Glass on Glass. How do you seperate?

Two suction cups on either side while running water across. This might let you slide the glass across instead of trying to pry it apart.

Always sounds good in theory

Hmmm,

maybe support them lightly on edge in a pressure chamber and suck the air out, maybe they'll separate...?

then agin ...maybe they won't.

The pieces are held together by air pressure. a vacuum chamber should be the first choice.

If it do not do the trick you may consider temperature difference.

Glass is a good heat isolator.

Suspending the big piece with the small piece hanging down.

apply a little heat to the small piece.

First try by just rubbing and increase if considered safe for the optical elements.

Try warm water and a drop of detergent then slide apart.

Will a penetrating oil damage the optical element? Try WD-40 or Liquid Wrench...spray some on and let the top piece "float" to one of the edges.

I would try an ultrasonic water bath, those cavitation bubbles get into everything. I'll bet they get in between the two sheets quite easily.

there is merit in using an ultrasonic bath that is .01 N in KOH because this vigorous action on the micro scale will lead to fluid mobility close to the adhered points and might erode the sticking points, which will be a little more mobile than bulk glass.

Of course, at some point you will have to make the hard decision and fabricate another...or was this a purchased part with optical patterns on it?

If so, the makers must have had this problem and know the precautions as well as the cures?

do you have a test piece to play with using the suggestions here?

yes I have a couple I have been playing with

it is called "wringing" it is very typical way to mate too gage blocks together for making measurements...........i am having a hard time getting a good definition on the web, but i think that it is a molecular bond that forms when too parts are polished flat

the blocks are difficult to get apart, but you simply "unwring them" by twisting them.

in the case here, i believe that becasue the glass was opically flat, tne bond is STRONG. also with the gage blocks, they are easy to hold and won't break if you drop them.........

Do you have access to lasers?

I can get access a a couple of local job shop. what is the suggestion?

I was thinking if nothing else worked, and you knew the chemical and mechanical differences of the glass, that you might be able to use an IR laser to both heat and help break the bonds between the pieces... On the other hand, I'm guessing that if these bonds are the real problem, putting the pieces under a relatively intense UV light would probably disrupt the bonds. UV is good that way.

True if it's a chemical bond, but what if it's just surrounding air pressure?

I'll vote for the water and detergent or my own suggestion, try blowing pressurized air or nitrogen between them.

I'm thinking the dish washing detergent should work. Detergent acts as a wetting agent to get water into places it doesn't usually get. If that doesn't work, go to your local nursery and buy a bottle of wetting agent - try that.

Surfactant - sa-fact-and it's true .

I'd also reccomend a coat hanger wedged in the mouth just in case.

More seriously , it will be interesting to hear what does the job . Good luck

Check tribology as well. Who knows where inspiration may come from.

Is it 2 pieces moulded into 1 piece with the optical element in between? In that case, u do not have a chance of separating the 2 pieces. However if u are sure that the 2 pieces were initially separate pieces and have been stuck together by a cynoacrylate or any other medium, then u can try keeping the piece in water / kerosene overnight. The pieces will automatically separate if they have been stuck together.

these were two separate pieces of glass cleaned and rinsed in DI water and the small, thinner one was laid on top of the larger. No glue or anything was added.

I know that a lot of heat would not be good, but perhaps drying the part out with low heat/air may loosen the bond. Also to station the .040" glass to a table andwith a suction cup "tool", push down while rotating the thin glass layer to mechanically separate.

Please let us know how this ends!

What may help is gently heating them. If one has a different coefficient of expansion, it may break the bond between them. However, as you're heating, I'm guessing you'll have to stay on top of it like cougar with a chicken (Where the hell did that come from?!), because you want to catch them just when the expansion is the most different. Shouldn't take too much heat, either.

mech eng: The larger piece of glass, if only 0.040" thick, may flex enough to allow you to get a razor blade under the edge of the small piece. You might try clamping the edge of the large piece in a vise and applying side pressure from the small glass side. Good luck. DickL

High pressure air through a small orifice aimed directly at the edge of the glass? I am thinking it would create a slight hovercraft effect.

Just a guess though.

What has been tried.

Bending the large wafer and trying to insert a razor blade. Broke the edge off the wafer.

Soaking 3 days in Alcohol. so far no luck even when trying to pick the edge with razor blade and bending the wafer.

Has anyone done this or something like it?

Mech Eng

lift a long edge of the wafer 1/100 inch and place a thin piece of stiff mylar under the glass. allow it to fall back. flood with detergent and water, advance the mylar a little and wait. This will allow a progresssive and patient wedging action to take place. advance it a millimeter every hour. At some point you will see a lift and release

i have no thoughts at the moment, but it is interesting enough, that i'll keep mulling it over,

PLEASE let us know if you are ever successfull

I have experienced this same phenomena with two polished ceramic disks. ONLY way I couls seperate them was soak them in water for a day or two. Then they would seperate. NOTHING else worked! Does anyone out there have an idea as to what mechanism is in play?

I suspect he has some bonding, since they are very flat and close on atomoc level there will be many touching points and he can get some local dissolution and rebonding with some mobile ions moving around and going into places in the matrix

This is an optical contact bond - I believe it is an ionic bond at the molecular level. It is used in many different products where the bond needs to be vaccuum tight with a very long shelf life. It requires very flat surfaces and a close CTE match (which I assume these two glass pieces are the same material) and it lasts a very long time. Heat will actually cause the bond to strengthen, unless you can create a large temperature gradient between the parts (which I highly doubt since the glass is so thin). Depending on how deperate you become, I would suggest cooling the whole set and then trying a torch (in ~3 second dwells) isolated on the thin piece to see if you can create a high gradient. The risk is of course fracturing the glass due to the fast change in temp.

Yikes! If he's right, throw the thing in the freezer, then try prying them apart.

What you have done here is inadvertently created a common optical element bonding technique. Your two sections of glass (what type of glass would be nice to know).

The very strong interfacal bond is created when you have two flat sections of glass (generally the same glass) that have been polished to < 0.1 um Ra. This type of bond requires no adhesives, bonding agents, or coupling agents. The bond is primarily an Si-0H to Si-O bond. With time and temperature the bond can become inter diffused.

Now how to get them apart with out killing them. Almost no way possible. The next time be sure to put a conformal release film down. You can use a number of agents, but a simple one is a 1% solution of micro 90. Use a polishing cloth, wipe on, rub off with a very clean soft lint free class 100 mate.

What you need to do is attack the Si-OH---O-Si bond, which is easy to break if you can get to it. Depending upon the glass, a 0.01% solution (by weight) of KOH (potassium hydroxide) will attack the Si-OH---O-Si bond. Be very careful if you have the concentration too high or the glass type is low temperature slump glass, and some borosilicate glass then the KOH solution will etch the glass.

Good luck, I think temperature or bending is going to just fracture the thin element.

I think he will have to try to force some of the dilute KOH bond dissolving fluid into the gap and gradually weaken the bond with a slight pressure to separate exerted by the thin mylar wedge, but not enough to break it upwards. Might take a while, but it should work.

the glass type is Corning glass 0211

Do you have access to and, a means to safely handle liquid mercury?

Does this occure with AlO3 ceramic?

If you are working on optics on a piece of glass.002" thick presumably you are using some precision tooling? Then possibly you have the means to grind a hole through the mounting plate to the under surface of your thin glass section, if you then apply a slight air pressure at the hole it should lift your thin section off. A piece of glass >002" thick should be flexible enough to allow a small bulge to start the separation process or alternatively try water for hydraulic separation.

Next time you mount your work piece always make sure you have an hole under the work.

Can you use an intermediate size piece of glass as a push-stick (hoping the thinner of the stuck pieces either slides or flexes enough to get lubrication in ) ? I have no idea if this will trash the stuck bits.

Can you try freezing it?? Or, place an ice cube on the samll piece, let it sit, and see if the cold shink the small piece enough to break the bind, and then try twisting the two pieces apart?? Just a thought.

MidniteFighter

Just a thought-I wonder if you could apply a magnetic repulsion to the two plates. Make both plates NORTH. Use the NORTH poles of 2 magnets?

How do you magnetise glass?

with a glass magnetizer. You show me magnetic glass and I will show you the magnetizer....you first...

I think mesmerize was intended.

Well , if we're going there , how about electro-static separation. A very big and opposing charge suddenly applied to each piece . If you fly a kite in stormy weather and touch one piece of glass with a free hand.........

Try touching a cube of ice onto the thin .

If no luck-try 200°C hot metal square cross section metal

I know a consultant you might want to get in touch with. His name is Whi Lee Coyote.

Don't try direct mail though !

I mentioned your problem to a friend last night and he immediately responded, "Soak it in 'Tween 20' surfactant". Apparently it is comprised of a very small molecule and penetrates easily with time. I have now told you more than I know about the subject, but he seemed convinced that it would work. Good luck. DickL

Better known as washing up liquid sort you use in a lab.

But more informatively expressed.

Yes ok I'll go with that.

Did you try to wedge a thin sharp blade to pry open?

Copy cat.

Glass cracks, finger gets cut, and tip of blade snaps off right into the eye!

It's all fun and games until someone...

........ does James Bond with a laser. Could be a real balls up.

For your eyes only or Gold b (sorry it must be finger).

Where does the balls fit in?

Definitely Connery

The balls wont fit anywhere after a close shave like that (unless he was a closet sumo wrestler)

What was the name of the one with the laser on the conveyor belt?

Dr No ?

Even Scaramanga did it to him on a beach !

Bad guys weapon of choice. Strange how vermin seems so savvy on them

Yes the thin glass just chips away

Enjoy some banter, but your not helping solve the problem.

I presume he has given up as he has not replied on this forum.

I think the banter occurs more when waiting for feedback from what the questioner has tried . Plenty of suggestions have been made . It would be kind of nice to know which has been tried out. I guess/hope that something is being tested out.

have not given up I still need to do this.

I had the assembly soak all weekend in isopropyl no luck

I'm going to look for some dry ice today

these bonds that have formed need to be broken chemically. You also need to protect the top surface with your optical pattern on it from any chemical etchant. An toluol soluble lacquer that does not cross link would do this. There are a number of etch resists you can use.

The etch resist can be paraffin wax dissolved in toluol or other solvent , which will resist all the etchants mentioned here and which can be rinsed off with toluol or other solvent later. bear in mind toluol is flammable and carcinogenic and so any personal contact via skin/lungs should be avoided by doing this in a fume hood.

As for the etchant. .01 potassium hydroxide is a mild base at that concentration it will be 100% ionized and will weaken these attachments....if you can get the solution in there. It seems to me that you will need to generate a cyclic flood and release to get fresh fluid into the space as the perimeter is the only apparent path into it. Can you see any sort of network of channels or other inhomogeneities in the bonded area that might be a channel you can flood? If that is the case, that gives you extra area to start the dissolution process.

There is another etchant you can try. This is ammonium fluoride or potassium fluoride. Bother of which will dissolve glass more aggressively than simple KOH, and should be tested on your test piece first. With the fluoride you MUST use an etch resist as it will make short work of your nice patterns on the top surface. I would start with quite a low concentration on the test piece first to see how it works.

Lots of stuff here about ammonium fluoride as etchant. Nice and toxic too. I can recall having some very aggravated finger nails from this as a dumb student.

http://www.google.com/search?hl=en&sa=X&oi=spell&resnum=0&ct=result&cd=1&q=%22ammonium+fluoride%22+%2Betching&spell=1

Keep at it mech eng , and good luck. It sounds like you're in the twilight zone of how a bond is defined . The solution you find could be of use to quite a few people .

What about a vacuum chamber, or its reverse, an autoclave? Lean the larger piece on something so it is sitting on a 45 degree angle with the small piece facing down so that if they do separate it will fall off.

True if it's about surrounding air pressure.

What if it's a chemical bond?

How do you slide a chemical solvent between the "sheets", not even knowing what it is that stuck the layers?

Someone near the start said

"

This is an optical contact bond - I believe it is an ionic bond at the molecular level. It is used in many different products where the bond needs to be vaccuum tight with a very long shelf life. It requires very flat surfaces and a close CTE match (which I assume these two glass pieces are the same material) and it lasts a very long time. Heat will actually cause the bond to strengthen, unless you can create a large temperature gradient between the parts (which I highly doubt since the glass is so thin). Depending on how deperate you become, I would suggest cooling the whole set and then trying a torch (in ~3 second dwells) isolated on the thin piece to see if you can create a high gradient. The risk is of course fracturing the glass due to the fast change in temp.""

That, and we don't know why the need to separate the two. If it's to expose an information surface such as photographic emulsion to a contact-press, than a simple optical system can reproduce by focusing the layer through the bonded layers.

If for something else, maybe another way found. We simply don't know. An optical bond is immensely strong, as it cures deep into the molecular level, and as the initial emulsion is transparent, the curing with a suitable wavelength (UV?) is very thorough.

When the cured substance remain transparent, it's a giveaway that it's bonding to the glass' surface is also thorough.

Very, very tough bonding and substance.

It's slight flexibility, will overcome the glass tendency to break due to it's rigidity and inner-tension properties.

Put a drop of liquid mercury at the edge.

RE:

Put a drop of liquid mercury at the edge.

??mechanism??

The glass floats on the mercury. Not sure about current practice but, glass used to be cast on mercury.

The problem here is the two glass flats have made a bond and the top glass is not free to float.

Someone mentioned that this was a common bonding procedure for two things that were flat to near atmoic level and the bond was made by van der waals forces or some such follwed by ion migration at stresssed areas to make a very strong bond that was not due to capillary action. Speculation are a large thermal stress from dry ice might cause a shearing of the bonds, others want to eth them in some manner. Machinist has a few test pieces to experiment with.

Interesting, as this thread play out a few questions do come to mind...

1. Why were these two pieces of glass put together in the first place?

2. Did you want them to bond, then decide something needed to be changed?

3. What is on the thin piece of glass?

4. Why can't the thin piece remain where it is?

5. Why should we insist on discovering a new way to dissolve an optically-cured chemical bond?

6. What's in it for us?

7. What's the hourly rate for some real sound advice these days?

8. Where is my lost dignity?

9. Are you listening?

8. You'll find it with your mother's shame.

9. Yes, for $145.00 an hour. So tell me about your dreams...

10. I'm in an empty forest, I mean forest with no trees, and it looks like a desert. It's night, and I can hear my mom calling:

"Yuval... Are you going to stay in the bathroom for long now?"

Then I wake up, wipe, and go to school.

Am I being paid 145 an hour for my story?

11. No , but that's what I'll make when I finnish cut'n'pasting you all !

No, but you owe me $145.00. And why do you spend so much time in the bathroom?

Ya know, we must be CR4's worst nightmare... Either that or they consider us as the carrion cleaners that chow down the remains of dead threads.

Fivestar vermin. This was your best yet.

I love that too. Is my seconding the admiration unnecessary bloat ?

Not if you really mean it.

Are you saying you don't get nasty fur-entaglement problems ?

I bet nutkin does

My prancing ability helps me - cats and tribbles can't do that.

A bear and a rabbit are next to each other in the woods.

Bear: don't you mind stuff getting stuck to you ?

Rabbit : no

Bear - picks up rabbit and...

Making up stories.

5. I'm an often contributor to the discussions and was hoping some one had general knowledge on the subject. If this becomes more than that let's all get paid for every entry.

6. What's in it for us? Why does anyone try to help someone else any way?

7. The last time I checked it depend on how badly someone needs to know and how much is worth to the buyer.

8. Sorry if you lost your dignity you have to find it yourself.

9. Yes, but I am about to put a ear plug in 1 ear.

Opps I wasn't logged on I was contributor 85 & 86 as guest

they had bought a costly optical ~interference filter on the thin glass. They cleaned it and inadvertently left one on another piece of similar very flat glass....they bonded.

Now they either buy another or fix this stuck problem. They have made some test pieces to try methods on so they brewak no costly optics.

1. the processing requires more stiffness than the .002" substrate has and the substrates can not handled by the existing equipment at the fabricators

2. they are mounted on standard wafers for ease of handling and to use existing equipment.

3 small optical filters

4. no the thick substrates do not work optically.

Opps I wasn't logged on I was contributor 85 & 86 as guest

Guest , your points seem to be consistent with the original question . If you are one and the same , it sounds as though the question should have related more to how can I stiffen the thin piece temporarily for processing . If the problem presented is not solvable (which seems likely) , it would be better to restate the situation clearly so you can find processing solutions rather than have a whole bunch of folk discussing a situation that could be avoided.Fixing something that's broken is fine (up to a point - which the thread seems to have reached ) , but avoiding the problem in the first place may be a better line of inquiry. Fixing one off mistakes is often taken lightly by people , but most people here see the seriousness and value of finding solutions for process engineering.

That's my conclusion re your earlier irritation , and suggestion if you are MechEng. I hope this can maybe help to re-focus on addressing the issue raised.

Only just saw your post MechEng (I was in Editor). Kris

The process handling / removal issues are being addressed for future runs at our supplier / vendor.

The Samples I have and would like to remove were test samples because earlier runs resulted in poor optical quality of the elements. Some samples were run to compare the affects of a bonding agent that was easy to release.

This lot without the releasable bonding agent were Very good but they are very stuck on to the carrier waffers. The problem is being worked by our supplier.

I was hoping for some suggestion outside of my normal experience to gain some time and test our system before more samples could be made (about a 3 month lead time).

Your strategy makes sense , I think we'd all try the same. Maybe some of the info here can help your supplier - they must be having a similar problem finding the answer . When I suggested temporary stiffness , I had something akin to a window frame in mind (the lesser contact area would make removal of the thin slice easier ). I'm not sure if this is practicable to the handling process , but you never know what far-out ideas may help (this isn't my field , but the problem is interesting). Best of luck to you

If absolutely (and I mean absolutely) nothing else works, you might try what diamond cutters do. A large thin blade (covers the entire area of the joint), set between the pieces and a quick, well placed, whack with a hammer. With diamond, this cleanly cleaves ionic bonds. The result, two separate diamonds. But we're talking desperation here.

The comparison is not quite fair. Diamond cutters may spend months/years before making the all important hit. Diamond are riddled with flaws in the natural state. Stereographic projection methods were often used to map the flaws/planes etc to aid visualization and solve 3D geometry problems. The method is great for any 3D geometry solutions if your drawing skill is better than your maths.

The drastic 'hit' solution may be as good as any - perhaps damage risk could be minimized by use of a lubricated load on the glass (to help resist buckling ) and a thin metal striking tool.

This is a little out of my field, but I wonder if you could take two pieces of metal, polish the surfaces, machine a .002" cavity in one the same shape as the thin piece, a .040" cavity in the other matching the shape of the thick piece. pressure could be applied to clamp the two pieces to keep the glass from breaking. At this point one could apply as much shear pressure as necessary to shear the bonds that have formed. It seems like kind of a brute force method but it is all I can come up with.

But how do you get the glass out of the machined metal ?