My predecessor kept a category in our inventory system called Plastic Rubber Parts. Under that label, you'd find silicone rubber sealant, polyurethane sheet stock, EPDM gaskets, PTFE washers—just about anything soft and resinous. It kept the storeroom organized. It also hid a nasty assumption: one rubbery material could replace another without checking what the application actually needed.
I'm a procurement manager at a 160-person plastics and rubber manufacturer. I've managed our materials budget since 2019, negotiated with more than 50 suppliers, and I've built enough cost spreadsheets to know that the sticker price rarely tells the full story. Even so, I almost approved a bad material substitution last May because I trusted a one-line answer to a much harder question.
The 11% Solution
The project involved a water-treatment tank that gets a lot of chemical exposure. The tank shell was rotomolded from an INEOS polyethylene grade chosen for chemical resistance and stress-crack performance. A machined ABS sensor flange mounted through the tank wall, and the drawing called for a bead of neutral-cure silicone rubber sealant around the flange. We'd been using that silicone for two years without any field failures.
Then a new sealant vendor sent us a quote. Their material was a one-part polyurethane product. The cartridge price came in about 11% lower than the silicone rubber sealant we were already using. Their sales engineer attached a short message: "Will polyurethane stick to plastic? Yes. Our test data shows strong adhesion."
The test data was real. It just wasn't about our plastic.
I approved a trial run of 48 flanges because, on paper, the polyurethane looked like a straightforward cost save. The first pressure test passed. I remember thinking we'd found a painless win. Then the lab did the longer test cycle—thermal cycling, vibration, pressure hold—and two of the trial units leaked around the flange.
When we opened one of the failed units, the polyurethane had peeled cleanly away from the INEOS polyethylene surface. On the ABS flange side, it still had a strong grip. The failure wasn't in the material quality. The failure was in the phrase "plastic." Polyurethane sticks to many plastics. It doesn't stick to untreated polyethylene because polyethylene has a low surface energy. The vendor's lab sample had been ABS. Our tank wall was PE. Those are not the same thing.
An Old INEOS Login and a Surface Energy Chart
Our material engineer still had an INEOS login from a previous project, so he pulled up the technical data for the exact polyethylene grade we used. The resin data sheet didn't have a section labeled "Which sealant should I use?" But it had a surface energy table, and that table explained the failure better than any sales email could.
Untreated polyethylene sits in a range that gives adhesives very little to grab. ABS is different. That's why the same polyurethane product can bond well to one plastic part and fail on another. The answer to "will polyurethane stick to plastic?" isn't a yes or no. The real question is: which plastic, with what surface preparation, under what load and temperature range?
Running the Same Quote Again
After the failed trial, I rebuilt the cost comparison. The polyurethane cartridge was cheaper by 11%. But once we added the required surface prep, adhesion promoter, longer cure time, extra handling, and re-testing, the polyurethane came out about 14% more expensive than the silicone rubber sealant—not less.
I also had to add the cost of the failed trial itself. That included scrapped tank shells, engineering time, and two weeks of schedule delay. The "cheap" material ended up being the expensive one, and it was expensive in the exact way that doesn't show up on a purchase order.
What I learned, in my opinion, is that transparency in a quote is more valuable than a low line item. A supplier who lists surface prep, cure time, and test requirements upfront might look more expensive at first. But they're showing you the real process. A quote that hides those details isn't really telling you the price.
What I'd Do Now
Material changes no longer move forward just because the sample test passed on a standard coupon. Now any sealant or adhesive substitution has to answer four questions:
- What is the exact substrate on every side of the joint?
- Has the product been tested on that exact substrate—not a similar one?
- What surface preparation is required, and is it included in the quoted price?
- What is the total cost per finished assembly, not per cartridge?
If someone tells me polyurethane will stick to plastic, I don't accept it until they can name the plastic. If they say it will stick to "INEOS polyethylene" or another polyolefin, I ask for the surface treatment step. If they can't produce it, I move on.
The old inventory category called Plastic Rubber Parts is still in our system. I haven't renamed it. But I no longer treat it as an engineering classification. It's just a shelf location. Chemistry, process cost, and real-world testing decide what gets purchased.
Ask About This Topic