Kettle steel: why silicon matters more than strength
The most common mistake in specifying a galvanizing kettle is asking for a stronger or more corrosion-resistant steel. For molten zinc service, that is backwards.
Molten zinc dissolves iron, and silicon speeds it up
Zinc attack on a kettle wall is not oxidation and it is not conventional corrosion. Iron dissolves into the molten zinc and forms an Fe-Zn alloy layer at the interface. Whether that layer protects the wall or not depends on the chemistry of the steel.
With low silicon, the alloy layer is dense and adherent, and it slows further attack. Raise the silicon and the layer becomes non-adherent — it sloughs off, exposes fresh iron, and the wall keeps dissolving. Phosphorus and aluminium do the same thing.
This is why stainless steels and heat-resisting steels — which outperform plain carbon steel almost everywhere else — resist molten zinc worse than a plain low-silicon structural steel does.
Our specification
There is no equivalent EN or ASTM grade designation for galvanizing kettle plate. Kettle steel is specified by chemistry, not by grade name. Anyone quoting you an “equivalent” pressure-vessel grade is quoting a steel with roughly ten times the silicon a kettle should have.
| Element | Our specification | Why |
|---|---|---|
| Si | ≤ 0.03 % | The single largest accelerator of zinc attack on the kettle wall |
| P | ≤ 0.02 % | Same mechanism as silicon |
| Al (Als) | ≤ 0.045 % | Same mechanism |
| Si + P + Al | ≤ 0.045 % combined | The controlling limit per BG/T 1501-2015 §3.2 |
| C | 0.05 – 0.11 % | Kept up, not down — carbon holds wall strength against hydrostatic load at 450 °C |
| Mn | 0.30 – 0.50 % | Same reason |
| S | ≤ 0.015 % | Cleanliness |
| Residuals | Ni ≤ 0.3 %, Cr ≤ 0.1 %, Cu ≤ 0.20 %, N ≤ 0.008 % |
These are the limits we order plate to. The evidence is the mill certificate that ships with the kettle — heat number, full analysis, signed independently of production. Do not take a chemistry table on any supplier's website as proof, including this one; take the certificate.
Mechanical properties: tensile strength ≥ 325 MPa, elongation after fracture ≥ 33 %, 180° bend test passed.
Steelmaking: aluminium-killed, hot-rolled. Converter or electric arc melting with secondary refining, and heavy reduction during rolling to close out centreline defects.
A note on very low carbon. It is tempting to read “low silicon, low carbon” and ask for the lowest carbon available. BG/T 1501-2015 explicitly advises against plate below 0.05 % C — the wall needs that strength to resist the static pressure of the zinc at temperature without bulging.
Verified twice, not once
The mill inspects every plate ultrasonically to GB/T 2970, Class III or better. We then re-verify on receipt: chemistry to GB/T 222, mechanical properties to GB/T 228.1, and ultrasonic re-check.
Plate that fails re-verification is not used in a kettle shell. It may be used for flanges or external supports, and it is specially marked if it is.
Mill certificates are supplied to EN 10204 Type 3.1 — signed by an inspection representative independent of production at the steel mill. Type 3.2, countersigned by a third party or by your own inspector, is available where your specification requires it.
How traceability actually works. The heat number on the mill certificate is transferred onto every piece cut from that plate, and carried through cutting, forming and fit-up (BG/T 1501-2015 §5.2.1). At the end you can point at a seam on your kettle and trace it back to a heat and an analysis.
That is what the paperwork is for. A certificate that cannot be tied to a specific plate in a specific kettle is a document, not evidence.
Questions
What is XG08?
It is the Chinese designation commonly used for galvanizing kettle plate. It is useful shorthand in China and meaningless in Europe, which is why we quote the chemistry instead. What matters is the numbers in the table above and the mill certificate that proves them.
Isn't a thicker wall better than a purer steel?
They solve different problems. Thickness buys you years of sacrificial material and resistance to bulging. Chemistry decides how fast those years are consumed. A 60 mm wall in a high-silicon steel can have a shorter life than a 50 mm wall in a low-silicon one.
Can you supply to a customer-specified chemistry?
Yes, within what the mill will roll. Send your specification and we will tell you whether it is available, what it costs and what the lead time on the plate is.
Source: BG/T 1501-2015, Steel Hot-Dip Galvanizing Kettles, China Surface Engineering Association — §3, Tables 1, 2, 3.
The weld is a different steel
Plate chemistry is half the story. Weld metal can carry thirty times the silicon of the parent plate depending on the consumable — and that is where kettles usually go through first.