Thermoplastic or Thermoset? The Real Cost of a Plastic Mug Mistake

In March 2024, I got a call at 4:30 PM on a Friday. A customer had 10,000 plastic mugs sitting in a warehouse—the kind with a soft-touch TPE grip over a polypropylene body. Forty percent had warped during the pre-production run. The line was booked for Monday morning. Normal troubleshooting takes five to seven days. We had about 36 hours.

These weren't cheap giveaway cups. They were retail products with a custom lid, a soft-touch surface, and the customer's brand molded into the base. The packaging alone cost $18,000. A missed ship date meant losing a grocery chain's shelf placement.

"It's the resin," the customer said. "The material must be bad."

In my role coordinating urgent material replacements, I've handled maybe 200 rush orders in 12 years. Maybe 180—I'd have to check the system. Doesn't matter. The pattern is the same. When a part fails, the material gets blamed first. But when I'm triaging a rush order, I ignore the accusation and look at the part.

The material wasn't the problem. What I mean is: the material was doing exactly what it was specified to do. The specification was the problem.

Everyone Jumps to "Bad Material"

When parts fail, the instinct is to blame the resin. "Cheap polypropylene." "Bad batch." "Not food grade." It's a satisfying explanation because it assigns blame somewhere else. But in my experience—and our internal numbers from 200+ rush jobs back this up—most failures trace back to an incomplete spec, not a material defect.

This customer sourced the PP through a broker at nearly 30% below the next quote. On paper, a win. The spec sheet said: "PP, food contact grade."

That's not a spec. That's a category. The problems began there.

Thermoplastic or Thermoset: The Fork in the Road

Let me back up. About half the time I ask a client what material they're using, I get a one-word answer: "Plastic."

As if that meant anything.

The first distinction you need to make is thermoplastic or thermoset. Polypropylene is a thermoplastic: it melts when heated, re-solidifies when cooled, and can be reprocessed. A thermoset cures—once set, it won't melt again. Think of an egg after frying. No going back.

That distinction matters beyond the obvious. Thermoplastics can be reground and reprocessed, which is worth a lot when your scrap rate hits double digits. A thermoset that fails is just landfill. For a mug that'll see hot water, microwaves, and hundreds of dishwashing cycles, you want a thermoplastic with the right heat deflection temperature and creep resistance. Polypropylene is a reasonable starting point. But then comes the question almost nobody asks: which polypropylene?

What Is PP Plastic Material? (It's Not One Thing)

People search "what is PP plastic material" because they want a simple answer. Here's the honest one: PP is a family.

There are at least three major branches:

• Homopolymer PP—stiff, good chemical resistance, more brittle at low temperature
• Random copolymer PP—lower melting point, better clarity, more flexible, common in food-contact products
• Impact copolymer PP—rubber-modified for toughness, less rigid

Then there's melt flow index (MFI). A grade with MFI 8 flows slowly and fills a mold deliberately. A grade with MFI 30 flows like water. According to ASTM D4101, polypropylene grades are classified by property ranges, not by name alone (Source: ASTM International, astm.org). And per ISO 1133, MFI is measured in grams per 10 minutes under defined conditions (Source: ISO, iso.org). The designation "PP" without a grade number is nearly meaningless to a molder.

The customer's PP was a random copolymer with a higher MFI than the part design expected. The thin walls filled fine. The thicker base sections cooled at a different rate. Result: internal stress, warpage, and a TPE grip that pulled loose because the substrate underneath was still moving.

TPE overmolding adds another layer. The bond between a TPE grip and a PP substrate isn't just chemical—it's thermal, mechanical, and process-dependent. Some PP grades are far more TPE-friendly than others. This customer's chosen PP was on the questionable end of that spectrum.

In an ideal world, every part gets validated with prototypes in the actual production material. In reality, most molding grades are chosen by habit or price. The broker knew the buyer needed a number on the PO, not an education on polymer physics.

That's the deep cause. Not "bad resin." A procurement decision based on a category instead of a properly specified material.

The Colorant That Cost $7,000

A few weeks later, the same customer called with a new issue. Mugs were dimensionally stable now, but the color was blotchy. Flow lines around the grip, visible under certain lighting.

The molder recommended a generic colorant. "Same pigment loading, a fraction of the price." The numbers looked fine. My gut said the generic product hadn't been tested in this particular PP. I let the customer make the call anyway.

Here's what I should have explained more clearly: colorants aren't cosmetic. They're part of the material system. Some pigments act as nucleating agents, changing how PP crystallizes as it cools. Different crystallization means different shrinkage. Different shrinkage means warpage, flow lines, and parts out of tolerance.

The flow lines weren't just ugly. On a mug, you're holding it, your hands feel every ridge. It reads as "cheap" to the end user.

The customer saved about $200 on that generic colorant. The scrap, rework, and expedited replacement resin cost close to $7,000 all-in. An Avient colorant, tested in the actual base resin, would have added maybe $300 to the original order. Total cost of that decision: $7,000.

I still kick myself for not pushing back harder. If I'd laid out the crystallization mechanics beforehand, they'd have made a different call. Instead, they learned—and I relearned—that unit price is the cheapest way to measure a material.

What This Actually Costs: TCO, Not Unit Price

Let me put some numbers on this, because this is the part that matters for your purchasing decisions.

Total cost of ownership includes more than price per kilogram:

• Processing performance: scrap rate, cycle time, energy per part
• Colorant and additive compatibility with the actual base resin
• Cost of field failures: returns, warranty claims, lost retail placement
• Rush shipping and overtime when something goes wrong

The properly specified PP cost about $0.05 more per part. On 10,000 parts: $500. The generic colorant saved $200. The failure cost $7,000.

The $7,000 wasn't a single bill. It was a thousand here, two thousand there. $620 for expedited resin delivery. $1,800 for the weekend molding line. $340 for a courier to the distribution center. $2,100 in scrapped parts. The rest disappeared into overtime, meetings, and stress nobody billed for.

The math isn't even close. If you ask me, that's not procurement. That's gambling.

TCO feels abstract because it includes costs that didn't happen yet. That's the trick of it. The $7,000 failure is a line item. The $500 you didn't spend feels like a win. Until it isn't.

The Fix: Four Questions Before You Order

By the time we solved this, we were hours away from the rescheduled run. The short-term fix was straightforward: a correctly specified PP, verified TPE overmolding process, and a colorant tested in the base resin.

The TPE portion of the product—the reason an Avient thermoplastic elastomer distributor got involved in the first place—was actually fine. The material was a good match for PP substrate adhesion. But when the substrate warped and shifted, no adhesive system on earth could hold the grip in place.

The long-term fix is asking four questions before any material PO goes out:

1. Thermoplastic or thermoset? Know which you need before you talk to any supplier.
2. Which PP, exactly? Homopolymer, random copolymer, or impact copolymer? What MFI range? Put it in writing.
3. Is the colorant tested in your base polymer? If not, assume it's a variable.
4. What's the total cost of failure? Not the unit cost of success—the all-in cost of getting it wrong.

That's the whole framework. Not complicated. But it requires a supplier who understands both polymer chemistry and production reality. We see these patterns constantly—TPE overmolding failures, colorant incompatibilities, vague PP specs. The fix is the same every time: slow down on the front end, or pay a lot more on the back end.

These questions don't take a materials science degree to answer. They take ten minutes with someone who knows materials. A supplier who can't answer them—that's useful information too.

The mugs shipped. The retailer kept the shelf placement. The customer now does TCO analysis before every material decision. They haven't called with a similar emergency since.

Not every material problem becomes an emergency. But the ones that do all start the same way: with a category on a spec sheet instead of a material.