Issue 78

A. A. Frantsuzov et alii, Fracture and Structural Integrity, 78 (2026) 474-488; DOI: 10.3221/IGF-ESIS.78.27

graph is constructed from crystallographically compatible potential parent orientations of neighboring martensitic grains. The graph is then clustered and refined by merging similarly oriented reconstructed regions and small inclusions, resulting in a reconstructed prior austenite grain map from which martensitic variants and packets can be calculated [14,15]. To observe the PAGs optically, polished samples were chemically etched using a freshly prepared picric acid solution at room temperature with the addition of surfactants. The sample was placed in the etchant heated to 70 – 80 ℃ for approximately 5 min. During the process, the surface of a sample was periodically gently polished using colloidal silica to prevent grains from becoming excessively etched to reveal the boundaries. The resulting PAG microstructure was investigated using optical microscope ZEISS Axio Vert.A1 with Thixomet [16] software installed for image capturing and linear dimensions calculation. Different approaches were used in the present study to evaluate PAGS on the basis of optical images: − direct measurement of selected grains, in which one linear dimension was measured for each identified PAG. All measurements were performed in the same direction, and the resulting dataset was treated as a list of individual grain-size values used to describe the PAGS distribution and calculate the mean; − the lineal intercept method, in which test lines of known total length were placed on the micrographs and the number of intersections with PAG boundaries was counted. In the classical ASTM E112 [8] Heyn lineal intercept procedure, the mean intercept length is evaluated by dividing the total length of the test lines by the number of boundary intersections. However, the image-analysis software used in this work also allowed the individual chord segments between adjacent boundary intersections to be extracted. Therefore, these chord-length data were used for additional statistical analysis of the PAGS distribution; − comparison with standard reference plates, performed according to ASTM E112 [8]. In this approach, the observed PAG structure was visually compared with standard grain-size charts, after which the corresponding ASTM grain size number was converted into an average grain diameter using the tabulated values provided in the standard.

R ESULTS

T

he microstructure of the C110-grade steel consists of tempered martensite, as shown in Figure 2. This microstructure was revealed by etching the sample in 4 % nitric acid. The same sample was then repolished to a mirror finish and etched with a picric acid solution to reveal the prior austenite grain (PAG) structure, shown in Figure 3. Although the PAG boundaries are clearly visible at first sight, closer inspection shows that some boundaries are poorly defined and cannot be assigned unambiguously to individual grains.

Figure 2: The tempered martensite microstructure of C110 grade steel revealed by nitric acid etch.

In addition to the diffuse PAG boundaries, larger scale dark banded regions were observed in the etched microstructure (Figure 3). These bands are associated with chemical segregation formed during casting and retained after rolling. Compared with the surrounding matrix, such regions are enriched in alloying elements, as shown in Figure 4 and described in several other works [17,18], which typically promotes the formation of finer austenite grains. Figure 5 shows a higher magnification view where the boundaries of a darker band are indicated by arrows. This effect is not uniform through the pipe wall thickness and is more pronounced near the inner surface. Furthermore, an increased alloying-element content promotes carbide precipitation during heat treatment, which reduces the etching contrast of PAG boundaries and complicates their detection.

478