Computer-Assisted Ultrasonic Analysis of Overlay Weld Delamination in Hydrogenation Reactors

Literature Overview

This 1998 study by Li Xiaogang, Fu Dongmei, Meng Qinghai, and Ke Wei, published in Petrochemical Corrosion and Protection (Vol. 15, No. 3, pp. 54-57), addresses a critical and recurring challenge in petrochemical equipment integrity management: the progressive delamination of overlay weld layers in hydrogenation reactors. The authors developed a practical computer analysis software for evaluating ultrasonic testing (UT) results of overlay weld delamination, and presented findings from a multi-year monitoring campaign. The work bridges metallurgical defect evolution with engineering inspection practice, offering both methodological and practical value.

Core Technical Context

Hydrogenation reactors in refineries operate under extreme conditions — high temperature (typically 350-450°C), high pressure (30-160 MPa), and hydrogen-containing atmospheres. To resist corrosion, these vessels are internally protected with overlay weld layers, commonly using austenitic stainless steel grades such as 304L, 316L, or 321 deposited by submerged arc welding (SAW) or manual metal arc welding (SMAW). The overlay layer must maintain metallurgical and mechanical integrity over the vessel's design life, typically 20-25 years.

Delamination between the overlay weld and the base steel substrate represents one of the most dangerous failure modes. Unlike surface corrosion, delamination is a subsurface defect that can propagate silently until catastrophic separation occurs, exposing the carbon steel base material directly to the corrosive hydrogen environment. This makes early detection and trend monitoring essential for risk-based inspection strategies.

Methodology and Key Findings

The authors developed dedicated software for analyzing UT signals from overlay weld delamination. The methodology involved:

• Standardized UT scanning procedures on the reactor inner surface, with the overlay layer serving as the acoustic interface
• Digitization of UT signals and automated defect characterization (size, location, and severity classification)
• Longitudinal tracking of defect populations across multiple inspection campaigns

The following table summarizes the principal findings:

These findings are significant because they suggest that overlay delamination in hydrogenation reactors follows a population-based degradation model rather than a single-defect-dominated model. The stabilization of individual defect sizes implies that the mechanical driving force for crack propagation is balanced by residual compressive stresses in the overlay, or by the toughness of the overlay/base interface.

Engineering Practice Implications

For engineers responsible for hydrogenation reactor integrity, several practical insights emerge:

1. Inspection frequency optimization: Since defect initiation rather than growth drives delamination severity, inspection intervals should be calibrated to detect new defect formation rather than monitoring individual defect growth. This supports the use of risk-based inspection (RBI) methodologies where inspection intervals are adjusted based on defect population trends.
2. Acceptance criteria refinement: The observation that defects stabilize at a certain size suggests that a maximum allowable defect area criterion may be more appropriate than a zero-defect approach. However, this must be balanced against the risk of undetected defect coalescence.
3. Software-assisted UT analysis: The development of dedicated analysis software addresses a real pain point in industrial inspection — the manual interpretation of large volumes of UT data is time-consuming and prone to inconsistency. Automated defect characterization improves reproducibility and enables trend analysis over multiple campaigns.

Key Questions and Reflections

The study raises several questions that remain relevant in modern practice:

• What is the metallurgical mechanism behind the arrest of individual defect growth? Is it related to hydrogen-induced cracking (HIC) behavior in the base steel, or to the fatigue behavior of the overlay weld interface?
• How does the defect population evolve beyond the 3-year observation window? Long-term data would be needed to validate the saturation hypothesis.
• Can the defect initiation rate be correlated with operating parameters such as temperature, pressure, and hydrogen partial pressure?

The work, though published in 1998, remains methodologically sound. The transition from manual UT signal interpretation to computer-aided analysis prefigured the modern practice of phased array ultrasonic testing (PAUT) with automated defect characterization. The fundamental insight — that delamination is driven by defect multiplication rather than individual defect growth — should inform current RBI frameworks for hydrogenation reactors.

Summary

This study provides a valuable contribution to the field of hydrogenation reactor integrity assessment by introducing computer-assisted UT analysis for overlay weld delamination and by identifying a population-driven degradation mechanism. The finding that individual defects stabilize in size while the overall defect population grows offers a new perspective on delamination behavior that challenges the conventional assumption of progressive crack growth. For engineers managing reactor integrity, the work underscores the importance of systematic trend monitoring and the value of automated data analysis in supporting informed inspection and repair decisions.