Rare earth content is the part of a MgFeSi specification that is easiest to accept on trust and hardest to prove after a casting fails. The published grades run from 0.5 to 1.5 percent Re in FeSiMg6Re1 to 6.0 to 8.0 percent Re in FeSiMg10Re7, a range wide enough that a mislabelled lot can be inside the supplier's capability and outside the foundry's requirement. Japanese precision foundries close that gap with three controls: a written sampling plan, a named analytical method with a stated tolerance, and a lot release decision that compares the measured value against the agreed window before the alloy is used.

Precision casting work tolerates less scatter than general jobbing work, because thin sections and machined faces show graphite morphology differences that a heavier casting would hide. Rare earth changes that morphology: too little leaves degenerated graphite in slow cooling sections, while too much promotes carbide in thin walls. Verification is therefore not paperwork, it is the step that keeps the alloy inside the process window.

Because the certificate describes the lot the supplier sampled, not the alloy in the foundry's warehouse.

Each gap is closed by procedure rather than by argument, which is what makes verification auditable.

The published table gives the answer, and the differences are large enough to be decisive in casting design.

Two adjacent grades can differ by a factor of two in rare earth while carrying similar magnesium. That is why grade identity has to be verified by analysis and not by the label or the colour of the fracture.

The plan fixes who takes the sample, from where, how much, and what happens to it.

1. Sample every lot, with a composite drawn from at least three bags across the container

2. Take material from the interior of the lump, avoiding surface oxide and mould coating

3. Collect a minimum chip mass agreed with the laboratory, typically several grams after preparation

4. Split the sample so that one portion is retained for the life of the lot

5. Record bag positions and container number against the lot number

6. Re-sample when the first result sits near a specification limit

The sixth rule prevents a single borderline reading from deciding a lot. Near a limit, one more sample is cheaper than a rejected casting batch.

The method has to be named in the specification, with a tolerance and a reporting format.

Reporting should state the measured value, the method, the sample identification and the uncertainty. A single number without those four elements cannot be defended in a customer audit.

Release is a decision with three possible outcomes, and it should be documented as such.

Conditional release is the outcome that needs the most discipline, because the alloy is physically present and production wants to use it. Assigning it to a different section thickness is a legitimate engineering decision, provided the decision is recorded.

Verification is only useful if the result feeds back into treatment practice.

• Grade assignment per pattern becomes a documented decision rather than a habit

• Rare earth above the window triggers a review of section thickness and inoculant balance

• Rare earth below the window triggers a check for graphite degeneration in heavy sections

• Trend charts show supplier drift before the casting does

• Customer audits can be answered with records rather than with assurances

No. Low rare earth grades such as FeSiMg6Re1 work where scrap quality is controlled and sections are thin. Rare earth becomes important when tramp elements or slow cooling threaten graphite morphology.

It should match the section range rather than the market fashion. Thin sections commonly use 0.5 to 1.5 percent Re, medium sections 1.0 to 3.0 percent Re, and heavy sections 2.0 to 4.0 percent Re or higher.

For routine screening, yes. For release against a tight window or for a customer audit, an ICP result with a named method is the defensible record.

Preparation, matrix effects and calibration standards differ. Agreeing the method and exchanging a reference sample resolves most offsets before they become disputes.

No. Excess rare earth can promote carbide formation in thin sections and does not compensate for low magnesium. It is a morphology control addition, not a nodularity substitute.

Re-sample before deciding. A lot sitting at the edge of the window is the case where a single reading is least reliable and where a wrong release is most expensive.

Lot number, container and bag positions, sample identification, method, measured magnesium and rare earth, decision, and the retained sample. The set should allow a casting to be traced back to a measured lot.

Whenever the supplier, production line or scrap source changes, and otherwise on a fixed annual review. Stable history justifies keeping the frequency, not reducing it silently.

ZhenAn produces FeSiMg grades with rare earth from 0.5 to 8.0 percent, ISO 9001:2015 certified, in 10 to 50 mm and 50 to 100 mm lump, packed in 1 MT big bags at 20 to 25 MT per 20 foot container, delivered 7 to 10 working days after payment. Lot level certificates and retained samples are available for incoming verification.

Send the grade, rare earth window, analytical method and monthly quantity to begin an RFQ.

Analytical verification is easier when the grade table is explicit, as on the 38 percent silicon FeSiMg grade table page, which lists rare earth from FeSiMg6Re1 through FeSiMg10Re7. Fine granule grades used in precision work appear on the 3-6 mm rare earth FeSiMg granule page, and heavy section options on the rare earth FeSiMg lump page.