Two-dimensional distribution analysis of inclusions in pressure vessel steel is a specialized metallographic evaluation that maps the type, size, morphology, and spatial arrangement of non-metallic inclusions on prepared cross-sections of carbon, low-alloy, and Cr-Mo pressure vessel steels. This analysis is essential for verifying steel cleanliness, assessing anisotropy introduced by rolling and forging, and predicting risks of laminations, hydrogen-induced cracking, and fatigue initiation in components such as reactors, boilers, and storage vessels. Accredited third-party laboratories perform this testing in accordance with internationally recognized standards including ASTM E45, ISO 4967, GB/T 10561, and EN 10247, providing manufacturers, fabricators, and plant operators with quantitative evidence of material quality and code compliance.
Technical Overview
Pressure vessel steels, including carbon steel grades such as SA-516 and low-alloy Cr-Mo grades such as SA-387, are required to sustain high internal pressures, thermal cycling, and often aggressive service environments over multi-decade design lives. Non-metallic inclusions such as manganese sulfides, alumina, silicates, calcium aluminates, and titanium nitrides act as stress raisers and hydrogen traps that can degrade transverse ductility, impact toughness, fatigue resistance, and resistance to hydrogen-induced cracking. Because these inclusions are inherited from deoxidation, refining, casting, and subsequent hot working, their two-dimensional distribution on a polished cross-section provides a direct record of steelmaking cleanliness and thermomechanical processing quality.
Two-dimensional inclusion distribution analysis goes beyond a simple cleanliness rating. It quantifies how inclusions are dispersed across the inspected plane, revealing elongated sulfide stringers aligned with the rolling direction, oxide clusters or bands located at mid-thickness, and locally enriched zones associated with solidification segregation. The resulting data support acceptance decisions against material specifications, root-cause investigation of defects, and input to extreme-value statistical models that estimate the largest inclusion likely to exist in a full plate or forging. Typical deliverables include inclusion classifications, size distributions, area fractions, number densities, aspect ratio statistics, spatial distribution maps, and thinness/thickness-direction gradient profiles.
Testing Samples
Plate Samples for Shell and Head Materials
Samples are taken from pressure vessel plates in accordance with product standards such as ASME SA-516 and SA-387 and sampling codes like ASTM A370 or EN 1011. Metallographic sections are typically prepared from quarter-thickness and mid-thickness locations, and both longitudinal (parallel to rolling) and transverse orientations are examined so that inclusion shape anisotropy can be characterized. For thick plates, multi-position through-thickness sampling allows the analyst to detect centerline segregation bands where inclusions tend to concentrate.
Forging Samples for Nozzles, Flanges, and Tubesheets
Forged components exhibit inclusion distributions governed by deformation flow during forging. Samples are extracted from locations specified in the forging specification, including the extension region, mid-radius, and near-surface zones, with examination planes oriented both parallel and perpendicular to the predominant metal flow direction. This helps identify inclusion stringers that could compromise the through-thickness properties of hub regions and nozzle necks.
Weld, Heat-Affected Zone, and Cladding Samples
For welded joints in vessel shells and for roll-bonded or weld-overlay clad plates, sections crossing the weld metal, heat-affected zone, and base metal are examined. The analysis distinguishes indigenous base-metal inclusions from weld-related oxides and slag islands, and evaluates whether coarse inclusions at the fusion boundary could act as crack initiation sites during pressure testing or post-weld heat treatment.
Defect and Failure Investigation Samples
Samples containing laminations, stepwise cracking, hydrogen blisters, or fatigue origins are sectioned so that the defect plane and the surrounding inclusion field are captured on the same polished surface. Mapping the inclusion distribution adjacent to the defect frequently reveals the causative inclusion type, for example elongated MnS chains responsible for lamellar tearing or oxide clusters at the origin of hydrogen-induced cracking.
Testing Items
The scope of two-dimensional distribution analysis generally covers the following quantitative and qualitative items, each tailored to the relevant acceptance specification and end-user requirements:
- Inclusion type identification: sulfides, aluminates, silicates, globular oxides, calcium-treated oxides, titanium carbonitrides, and complex multiphase inclusions.
- Standardized classification ratings per ASTM E45 Method A thin, thick, and worst-field ratings for A-type sulfides, B-type alumina strings, C-type silicates, D-type globular oxides, and DS-type single large globular inclusions.
- Morphometric parameters: length, width, equivalent circle diameter, aspect ratio, projected area, and perimeter for individual inclusions and for populations by class.
- Distribution metrics: number density (count per square millimeter), area fraction (volume-free 2D cleanliness index), nearest-neighbor distance, inter-particle spacing, and coefficient of variation of local density.
- Spatial distribution characterization: uniformity assessment, identification of clustering, banding, centerline enrichment, and surface-to-core gradients across the inspection plane.
- Orientation analysis: measurement of the alignment angle distribution of elongated inclusions relative to the rolling direction, quantifying anisotropy relevant to transverse and through-thickness properties.
- Chemical composition of individual inclusions determined by energy-dispersive spectroscopy, enabling deoxidation practice diagnosis and calcium treatment effectiveness evaluation.
- Extreme-value statistics: application of the statistics of extreme values (SEV) or generalized Pareto distribution to predict the maximum inclusion size expected within a defined inspection volume or component surface.
Testing Methods
Specimen Preparation
Representative sections are cut with abrasive or low-speed saws using abundant coolant to avoid thermal alteration of inclusions. Specimens are mounted, ground through successive silicon carbide grits, and polished with diamond suspensions down to one micrometer or finer, following guidelines of ASTM E3. Final polishing is performed with minimal etching or none at all, since inclusion ratings must be made on the unetched condition; a light nital etch may be applied afterwards solely for correlating inclusions with microstructure. Careful preparation prevents artifacts such as sulfide pull-out, edge rounding, and oxide smearing, all of which would bias the two-dimensional distribution data.
Light Optical Microscopy and Standard Diagram Rating
Polished sections are surveyed with a calibrated research-grade light optical microscope at the magnification prescribed by the applicable standard, commonly 100x with a field area of 0.50 mm² per ASTM E45 and ISO 4967. Analysts compare observed inclusions against the standard diagrams of ASTM E45, ISO 4967, GB/T 10561, EN 10247, or JIS G 0555, assigning severity numbers by inclusion type and series, and recording the worst field and worst-field-plus-frequency information required by many pressure vessel material specifications.
Automated Image Analysis
Computerized image analysis in accordance with ASTM E1122 and ASTM E1245 provides fully quantitative two-dimensional characterization. The system captures contiguous fields across the defined inspection area, segments inclusions by gray-level thresholding and morphological filtering, discriminates sulfides from oxides by gray level and aspect ratio rules, and outputs size distributions, area fractions, number densities, and spatial coordinate data. Because every detected feature is stored with its centroid coordinates, the method directly supports distribution mapping, clustering statistics, and gradient profiling that manual chart rating cannot supply.
Scanning Electron Microscopy with EDS
Scanning electron microscopy combined with energy-dispersive X-ray spectroscopy, executed per ASTM E2142, is used to confirm inclusion chemistry, distinguish chemically similar phases, and rate inclusions whose morphology is ambiguous in the light microscope. Automated SEM-EDS feature analysis extends two-dimensional mapping to composition-resolved inclusion populations, differentiating, for example, calcium aluminate from pure alumina and manganese sulfide from calcium sulfide. This compositional dimension is particularly valuable for evaluating calcium treatment of sulfur-bearing pressure vessel steels intended for sour service.
Statistical and Spatial Evaluation
Raw particle data are processed into engineering-relevant summaries: histograms and cumulative distributions of size and aspect ratio; two-dimensional density maps and contour plots revealing bands and clusters; and directional statistics quantifying alignment with the principal working direction. Extreme-value analysis of the largest inclusions measured over a defined total inspection area enables estimation of the maximum inclusion size in the component, a key input for fatigue assessment and hydrogen-induced cracking susceptibility studies.
Applications
Two-dimensional inclusion distribution analysis is applied throughout the life cycle of pressure vessel materials and components:
- Qualification of plate, forging, and tubular heats against ASME, EN, and GB pressure vessel material specifications, including supplementary cleanliness requirements for critical service.
- Quality control of steelmaking and calcium treatment practices, providing feedback on deoxidation, ladle metallurgy, vacuum treatment, and continuous casting parameters.
- Assessment of resistance to hydrogen-induced cracking and stress-oriented hydrogen-induced cracking in wet H2S service, where elongated sulfides and oxide stringers control crack path development.
- Through-thickness property qualification of shells, tubesheets, and nozzle attachments loaded in the short-transverse direction, using inclusion distribution data to support reduction-of-area and lamellar tearing evaluation.
- Fatigue design substantiation for vessels subject to pressure and thermal cycling, using maximum inclusion size estimates derived from extreme-value statistics.
- Surveillance, repair qualification, and remaining-life assessment of operating vessels, including evaluation of replacement plate and weld repairs for older units.
- Root-cause failure analysis of laminations, blisters, hydrogen stepwise cracking, and brittle fracture initiation where inclusions are suspected contributors.
FAQ
What standards govern two-dimensional inclusion evaluation of pressure vessel steel?
The most widely applied standards are ASTM E45 for chart rating, ASTM E1122 and ASTM E1245 for image-analysis-based measurement, ASTM E2142 for SEM-based rating and classification, ISO 4967 and its national equivalents GB/T 10561 and EN 10247 for micrographic rating with standard diagrams, and JIS G 0555 for Japanese practice. Specimen preparation follows ASTM E3, and terminology is defined in ASTM E7. Material product specifications then invoke one of these methods together with acceptance limits.
Why is two-dimensional distribution analysis performed instead of relying on a single cleanliness number?
A single rating field describes severity at one location but ignores spatial structure. Two-dimensional analysis reveals banding, clustering, and through-thickness gradients that control anisotropy, lamellar tearing susceptibility, and hydrogen-induced cracking paths. Two plates with identical average ratings can show very different local inclusion densities, and the distribution information identifies the plate that presents the greater engineering risk.
How are sulfides, oxides, silicates, and globular oxides distinguished?
In the light microscope, sulfides appear as elongated gray stringers, alumina forms discontinuous stringer-like chains of dark particles, silicates appear as elongated dark streaks, and globular oxides are randomly dispersed dark rounded particles, matching the A, B, C, and D series of the standard diagrams. Where morphology is ambiguous, SEM-EDS per ASTM E2142 confirms the chemistry, for example distinguishing calcium aluminate from titanium-bearing inclusions.
Can two-dimensional results predict the largest inclusion in the whole plate?
Not deterministically, but statistically. By applying extreme-value statistics to the largest inclusions measured over a known total inspection area, laboratories can estimate the maximum inclusion size expected at a defined confidence level for a larger surface or volume. This estimate is routinely used in fatigue and bearing-quality assessments and is increasingly requested for high-duty pressure vessel applications.
What sample size and inspection area are required?
Requirements depend on the governing standard and specification. Chart rating methods typically examine at least the number of 0.50 mm² fields needed to cover approximately 160 mm² or the area stated in the specification, while image analysis methods commonly survey 100 to 300 mm² or more, depending on inclusion density and the statistical confidence required for distribution and extreme-value analysis. For critical components, multiple locations spanning surface, quarter-thickness, and mid-thickness planes are recommended.
How does inclusion distribution relate to wet H2S service performance?
In sour environments, atomic hydrogen enters the steel and recombines at inclusion interfaces, especially elongated manganese sulfides and hard oxide clusters, producing hydrogen-induced cracking and stepwise cracking. A refined, uniformly dispersed inclusion population with controlled sulfur and effective calcium treatment measurably lowers cracking susceptibility, which is why inclusion distribution analysis is often paired with NACE-type hydrogen-induced cracking tests during material qualification for such service.
