When non-stick lifts off hard anodised cookware, the reflex is to blame the PTFE. In production the coating is almost never the culprit. The failure is nearly always in the surface underneath it — a hard anodic layer that was sealed, contaminated, or finished in a way that gave the resin nothing to key into.

This matters commercially because the two halves are usually done by different people. The anodiser hands over parts that pass every test on the spec sheet, and the coater applies a resin that works everywhere else. Both are right, and the pan still fails.

The mechanism: PTFE holds on mechanically

A hard anodic coating is a dense oxide layer full of fine pores running perpendicular to the surface. A non-stick topcoat does not chemically bond to aluminium oxide in any meaningful way. It keys into those pores and grips mechanically once cured.

Everything that determines adhesion follows from that single fact. If the pores are closed, filled, contaminated or absent, there is nothing for the resin to hold and the coating is sitting on a smooth ceramic surface waiting to peel.

The four causes, in order of how often we see them

1. The coating was sealed

This is the big one, and it is a process-sequence error rather than a mistake in any single step. Sealing exists to close the pores. It is exactly the right thing to do on decorative or corrosion-critical anodising, and exactly the wrong thing to do before applying a topcoat that needs those pores open.

It happens because the anodising line has one standard route and sealing is the last tank on it. Parts destined for non-stick have to bypass that tank. If your line cannot skip sealing for a specific load, you have a routing problem that will keep producing peeling pans no matter what the coater does. See sealing methods compared.

2. Contamination between the two steps

An unsealed hard anodic layer is porous and absorbent — it will take up whatever it meets. Handling with bare hands puts skin oils into the pores. Storing parts uncovered in a workshop puts airborne oil and dust in. Water left in the pores from a final rinse and dried in still air can leave mineral residue.

The practical rule is that the gap between anodising and coating should be short, and the parts should be handled with gloves and kept covered. Where the two operations are in different premises, this is the step that fails most often.

3. Coating thickness and pore structure

Very thick hard anodic coatings run at high current density and low temperature, which produces a dense layer with relatively fine porosity. Very thin coatings do not offer enough pore depth for the resin to anchor into. There is a working band, and it is narrower than most specifications acknowledge — the target is a coating thick enough to key into but not so dense that the resin cannot enter.

If adhesion is marginal and everything else is controlled, coating thickness and current density are the parameters to examine. Measure before you argue: see eddy-current thickness measurement.

4. Cure schedule

PTFE topcoats need a defined time at a defined temperature to flow into the pores and coalesce. Under-cured coatings look finished and fail early, and the failure looks exactly like an adhesion problem because it is one. Oven profiling — actual part temperature over time, not the oven setpoint — is the check, and it is rarely done.

Telling the causes apart

Peels in clean sheets, bright surface beneath

The coating never keyed in. Sealed substrate, or a surface so dense the resin could not enter. Check the routing first: did the load go through the sealing tank?

Lifts in patches, no pattern

Contamination between operations. Look at handling and storage between the two lines, not at either process.

Fails at the contact point

The rack mark is bare or thinly coated aluminium, so there are no pores there at all. The weakest adhesion site on the piece is always the jig contact — another reason to hide it under a handle rivet. See rack marks on anodised cookware.

Wears through rather than lifting

Not an adhesion failure. That is a thickness or cure problem, and Taber testing per ASTM D4060 will separate them.

What to control, in order

  1. Route non-stick work around the sealing tank. If the line cannot do that reliably, fix it before anything else.
  2. Close the gap between anodising and coating, with gloved handling and covered storage.
  3. Hold coating thickness in band and verify it rather than assuming it.
  4. Profile the cure oven with a part-mounted probe, not the setpoint.
  5. Test to a standard — Taber wear per ASTM D4060 and cross-hatch adhesion per ASTM D3359 — so that "it peels" becomes a number both sides can act on.

The full process and test-method detail is in PTFE impregnation on hard anodize, and the overview of the stack is in hard anodizing with PTFE. For how this fits a cookware line as a whole, see hard anodising for cookware.

Frequently asked questions

Why does non-stick coating peel off hard anodised cookware?

Almost always because of the surface underneath rather than the coating itself. PTFE does not chemically bond to aluminium oxide; it keys mechanically into the pores of the anodic layer. The most common cause is that the parts were sealed, since sealing exists specifically to close those pores. Other causes are contamination between anodising and coating, coating thickness outside the workable band, and an under-cured topcoat.

Should anodised parts be sealed before applying a non-stick coating?

No. Sealing closes the pores in the anodic layer, and those pores are what the PTFE keys into. Parts destined for a non-stick topcoat must bypass the sealing tank. Because sealing is usually the last stage of a standard anodising route, this is a line-routing problem as much as a process one.

How can I tell whether it is an adhesion problem or a wear problem?

Look at how it fails. Coating that lifts in clean sheets leaving a bright surface underneath never keyed in, which points to a sealed or over-dense substrate. Coating that wears through gradually is a thickness or cure issue, not adhesion. Taber abrasion per ASTM D4060 and cross-hatch adhesion per ASTM D3359 separate the two objectively.

Why does non-stick fail first at the rack contact point?

The contact point is where the jig touched the part, so no oxide grew there. With no pores, there is nothing for the resin to key into, making it the weakest adhesion site on the piece. It is one more reason to place contacts where a handle rivet or base plate will cover them.