Stainless Steel Non-Metallic Inclusion Rating Analysis

Non-metallic inclusion rating is a fundamental cleanliness test that determines whether stainless steel meets the quality requirements of demanding applications in aerospace, medical, automotive, and chemical engineering. This article provides a comprehensive overview of inclusion classification (Types A, B, C, D, and DS), the governing standards such as ASTM E45, ISO 4967, and GB/T 10561, specimen preparation, rating procedures, and the practical application of test results in material qualification and failure analysis.

Technical Overview

Non-metallic inclusions are foreign particles embedded in the stainless steel matrix, originating from deoxidation products, refractory erosion, slag entrapment, reoxidation during teeming, or sulfide formation during solidification. Although often microscopic, these particles have a disproportionate influence on service performance. Elongated manganese sulfides act as initiation sites for pitting corrosion in chloride environments, hard oxide stringers reduce fatigue life, and coarse silicates or globular oxides cause surface defects, tearing during deep drawing, and polishability problems that are especially critical for stainless steel products.

Inclusion rating analysis quantifies the amount, size, morphology, and distribution of these particles using standardized microscopic comparison charts. The evaluation classifies inclusions into five characteristic types based on their shape and deformation behavior:

• Type A — Sulfides: ductile, elongated grey inclusions, typically manganese sulfide (MnS), which deform plastically during hot working.
• Type B — Alumina: discontinuous stringers of fragmented, non-deformable aluminum oxide particles aligned in the working direction.
• Type C — Silicates: elongated, ductile inclusions that were plastic at hot-working temperature and appear as isolated streaks with rounded ends.
• Type D — Globular oxides: randomly dispersed, non-deformable round particles, often calcium-aluminate or spinel oxides.
• Type DS — Single large globular particles: isolated oxide particles exceeding roughly 13 micrometers in diameter, rated separately because of their strong effect on fatigue and polishability.

Because stainless steels rely on a continuous passive chromium-oxide film for corrosion resistance, inclusion control is often more critical than in plain carbon steels. The most widely applied standards are ASTM E45, ISO 4967, GB/T 10561 (the Chinese equivalent of ISO 4967), JIS G 0555, and image-analysis-based methods such as EN 10247 and ASTM E2142.

Testing Samples

Correct sampling is decisive for a representative rating. Specimens are normally taken in the longitudinal orientation — that is, the polished surface is parallel to the principal hot-working direction — so that the characteristic inclusion morphology created by deformation can be identified. Typical sample categories include the following:

Plate, Sheet, and Strip

Samples are cut from locations specified by the product specification, usually at quarter-width or mid-width positions. The polished face is oriented in the rolling direction so that sulfides and silicates appear as elongated strings, allowing accurate distinction between ductile and brittle inclusion types.

Bar, Rod, and Wire

Longitudinal sections are commonly removed from mid-radius or between surface and center, where inclusions are statistically representative of the billet. Center-to-surface gradient studies may require multiple parallel sections, particularly for free-machining grades such as 303 and 416, in which deliberately high sulfur content produces long sulfide strings.

Seamless and Welded Tubing

Longitudinal wall sections are prepared for cleanliness assessment, and welded tubes receive additional attention in the weld seam area, where reoxidation during welding can generate localized oxide clusters not present in the base metal.

Forgings and Castings

Forgings are sectioned parallel to the grain flow, while castings are sampled at thick sections where solidification is slow and inclusions have more time to agglomerate. Investment-cast stainless components often require stricter oxide ratings because ceramic shell fragments may be entrapped in the metal.

Weldments and Additive Manufactured Parts

Dissimilar-metal welds, overlay cladding, and laser or electron-beam powder-bed fusion parts contain unique oxide populations. Sectioning through the fusion zone allows evaluation of oxide inclusions, lack-of-fusion particles, and process-related contamination.

Testing Items

The core deliverable of the analysis is a severity rating for each inclusion type, reported separately for thin and thick series and normally expressed in half-step increments from 0 to 5. Typical reported items include:

• Worst-field ratings for Type A, B, C, and D inclusions, thin and thick series (ASTM E45 Method A style).
• Severity indices and total inclusion counts per unit area (ASTM E45 Method B style).
• Single large particle (Type DS) ratings for critical fatigue applications.
• Cleanliness indicators such as area fraction, total inclusion length per square millimeter, and number of fields exceeding specified limits.
• Extreme value analysis results predicting the size of the largest inclusion expected in a large production volume (ASTM E2283).
• Optional inclusion chemistry and morphology identification by scanning electron microscopy with energy-dispersive spectroscopy (SEM-EDS), for example to confirm MnS, alumina, spinel, calcium-aluminate, or titanium nitride.

Acceptance criteria are defined by the purchaser. A typical specification for a high-cleanliness austenitic grade might read A thin ≤ 2.0, B thin ≤ 1.5, C thin ≤ 1.5, D thin ≤ 1.5, and DS ≤ 1.0, while free-machining grades tolerate much higher A ratings by design.

Testing Methods

Specimen Preparation

The specimen is sectioned without overheating, mounted, ground through successive abrasive grades, and polished to a scratch-free mirror finish. Rating is performed on the unetched polished surface, because etching can attack inclusions and alter their apparent size. The polished area is typically on the order of 100 to 200 square millimeters to ensure statistical validity, and the examination is carried out at 100× magnification over contiguous fields with a viewing area of approximately 0.71 square millimeters per field.

Manual Chart Rating (ASTM E45 Method A, ISO 4967, GB/T 10561)

The operator systematically traverses the entire polished surface and compares each field against the reference charts originally derived from the Jernkontoret (JK) diagrams. The worst field observed for each inclusion type and thickness series is recorded as the rating. This method is fast, reproducible, and directly compatible with most purchase specifications, and it is the default approach for routine quality control of stainless steel bars, plates, and forgings.

Severity Index Method (ASTM E45 Method B)

Method B rates every field rather than only the worst one, estimating the length or number of inclusions per field against calibration charts and converting the results into severity indices. The summed indices give a more complete statistical picture of cleanliness, which is useful when comparing competing suppliers or tracking long-term process improvements in secondary metallurgy such as argon oxygen decarburization (AOD) refining and ladle furnace treatment.

Automatic Image Analysis

Computerized systems capture digital images of each field and classify inclusions automatically by morphology, contrast, and size according to EN 10247, ASTM E2142, or the image-analysis annexes of ISO 4967. Automated rating eliminates operator subjectivity, generates full size distributions, and can examine thousands of fields, making it the preferred method for research, statistical process control, and high-volume mill certification.

SEM-EDS Inclusion Identification

When the origin of inclusions must be determined — for example, during a supplier quality investigation or failure analysis — individual particles are analyzed by SEM-EDS. Composition reveals whether the inclusion is a deoxidation product, refractory contamination, slag entrapment, or reoxidation, and this information guides corrective action in steelmaking. Titanium-stabilized grades such as 321 are also examined for titanium nitride particles, which are assessed separately from the standard A to DS categories.

Statistics of Extreme Values

For fatigue-critical parts such as bearing components and valve springs, the size of the largest inclusion governs life. Extreme value methods per ASTM E2283 extrapolate the maximum inclusion size expected over a defined production tonnage from a measured sample, providing a quantitative link between inclusion ratings and predicted fatigue performance.

Applications

Inclusion rating results directly support material selection, supplier qualification, and root-cause failure analysis across the full range of stainless steel end uses:

• Medical implants and instruments, where surface-critical 316L, F138 bar, and custom alloys must meet stringent oxide limits to guarantee polishability, biocompatibility, and fatigue durability.
• Aerospace fasteners, structural forgings, and high-temperature components requiring documented cleanliness to ensure reliability under cyclic loading.
• Automotive fuel-injection and valve components, exhaust systems, and bright-annealed trim strip, where inclusions would cause pinholes, lines, and corrosion initiation on visible surfaces.
• Chemical, petrochemical, and nuclear pressure-boundary components, for which coarse silicates and sulfide clusters are restricted to limit pitting, hydrogen-induced cracking, and weld defects.
• Deep-drawn and spun household products, watch cases, and sanitary tubing, where hard oxides produce tool marks and scoring during forming.
• Melt-shop process optimization, in which AOD, ladle, tundish, and casting practice changes are verified by statistically comparing inclusion ratings before and after implementation.

FAQ

Which standards are used for stainless steel inclusion rating?

The most common specifications are ASTM E45, ISO 4967, GB/T 10561, JIS G 0555, and EN 10247 or ASTM E2142 for automatic image analysis. ASTM E45 and GB/T 10561/ISO 4967 are mutually compatible and use the same A, B, C, D, and DS classification, so results are generally transferable between these standards.

Is etching required before rating?

No. Rating is always performed on the polished, unetched surface because etching preferentially attacks inclusion-matrix boundaries and changes the apparent inclusion width, which would distort the thin/thick classification.

What is the difference between ASTM E45 Method A and Method B?

Method A records the worst field for each inclusion type and thickness series and is the format used in most purchase specifications. Method B rates all examined fields and reports severity indices, giving a more complete measure of overall cleanliness for supplier comparison and process development.

Why do free-machining stainless grades show high inclusion ratings?

Grades such as 303 and 416 intentionally contain elevated sulfur, which forms elongated manganese sulfide inclusions that chip-break during machining. High Type A ratings in these grades are expected and acceptable; the same ratings in 304L or 316L would indicate a cleanliness problem.

How do inclusions affect corrosion resistance?

Manganese sulfide inclusions dissolve or form micro-crevices in chloride environments and are the most common initiation sites for pitting corrosion in stainless steel. Oxide inclusions also disrupt the passive film locally, so cleanliness ratings correlate strongly with pitting resistance in critical service.

What sample size and how many fields are examined?

A typical polished area is 100 to 200 square millimeters on a longitudinal section, and a full manual rating covers the entire area in contiguous fields at 100×. The exact sampling location, number of specimens, and minimum fields are defined by the applicable product specification or by agreement between purchaser and supplier.