would clarify things.

all the best

Jens

That's about right, we use a glass/metal bond but only because we need a vacuum seal

Try to ignore the extraneous bits in this sketch I have pulled from one of our drawings. The convoluted part is to protect the glass/metal seal from stress & to give a long thermal path to avoid the weld heating the glass/metal bond. The metal parts are about 0.5mm thick.