Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >

Environmentally Assisted Cracking in Duplex Stainless Steel Welds

Elucidating the process of environmentally assisted cracking in duplex stainless steel weld microstructures

TWI Core Research Project 1124/2019

Overview

Use of duplex stainless steels (DSS) in the energy and petrochemical sectors has increased significantly in recent years due to several advantages the material has over austenitic stainless steel grades, including their high mechanical strength, and localised corrosion and stress corrosion cracking resistance in chloride containing environments.

However, a better understanding is needed of the effect of different phases in DSS parent/weld heat affected zones on localised corrosion in chloride-containing environments and when hydrogen charged. High resolution microstructure and electrochemical characterisation methods were used.

Approach

Various tools were applied to simulate heat affected zone microstructures, and identify the inherent anodic and cathodic behaviour of different phases during environmental exposure.

First, procedures were defined to produce representative, simulated heat affected zones, equivalent to a range of heat inputs in 2205 duplex stainless steel pipe (38mm wall thickness), using a GLEEBLE 3500 thermo-mechanical simulator. This included the formation of third phases and precipitates that might accompany the reduction in austenite content, compared with the starting parent material. A range of techniques was applied including electron backscatter diffraction, scanning Kelvin probe force microscopy, and high-resolution energy-dispersive X-ray microanalysis.

The simulated heat affected zones were then characterised, and the effects of the thermal cycles  on the austenite morphology investigated. The effects of elemental partitioning on the electrochemical properties of a simulated heat affected zone were also investigated.

A multi-pass TIG/GTA weld was made in the form of a girth weld in a 2205 DSS pipe (406mm OD, 9.2mm wall thickness), and used to assess the influences of different austenite morphologies on the hydrogen-induced stress cracking (HISC) behaviour.

Figure 1. Chromium (II) nitrides at the ferrite-to-ferrite phase boundaries. Overlapped, nitrogen elemental mapping with windowless EDX
Figure 1. Chromium (II) nitrides at the ferrite-to-ferrite phase boundaries. Overlapped, nitrogen elemental mapping with windowless EDX

Outcomes

The work undertaken in this project produced the following findings:

  • The simulated high-temperature, heat affected zones corresponding to different heat inputs revealed the progressive evolution of the microstructure, with slower cooling, equivalent to higher heat input, promoting coarsening of the austenite plates and a reduction in intragranular austenite content.
  • Chromium nitrides were found in ferrite in the simulated high temperature, heat affected zones (Figure 1), although their appearance is normally associated with faster cooling rates, i.e. low heat input and reduced austenite content. Despite the advantage of a coarse austenite morphology, the improved properties may be compromised by Cr2N, potentially offsetting any benefit.

  • changes and chemical partitioning in the high temperature, heat affected zone that have an influence on the electrochemical properties. In particular, a reduction of potential difference between the ferrite and austenite was measured, suggesting that the effectiveness of any galvanic protection of ferrite by austenite is reduced in a heat affected zone.

  • HISC in DSS heat affected zones is strongly influenced by the ferrite morphology, with intragranular and grain boundary austenite offering limited HISC resistance (Figure 2).

  • Coarser austenite grains are beneficial for HISC resistance, but austenite can also become embrittled by high concentrations of hydrogen, e.g. in environments in which hydrogen recombination poisons are present and/or high charging currents are employed.

  • There are different experimental techniques available with which to evaluate the environmental properties of DSSs. The methodology used strongly influences the observations, which are not always transferable between different methods.

 

Conclusion

Conventional methods of characterising DSS heat affected zone microstructures, particularly a focus on ferrite volume fraction and intermetallic phases, do not capture all features that influence heat affected zone environmental properties.

The range of environmental properties measured in the simulated heat affected zones confirms the logic of restricting welding conditions for DSSs, to control the range of resulting properties. In order to better define the relationship between welding procedure variables and properties, there is a need for study beyond the typical arc energy windows used in industry. For such studies, the characterisation methods developed here may be used to measure the fundamental properties of the various heat affected zone microstructures developed.

 

This project was funded by TWI’s Core Research Programme

Figure 2. Quasi-in-situ scanning electron microscopy straining of specimen taken form a TIG welded pipe. Detail of the heat affected zone with crack nucleation at 6% strain
Figure 2. Quasi-in-situ scanning electron microscopy straining of specimen taken form a TIG welded pipe. Detail of the heat affected zone with crack nucleation at 6% strain
Media /

Related Insights

View all insights
}