Bonding FFKM & PCTFE at temperature in solvent atmosphere 1

[KENAT]

Bonding FFKM & PCTFE at temperature in solvent atmosphere

I’m involved in a project where we need to bond plastics like FFKM (Kalrez) & PCTFE (Kel-f) in a device that in service will see saturated solvent atmosphere ( DMSO, Acetone, Acetonitrile, toluene, dichloromethane among the most significant) to 100°C, or with less aggressive environment will see close to 250°C.

Previous research by a colleague (including some testing) for a slightly different application suggests that Torr-Seal has the best resistance to the solvents (out of adhesives he looked at) but possibly doesn’t bond the FFKM & PCTFE materials well and has a temperature limit of 120°C.

We typically use Chemgrip for bonding plastics like these but it does not stand up well to the solvents. We typically use MASTERBOND 10HT for high temp, but again not great with some chemicals.

I’m doing some digging and came across MasterBond EP45HT which claims solvent compatibility & good temperature performance does anyone have any experience using this

http://www.masterbond.com/tds/ep45ht?

(I’ll try contacting them but apparently my colleague had trouble getting information from them previously and had concerns about their claims for some other formulations so I thought I’d see if anyone here has used it.)

In my colleagues research he also came across D.E.N. 438 Epoxy Novolac Resin but didn’t get as far as testing it, and had concerns about it being tricky to work with in production.

A related formula D.E.N. 431 looks even better.

Another company they found was Arcor especially EE-101

Quick Google search turned up Loctite 325

If any one has any experience with these or suggestions of alternatives I’d appreciate it.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[KENAT]

Got some additional info from my marketing colleague. The 100°C with solvents is misleading. While some components/glue joints in my assembly would be heated to 100°C the saturated solvent vapor would be nominally room temperature (though obviously may warm up some around the heated components).

Any suggestions appreciated, thanks.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

Ick, you sure get the tough ones, Kenat. No real advice here, other than I doubt you will find much useful information from any method other than testing.

I would have my doubts about the loctite acrylic adhesive, but have not checked your list of solvents to know for sure.

The biggest problems with any adhesive is the surface energies of the parts you are trying to get to adhere, both of which are some of the toughest to make work. The typical method is to etch the surfaces with things like ion beams and hydrofluoric acid, but if these are small parts, it can be tough to control where the etch is being done, and avoid damaging the non-bonding surfaces of the parts to the point that their chemical stability is compromised. As I'm sure you know.

FFKM is an elastomer, to be picky, but my point is that it is heat cured/vulcanized. This opens an opportunity to have it over-molded to the PCTFE part, with a suitable bonding primer/etch process done to the pctfe to enhance bonding. The process may not be feasible for your part, but it will give you the best results if you can make it work, IMO. Lord chemicals and molding shops with experience in FFKM elastomer molding will be your best friends here. I could probably point you to some people that could help, if you want a name, but I'm also fairly certain, given what I know of your location, that you know those Green WoolenFabric people quite well.

Good luck and don't sniff any open bottles.

 

[KENAT]

Thanks Btrue, a colleague of mine has been fine tuning some relatively precision sodium etching techniques for a similar application, so hopefully we can learn from that.

We will be doing some testing, just deciding what to test is a good first step though.

I'll bring up the overmolding issue with the guy doing the detail design but I have a feeling it's a no go as the FFKM part is actually a bellows. I'm afraid I'm being dumb and didn't get your reference to the green woolen fabric folk, if you could be more obvious I'd appreciate it.

I looked again at some of the glues I listed, the Loctite is a no go on the temp range anyway (I was mixing up °C & F) and the Arcor EE-101 probably isn't good enough on temp either.

I'm waiting to hear back from Masterbond.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[dhengr]

Kenat:
Could you mold or form a lip on one of the parts which acted as a sealing lip around the glue/bond surface? Thus keeping the solvents away from the bond surface and the bonding agents. This lip might fit in a groove in the other part, and this groove could be filled with some material which did bond minimally and seal, but tolerated the solvents better, and was not the primary bonding agent, when the two parts were pressed together to form the primary bond. Alternatively, if you had a rubber (maybe not the best mat’l., but flexible enough) strip with a round hollow tube at each edge; the edges of this trip would be presses into a T-slot groove in each part and be held there by the slot, spanning the bond joint and keeping the solvents away. Maybe these ideas are too complicated if you can find a bonding agent which will do the bonding and tolerate the solvents.

 

[KENAT]

Doh, I forgot to mention that to the guy doing the detail work when we were talking this afternoon.

I have a feeling that the space limitations and bellows complexity etc. may rule that out but I like the idea - like btrues 'molding over' idea.

An a similar product we were able to protect most of the adhesive seals using o-rings or gaskets etc. but we're having trouble doing that on this one.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[TVP]

Kenat,

I would talk to Lord about your application. They have a lot of experience with bonding to elastomers, and their primers have to withstand subsequent moldbonding temperatures.

 

[KENAT]

Lord Chemicals - got it now Btrue & TVP.

For some reason my initial search after Btrue's suggestion didn't get good results - user error I'm sure.

Thanks.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MiketheEngineer]

Good luck - bolt them together??

 

[KENAT]

Maybe I should have given some idea of scale.

This is for use on a special type of microsocope as a mini environmental chamber for samples. One end of the kalrez bellows is attatched to a kel-f part maybe an inch or so in dia, maybe .125" thick which has a bunch other stuff allready integrated in it and whose dynamics are fairly sensitive. One of the items mounted on this kel-f part will need to be heated to close to 250°C.

The other end has a bit more room and less stuff cluttering it and is further away from its heat source but still pretty cramped, especially vertically.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

Sorry, Kenat. Green Tweed (

http://www.gtweed.com)

is who our oil patch sister company works with for high-end (KelF, Chemraz, etc.) rubber seals and molded parts.

I figured you were talking small.

If the bellows is being molded now, it may still be possible to overmold/bond to your part; whenever you are talking elastomers it's really the best approach for strength and chem resistance, i think. Might be worth talking to the people molding the bellows to see if its feasible, and you would probably need to be willing to pay for modifying or making a new mold/tooling. If the mfgr. of the bellows doesn't want to do it (part count just too low, and they likely don't want to be a custom molding shop) then still consider finding a custom shop, and working with their mold/tooling guys to figure out a way. I can think of a couple of ideas off the top of my head that might work, but I'd need to talk to somebody who knows about molding/bonding FFKM to know for sure.

 

[KENAT]

I'd love to share a cutaway picture that gives some idea of what's going on but we're pretty IP sensitive.

Our purchasing/accounting also don't like adding new vendors - we have to make a pretty compelling case. However, I'll let the guy doing the design know about green tweed thanks.

As you guessed we're talking pretty low volumes - low double digits per year probably.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

Hmm. Not sure if it helps, but I posted my contact info to a forum/FAQ area I know you frequent. A sketch would help, and I could share back some ideas for molding tooling that might help...but if we need NDA's and a team of lawyers maybe it's best left at this level.

In general terms, if a bellows has narrow enough convolutions, the core used to form the internal hollow can be a solid piece, and the bellows is stripped off of the internal core after molding by peeling/rolling it back or by injecting/blowing air under the edge. So, tooling can be pretty simple for a mold like that.

Adhesives - I'd get samples in of as many different breeds and varieties (epoxies like you are looking at would be my first choice too) and come up with some kind of simple test fixture that could be pulled/peeled or given the kind of stresses you think the bond will need to survive...then make a bunch of samples with different adhesives, soak them in your witch's brew of solvents, and go pull/peel/bend/fatigue them to failure and see what you learn. Assuming, of course, infinite time money and help, you should easily reach a solution...

 

[KENAT]

Thanks btrue, while I have no doubts of your trustworthiness I think I probably need to be a bit careful on this one. Plus the guy doing the work is a contractor (who used to be a direct super senior/principle engineer) and can be a little tricky to direct (I really have to use my apparent people skills!;-)) and doesn't always take suggestions well.

For 'rigid' materials another colleague developed a standard sample shape which is basically a small 'dish' with some holes in the bottom. Idea is that you plug the holes with the adhesive of interest, then fill the dish with solvent and place on top of blotting paper or similar indicator. Then leave for X hours/days... and come back to check for leaks.

I have some concerns about the solvent evaporating - especially if forced to some temperature - though maybe he tops them up or something.

Our loads are relatively low so for our applications the 'sealing' aspect of the adhesive is more significant than the 'strength' aspect - hence this test set up. Plus this set up maximizes the area open to solvent attack.

I may have an intern to help with this testing - I checked and he doesn't have a history of huffing.;-) I already suggested he start thinking about a test for the kalrez parts that aren't as well suited to the same dish test.

Thanks again.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

"no doubts of your trustworthiness "

Guffaw. No worries, Kenat.

 

[KENAT]

I meant trustworthiness regarding IP, not kegs of Guinness or similar;-).

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[moltenmetal]

I think that talking to THE LORD might be necessary for this one...gluing together non-stick surfaces is pretty tough!

I saw a presentation about 25 years ago from a guy who was doing extremely violent surface treatments to fluoropolymers to make them susceptible to bonding. He was using very, very nasty reagents that made sodium look about as tame as water, but they were very effective at modifying the surface and didn't do anything beyond the first few molecular layers of the polymer. Bulk properties remained intact. He was a uni professor and I don't know if his work went anywhere, so I know it's not much good to you, but there it is FYI.

 

[btrueblood]

Kenat, you are right not to trust me with any spare kegs of Guinness you have lying around.

Molten, yes, hydrofluoric acid etch on pctfe is what I remember too, for bonding with an epoxy.

Yghr...sometimes it takes awhile for the memory cells to wake up, in this case 2 days after the OP prompt. Well, not bad for something I learned 20+ years ago. Kenat, the epoxy we used (epoxy potting/bonding/strain relief of pctfe jacketed leadwire) was a product from the National Starch Company. Google turns up that the adhesive product brands for that company were purchased by Henckel, which is also loctite's parent now I think. The brand name was Ablestik, which is still available on the Henckel website if you dig for it. The stuff would stick darn near anything to anything else, polyimide films, pctfe films, etc. Like an epoxy/nonwoven glass pre-preg, that could be cut to shapes and molded, then clamped and heat cured (200 F or thereabouts). Don't know much about solvent resistance, but it should be pretty good, as it's an epoxy base.

 

[KENAT]

I had a look at Henkel's site and found the Ablestik & nothing explicitly jumped out at me - I did forward to my contractor colleague though to look at.

I heard back from Masterbond and they sent me the data sheet for ep125 which looks promising. Not sure it's going to meet DFMA goals as it's a paste with fairly high & long cure/post cure but hey, you can't have everything;-).

http://www.masterbond.com/tds/ep125

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[KENAT]

Doh,

Having finally made some traction on the high temp/thermally stable adhesive front, the colleague that identified Tor seal for an earlier application has just pointed out that PCTFE is only good up to about 120-130°C and may show dimensional instability below that!

I'm amazed I haven't had a panicked phone call from my consultant!

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​