Does Epoxy Resin Expand When Drying? A Quality Manager's Take on Gold Acid Tests, Medical Tape, and Total Cost

First, the short answer: epoxy resin doesn't expand when drying

I work in quality at a coatings and specialty chemicals company. The PPG name is the one on the arena in Pittsburgh, if that helps orient you. So if you're reading this while checking hotels near PPG Paints Arena in Pittsburgh, PA, or skimming it on your phone before a Disturbed show at PPG Paints Arena, this page probably isn't what you meant to open. I don't handle hotel lists or concert schedules. What I do handle is questions like the one this article answers: does epoxy resin expand when drying?

Short answer: in ordinary use, no. Epoxy resin doesn't dry like latex paint dries; a two-part system cures by crosslinking. Solvent-based epoxies lose solvent and pull together slightly. Solvent-free systems shrink very little. If you're seeing bubbling, blistering, or lifting, the usual explanation isn't resin expansion while it dries—it's trapped air, a damp substrate, or a thick mass that overheated and produced exotherm bubbles. Those are different failure modes with different fixes.

That's a clean short answer. It still doesn't solve the problem, because the real problem isn't expansion. It's how quickly we simplify a complex material decision into a single yes/no question.

The question under the question

When a customer sends me an email asking whether epoxy expands, there's usually a damaged floor or a warranty claim behind it. They want one cause they can circle. I understand, because I did the same thing early in my career. I've spent over four years in quality, reviewing roughly 200 unique incoming lots, batch records, and supplier audits a year. In my first year, I approved a material lot because the certificate matched our one-line spec. I didn't ask how the vendor tested it or whether the test matched our application. The answer turned out to be 'not really.' The lot failed at the line, and I learned to ask a different set of questions.

I see that same shortcut show up in three places over and over: acid testing of gold, medical adhesive tape, and coating failures. Different industries, same mental move.

Gold acid test: color reactions are clues, not conclusions

I'm not a metallurgist. When we needed to qualify plated parts for one of our processes, a chemist walked me through gold acid test color reactions with nitric acid on a touchstone. The colors are genuinely useful. Copper-heavy alloys usually show a green or blue-green edge as copper dissolves. Silver alloys can leave a pale, milky-looking film. A streak that stays unchanged tells you the metal resisted that particular acid strength.

Here's the trap: a reaction is not a verdict. One scratch samples one tiny point on one part. A plated surface can pass while the base metal underneath is something else. If the acid strength doesn't match the alloy you think you're testing, you'll misread a perfectly good reaction. The test is a screen. When the part is expensive or safety-relevant, confirm with a method that looks past the surface.

I've started calling this the 'one-drop' problem in my own field. A quick rub test on a coating sample is fine for screening. It is not fine as the sole basis for accepting 10,000 units. The sampling plan is part of the test, and people forget that half.

Medical adhesive tape: 'medical grade' tells you less than you think

Medical adhesive tape is the second place the same pattern shows up. I'm not a clinician, so I'll stay on the industrial side. In the chemical industry, we supply materials used by medical tape converters, and we get pulled into their quality investigations. From that seat, 'medical grade' has become nearly meaningless as a specification.

It doesn't tell you what failure mode matters. Is it initial tack? Peel strength after sterilization? Removal force, aging behavior, residue, compatibility with the sterile pouch? A tape that performs well for one application can be completely wrong for another. The polymer, the adhesive coat weight, and the crosslinking are different, even if both products are called 'medical tape.'

In Q1 2024, a tape converter asked us to help investigate 8,000 finished units stored for three weeks after sterilization. The tape looked fine. The supplier's data were technically within tolerance on its standard test panel. But the peel strength on the actual product, after the actual sterilization cycle and storage period, had dropped below the agreed limit. The converter had to quarantine the units and rework them. The redo cost about $22,000 and delayed a launch.

Put another way: the supplier was right, and the product was still wrong. Nobody specified what the tape had to survive.

The common thread, and the total cost nobody quotes

Epoxy expansion, gold acid tests, adhesive tape—the chemistry is unrelated. The decision pattern isn't. Each example starts with a useful question and collapses it into a simpler one. Does it pass? Is it within spec? Does the resin expand when drying? The simpler question is easier to type into an email. It also ignores the conditions that decide whether a material actually works.

The result is what I call the hidden invoice. Unit price is the number at the top of the quote; it is not the cost of ownership. The cost of ownership includes freight, incoming verification, line stoppages, rework, expedited shipping, complaint investigations, and the hours your team spends assigning blame. In the tape case above, the savings from choosing a lower-cost material disappeared by the time the quarantine started.

I made this mistake myself before building our verification protocol in 2022. Now I ask every supplier for the test method before I ask for the pass number, and I require a confirmation test on our actual application before committing to a large order. It feels bureaucratic right up until the moment it saves a launch.

What I would write into a specification instead

If you want to avoid the same invoice, stop specifying by product name and price. Specify the condition the material has to survive.

Instead of 'epoxy floor coating,' write 'epoxy coating applied over primed concrete at no more than 4% moisture, verified by adhesion testing on the actual substrate after seven days.' Instead of 'medical grade tape,' write 'peel adhesion after sterilization and 30 days of storage, tested on the actual substrate per ASTM D3330.' Instead of comparing quote prices, compare total cost of ownership: material plus freight plus verification plus rework risk. Make those tests part of the purchase order, not part of a post-failure conversation.

So, does epoxy resin expand when drying? For most normal coating jobs, no. But if you have a failure, stop measuring the resin and start measuring the moisture, temperature, and thickness around it. Use acid tests as screens, not verdicts. Write specifications that describe what the product has to survive. The material is rarely the whole story. The vague sentence in the specification usually is.