Metallographic examination of non-metallic inclusions in pressure vessel steel is a fundamental cleanliness test that safeguards the integrity of boilers, pressure vessels, reactors, and petrochemical equipment operating under sustained pressure and aggressive service conditions. Using standard chart comparison or automated image analysis in accordance with ASTM E45, ISO 4967, EN 10247, and GB/T 10561, accredited third-party laboratories rate sulfides, aluminates, silicates, and globular oxides to verify that materials conform to product specifications such as GB 713 and ASTM A20/A20M. This article explains the sampling, preparation, rating methods, and acceptance criteria used in pressure vessel steel inclusion inspection.
Technical Overview
Pressure vessel steels are safety-critical materials that must maintain adequate strength, ductility, and fracture toughness under internal pressure, cyclic loading, and often elevated temperature or hydrogen-bearing media. Non-metallic inclusions are inevitable constituents of steel, formed primarily during deoxidation and solidification, and their amount, size, morphology, and distribution directly influence the performance of the finished vessel. Elongated inclusions reduce through-thickness ductility, promote lamellar tearing at welded attachments, act as initiation sites for fatigue cracks and hydrogen-induced blistering, and increase the anisotropy of impact toughness.
Metallographic inclusion examination is therefore specified in material standards and vessel fabrication codes as a routine acceptance or qualification test. The evaluation is performed on polished, unetched cross-sections at a defined magnification, and the inclusions are classified into standardized categories:
- Type A – Sulfides: manganese sulfide and complex sulfides, typically highly deformable and elongated into stringers during rolling.
- Type B – Aluminates: alumina-based, brittle particles that break up into discontinuous chains aligned with the working direction.
- Type C – Silicates: deformable silicate inclusions that elongate into lines but remain continuous and ductile at rolling temperatures.
- Type D – Globular oxides: discrete, more or less equiaxed oxide particles that do not deform during hot working.
- Type DS – Single large globular oxides: isolated large-diameter oxide particles rated individually by size, recognized in ISO 4967 and GB/T 10561.
Results are reported as severity ratings that reflect inclusion length, number, and width, allowing objective comparison against specification limits and providing traceable evidence of steelmaking cleanliness.
Testing Samples
Boiler and Pressure Vessel Plates
The most frequently tested products are hot-rolled carbon, carbon-manganese, and low-alloy plates supplied to GB 713 (for example Q245R, Q345R, Q370R, 13MnNiMoR, and Cr-Mo grades such as 15CrMoR and 12Cr2Mo1R), ASTM A516/A516M, ASTM A387/A387M, and the EN 10028 series (P265GH, 16Mo3, 13CrMo4-5, 11CrMo9-10). Each plate heat lot is represented by specimens taken at the positions defined by the product standard or the purchase order.
Forgings and Bars
Pipes and Tubes
Seamless pipe for pressure service, including grades under ASTM A106, A335, and A333, may be subjected to inclusion rating to qualify steelmaking practice and to support toughness and sour-service performance assessments.
Weldments and Failure Samples
Although inclusion rating is primarily a base-metal test, polished sections from welded joints, delaminations, blisters, and fractured components are examined metallographically during failure analysis to identify inclusions as crack initiation sites.
Testing Items
A typical metallographic inclusion inspection report for pressure vessel steel covers the following items:
- Identification and classification of inclusion types (A, B, C, D, DS) on the polished longitudinal section.
- Severity ratings for each type, separated into thin and heavy series, in increments of 0.5 within the standard chart range.
- Worst-field rating for each type and series (chart comparison method), together with the field coordinates.
- Statistical ratings such as the mean severity over all examined fields, the number of inclusion-bearing fields, and cumulative indices where specified.
- Quantitative image-analysis outputs, including total inclusion length, average and maximum width, particle count, aspect ratio, and area fraction.
- Observations on inclusion distribution and orientation, banding, and the location of the worst field relative to the plate surface and mid-thickness.
- Comparison of results with the acceptance limits of the applicable material specification or the client's supplementary cleanliness requirement.
Testing Methods
Sampling and Specimen Preparation
For plates, specimens are normally cut with the polished plane parallel to the rolling direction (longitudinal section), at the quarter-width position from the plate edge and at the depth required by the governing specification. The polished area is prepared to at least the minimum size demanded by the rating standard so that a statistically valid number of fields can be examined. Preparation quality is decisive: inclusions must be retained without pull-out, smearing, comet tails, or relief. Recommended practice follows ASTM E3 and ASTM E45 or the equivalent clauses of ISO 4967 and GB/T 10561:
- Cut with abrasive sawing under abundant coolant to avoid heating or deformation of the inclusion population.
- Mount specimens, if required, taking care that elongated inclusions remain parallel to the polished face.
- Grind through a progressive silicon-carbide sequence with thorough cleaning between steps.
- Polish with diamond or oxide suspensions to a scratch-free, flat, undistorted finish.
- Examine in the unetched condition; etching is used only subsequently, if microstructure verification is also required.
Chart Comparison Rating
The classical micrographic method compares each microscope field, viewed at 100× and covering a defined 1 mm² viewing area, with the standard rating diagrams of ASTM E45 or the pictorial charts of ISO 4967 and GB/T 10561. Successive contiguous fields are examined until the specified area has been covered, typically at least 100 mm² for a full evaluation. Depending on the selected procedure, the result may be expressed as the worst field found for each inclusion type (the common approach for pressure vessel plate acceptance), as field-by-field statistics, or as length and count totals for products with special cleanliness requirements. Thin and heavy series charts differentiate finer from coarser, thicker inclusions, and the DS diagram assigns ratings according to the diameter of single large globular oxides.
Automated Image Analysis
Many accredited third-party laboratories supplement or replace manual comparison with quantitative image analysis in accordance with ASTM E2142 or ASTM E1245. Digitized images of the polished section are thresholded to isolate inclusions, and morphological parameters — length, width, aspect ratio, count, and area — are used to classify particles into categories equivalent to the A, B, C, D, and DS chart types. Image analysis improves repeatability, removes operator subjectivity, and yields additional statistical measures such as area fraction and size distribution, which are valuable for clean-steel development and data retention.
Evaluation and Reporting
The laboratory rates each field or the worst field, applies the calculation method defined by the specification (worst field, mean value, or cumulative index), and compares the outcome with the acceptance limits. A complete report identifies the steel grade, heat number, product form, sampling location and orientation, preparation procedure, standard and method applied, ratings for every inclusion type and series, the examined area and magnification, representative photomicrographs of the worst field, and a clear conformity statement against the specified limits.
Applications
Inclusion metallography is applied throughout the life cycle of pressure vessel materials:
- Mill acceptance testing: verification of plate, forging, and pipe cleanliness against GB 713, ASTM A20/A20M supplementary requirements, EN 10028, and purchase specifications before release for fabrication.
- Code fabrication support: documentation of material quality during vessel manufacture to the ASME Boiler and Pressure Vessel Code, EN 13445, or GB/T 150 design frameworks.
- Sour service qualification: assessment of inclusion populations, especially elongated sulfides, in materials intended for wet hydrogen sulfide service where inclusion stringers are prime initiation sites for hydrogen-induced stepwise cracking.
- Through-thickness (Z-quality) qualification: cleanliness screening supporting improved deformation properties perpendicular to the plate surface per EN 10164, reducing the risk of lamellar tearing in heavily restrained welded joints.
- Failure analysis: identification of inclusions at the origin of fatigue fractures, blisters, delaminations formed during hot forming of heads, and brittle fractures.
- Steelmaking and supplier development: benchmarking of calcium treatment, ladle refining, and vacuum degassing practices through statistically robust inclusion statistics.
FAQ
Why are inclusions rated on unetched polished specimens?
Inclusion size, shape, and contrast against the steel matrix are best revealed in the polished, unetched condition, as required by ASTM E45, ISO 4967, and GB/T 10561. Etching would reveal the microstructure but can obscure inclusion boundaries, exaggerate apparent widths, and bias the rating.
What magnification and examined area are used?
The standard rating magnification is 100×, with each field corresponding to a 1 mm² viewing area. A full evaluation normally covers at least 100 contiguous fields, i.e., about 100 mm², taken from a polished surface prepared with an adequate margin beyond the rated area.
What is the difference between the worst-field method and statistical methods?
The worst-field method reports the most severe rating observed for each inclusion type over the whole examined area, which suits acceptance of pressure vessel plates where a single dense inclusion cluster can be deleterious. Statistical methods aggregate every field to produce mean values, totals, or inclusion counts, and are preferred for evaluating steelmaking consistency and clean-steels with low inclusion populations.
What acceptance limits are typical?
Unless a purchase order imposes stricter values, common supplementary requirements call for each inclusion type and series not to exceed approximately severity 2.0 to 3.0 at the worst field, with tighter limits for heavy series and for grades destined for sour or low-temperature service. The governing value is always the limit stated in the applicable product standard or specification.
Are ASTM E45, ISO 4967, and GB/T 10561 interchangeable?
All three standards use comparable chart-based classification of A, B, C, D inclusions at 100×, and GB/T 10561 is technically aligned with ISO 4967, including the DS single-globular-oxide category. However, sampling rules, calculation options, and rounding differ in detail, so the report must state which standard and which method within it were applied.
Can image analysis replace manual chart comparison?
Yes, when performed per ASTM E2142 or ASTM E1245 by qualified personnel with validated software and calibrated systems. Image analysis classifies particles by measured morphology and provides reproducible quantitative results, and many specifications accept it as an equivalent or supplementary method, provided the correlation with chart ratings is demonstrated.
How should samples be taken from a plate?
The customary location is a longitudinal section, polished parallel to the rolling direction, cut at the quarter-width position from the plate edge, with the sampled depth (surface, mid-thickness, or full-thickness) as specified by the product standard or agreement, because inclusion stringers are longest and most consequential in this orientation.