Brand Logo Contact Us

News detail

7 Checks Before You Sign Off an Adhesive Batch: A Quality Reviewer's Checklist

When this checklist applies

If you're specifying or receiving adhesives on an industrial line — tiling adhesives on a construction project, fast-curing epoxy adhesives on an assembly bench, cyanoacrylate on a high-throughput packaging cell — this is the checklist I run before I sign off on any batch. Seven steps, each with a specific checkpoint you can actually verify.

I've been doing quality and brand compliance review at a specialty materials company for a little over four years. I clear roughly 200 unique batches a year. In our Q1 2024 audit, we rejected about 12% of first deliveries because curing behavior didn't match the technical data sheet — and in one case, the label matched an older revision than the SDS we'd been quoted against. Every step below exists because of a specific rejection, not because it's best practice in a textbook.

If you only have time for three steps, do 2, 4, and 6. Those are the ones that catch real problems.

Step 1 — Select by failure mode, not by product shorthand

Before you can inspect anything, you have to know what you're actually buying. "Adhesive" covers a wide range, and the names people use on the floor don't always map cleanly to chemistry.

The categories I actually see on purchase orders:

  • Cyanoacrylates — fast set, low gap-fill, poor peel. Fine for small assemblies, bad for anything that flexes.
  • Epoxies — structural, high strength, rigid. Fast-curing epoxy adhesives trade pot life for throughput; check how that affects your dispense window, not just your cycle time.
  • Polyurethanes — flexible, good for dissimilar substrates, moisture-sensitive during cure.
  • Tiling adhesives — usually cementitious or dispersion-based. Coverage rate and open time matter more than shear strength on the label.
  • Silicones and hybrid sealants — elongation and UV stability are the spec that matters, not the headline tensile number.

Checkpoint: write down the failure mode you're trying to prevent — peel, shear, thermal cycling, water ingress — before you write the spec. I've rejected more batches because the spec listed a tensile number while the actual failure was peel than for any other single reason.

Step 2 — Pull the TDS and SDS from the source, and lock the revision

This is the step where huntsman download searches go wrong for a lot of buyers. The product page on the manufacturer's site has the current revision. The PDF your distributor emailed you six months ago probably doesn't. Those two documents can disagree on something as basic as recommended cure schedule.

What I do: open the manufacturer's document portal, search the product code (not the trade name — the code), and download both the TDS and the current SDS. Then I write the revision number and download date onto our internal spec sheet. When a batch arrives, the first thing I check is that the certificate of analysis references a revision that matches or supersedes ours.

From the outside, it looks like a paperwork step. The reality is that Huntsman chemicals — and any large specialty supplier — run multiple product grades under similar names, and a grade mix-up is a real failure mode, not a theoretical one. I've seen a buyer receive a lower-viscosity grade because the PO used the trade name without the grade suffix.

Checkpoint: the revision on your internal spec sheet matches the revision cited on the COA. If it doesn't, stop.

Step 3 — Compare the COA line-by-line against the TDS spec range

Not every property on the TDS has a release spec attached. That's fine. But the ones that do — viscosity, solids content, gel time, density, color — need a value on the COA that falls inside the published range.

I use a simple two-column check: for each spec'd property, the published range on one side, the COA value on the other. Anything outside range gets flagged. Anything "typical" but not spec'd gets noted but not held.

Two things trip people up here. First, viscosity is usually measured at a stated temperature. A COA that says "viscosity 4,200 cps" without a temperature is not a usable value — call and ask. Second, some ranges are wide enough that "in spec" doesn't mean "suitable for your process." A gel time of 4–8 minutes is compliant at either end, but a 7-minute gel on a line with a 5-minute dispense window is a problem regardless of the COA.

Step 4 — Run the cure test under your line conditions, not the lab's

This is the step most plants skip, and it's the one that has caught the most problems for me.

The TDS cure schedule is a reference condition. Your line runs at a different ambient temperature, a different humidity, a different dispense bead geometry. Fast-curing epoxy adhesives are especially sensitive to this — a 30-second difference in gel time is invisible on the TDS and very visible when your fixture can't hold position long enough.

What we do: three specimens, cured under production conditions, tested for hand-load strength at the interval your process actually requires. For tiling adhesives, that's coverage rate and open time on the actual substrate — not on a glass plate in the lab. For structural epoxies, it's a lap shear pull after the minimum cure your line allows.

I should add that we don't run full mechanical testing on every batch. We run a quick functional check, and full ASTM D1002 or ISO 4587 lap shear testing only on new grades or new suppliers. If you're not set up for full mechanical testing, a simple pull-to-failure on a production-representative coupon tells you more than a COA ever will.

Checkpoint: every new grade gets a functional cure test before first production release. No exceptions. This is where the 12% rejection rate I mentioned drops to near zero on repeat suppliers.

Step 5 — Check packaging, labels, and storage history

By the time a drum or pail reaches your dock, it may have sat in a warehouse, a freight terminal, and a truck. Moisture-cure polyurethanes and cyanoacrylates are sensitive to this. Epoxies are usually more forgiving, but freeze-thaw cycles still matter for some formulations.

What I look for on receipt:

  • Label legibility and batch/lot number — matches the COA? If not, quarantine.
  • Drum or pail integrity — dents, seal damage, residue at the closure.
  • Temperature indicator or any shipping documentation that shows exposure outside the storage range on the TDS. If there's no record, I assume ambient and test accordingly.

Looking back, I should have made temperature documentation a contract requirement two years earlier than I did. At the time, it seemed like overkill for a supplier we'd used for years. Then a batch of moisture-cure product arrived with a compromised seal, and we spent three weeks proving it wasn't our storage that caused the issue. Now every contract includes a temperature-recording requirement on inbound shipments of moisture-sensitive grades.

Step 6 — Run a small-scale trial before releasing the full batch

This is the step that sounds obvious and gets skipped under schedule pressure. Do not release a full batch to production based on COA conformance alone.

Our protocol: pull one production-representative sample, run it through the actual process on a small run — 20 to 50 units, depending on the line — and inspect the output. For tiling adhesives, that means laying a test section and checking open time and coverage. For fast-curing epoxies, it means running the dispense and cure cycle at line speed and checking the bond.

If the trial output meets your acceptance criteria, release. If it doesn't, quarantine the batch and open a supplier conversation before you consume any more of it. That conversation is much easier with a trial result in hand than with a hunch.

Step 7 — Sign off, and leave a trail

The last step is administrative and it's the one that protects you six months later. Record the batch number, the COA revision, the TDS revision, the trial result, and the reviewer name and date. If a customer complaint comes in referencing a specific batch, this record is what lets you answer in an hour instead of a week.

I'm not a polymer chemist — I can't speak to why a given formulation drifted on gel time, and I don't try to. What I can tell you from a quality review role is that traceability is what separates a reject-and-redo (at the supplier's cost) from a reject-and-eat-the-loss (yours).

Common mistakes worth flagging

A few things I see repeatedly that aren't covered above:

  • Assuming "in spec" means "suitable." It doesn't. A batch can sit at the edge of a published range and still fail your process. Spec conformance is the floor, not the ceiling.
  • Relying on a single document source. If the TDS, SDS, and COA don't agree on grade and revision, you don't have a spec — you have three documents.
  • Skipping the trial on "known" suppliers. The rejection I remember most was from a supplier we'd used for five years. Known suppliers drift too; the trial is cheaper than the recall.
  • Not documenting ambient conditions during the cure test. A cure test without temperature and humidity recorded is not reproducible, and an unreproducible test won't hold up if you ever need to dispute a batch.

None of this is exotic. It's the routine that keeps a production line from inheriting a supplier's bad day. If you're setting this up from scratch, start with steps 2, 4, and 6 — those three catch the failures that actually cost money.

Need the related report or SDS pack?

Request the current document set before forwarding this topic to procurement, regulatory, or formulation colleagues.