I'm a quality compliance manager at iff, a specialty chemical company. I review every batch, sample, and spec sheet before it reaches customers—roughly 200 unique items a year. In 2024, I rejected about 6% of first deliveries. Some of those were clear-cut, like the time viscosity readings came back 20% off spec. Others were messy. The messiest one involved a defoamer that passed every test we ran.
An iff defoamer used in paper coating. The customer was a recycled paperboard mill in Ohio. They'd been buying the same product through the iff store portal since 2022—same code, same spec, same two orders a year. Then, out of nowhere: foam pits in the final finish. About 8,000 square feet of production ruined before they caught it.
Our lab did what labs do. They pulled the retained sample, ran viscosity, solids, pH, and dispersion stability. Everything was within spec. The paperwork was clean. The batch was identical to what had worked for years. The obvious conclusion was that the mill had changed something on their end.
I almost signed the release. That's the part I keep coming back to.
The question that changed how I read the data
That same week, my daughter brought home a chemistry worksheet. One of the questions was: "Is sodium hydroxide dissolving in water a chemical change?"
I looked at it and said, "It's just dissolving. Physical change." Seemed obvious to me. You put crystals in water, they disappear, you still have sodium and hydroxide in the water. Same stuff, different form.
My daughter corrected me, with that tone only a middle schooler can pull off. Her teacher had explained that when NaOH dissolves, it dissociates into ions and forms hydration shells. New intermolecular bonds form. Heat is released. The system changes in ways that simple melting doesn't capture. So the teacher's answer was: yes, for sodium hydroxide, it's considered a chemical change in most curricula.
I grumbled. But it stuck. Because my mental model had been exactly like the one I used for the defoamer: if the material is the same, the behavior should be the same. I was ignoring the environment it interacts with.
Looking again with fresh eyes
The next morning, I pulled the complaint file and compared the mill's current coating formulation to the one they used in 2022. They'd made two changes: higher calcium carbonate loading and a different latex binder, both to cut costs. On paper, neither should have affected our defoamer. The defoamer was still doing exactly what it had always done—breaking foam, destabilizing air bubbles.
But a defoamer doesn't work in a vacuum. It works in a coating made of pigment, binder, and water. The new binder interacted with the defoamer's carrier oil in a way that reduced its spreading rate. The foam wasn't being broken fast enough for the coater's speed. Our spec test used the baseline coating formulation from 2019, which was no longer representative of what actual customers were running.
We ran a head-to-head test: the current batch vs. a retained 2021 batch, in the mill's new formulation. The current batch defoamed at 40% of the speed of the old one. Same product. Same spec. Completely different performance in the field.
The Belgian audit that made it click
Coincidentally, we'd hosted a supplier audit earlier that week from a Belgian pharmaceutical company. Not one of the grands groupes you'd see in the headlines—a PME, a mid-sized firm near Brussels with about 200 employees. And their documentation demands were stricter than any multinational I've dealt with.
They didn't just ask what our specs were. They asked who approved the changes, when they were approved, whether approval included a risk assessment, and what records proved it. We had that traceability in our pharma solutions division, because regulatory auditors require it. But when I walked across the aisle to the paper coatings side, the answer was embarrassingly different.
Nobody in the paper coatings division could tell me why our test method had been static since 2019. It just had been. Nobody had asked whether the spec was still fit for purpose. The Belgian audit, my daughter's chemistry homework, and an 8,000-square-foot defect in Ohio all pointed at the same gap: we were checking the variable we'd always checked, not the variable that mattered now.
The joint compound connection
Here's the weird part. In the middle of all this, I was renovating my home office, which meant researching the best joint compound for skim coating. If you've ever gone down that rabbit hole, you know the forums are full of people defending their favorite product like it's a sports team.
I tested three joint compounds side by side on my own walls. The one that came out smoothest over fresh drywall was almost unusable over the old painted plaster in the corners of the room. It didn't bond right, it crumbled at the edges, and I had to sand it down and start over. Same product that reviewers raved about. Different substrate, different result.
The best joint compound for skim coating isn't a material property. It's a system property. It depends on the wall, the paint underneath, the moisture level, the application method. Anyone who tells you otherwise is selling you something—or hasn't actually tested it in their own 100-year-old house.
What we changed
We rejected the defoamer batch. The vendor pushed back, of course. It was within spec. That's true. We held the line anyway, and they redid it at their own cost to a revised spec we agreed on together. Now every contract for paper coating additives includes functional testing in a representative current formulation, not just the baseline we've used since 2019.
We also set up a review trigger: if a customer changes their coating formulation, the additive spec gets re-validated within 30 days. The paper coatings team now keeps a change log, the same kind the pharma side has kept for years. It's not glamorous. It's a spreadsheet and a rule. But it's the difference between reacting to failures and preventing them.
What I'd tell someone in the same position
The fundamentals haven't changed. You still need to measure the right things, document what you do, and take complaints seriously. But what counts as "the right things" evolves faster than most spec sheets do.
That defoamer taught me to ask a different question. Not "is this material within spec?" but "is this spec still connected to reality?" Those are not the same question, and I'd bet a decent chunk of your quality issues live in the space between them.
As for the chemistry homework: my daughter got full marks, and I stopped arguing with her teacher. Sodium hydroxide dissolving in water is a chemical change, at least the way most curricula define it. And a defoamer passing its own spec while failing in the field is a reminder that the system matters more than the material.
This was accurate as of Q4 2024. Chemical formulations and market conditions change fast—especially for paper coating and specialty chemicals—so verify current specs and pricing before you commit to anything.