Experimental evaluation of localized creep deformation in grade 91 steel weldments

https://doi.org/10.1016/j.msea.2020.140356Get rights and content

Abstract

Spatially resolved measurement of localized creep deformation in heterogeneous creep resistant steel weldments is crucial but challenging for lifetime assessments of critical steam components in power plants. In this work, experimental approaches were established to quantitatively evaluate commonly observed localized creep deformation in multi-pass Grade 91 steel weldments. An in-situ digital image correlation (DIC) system was utilized with a creep testing frame to monitor and measure both full-field strain and localized strain accumulation across the weldments during long-term creep testing at elevated temperatures. The in situ DIC method measured not only the creep deformation behavior of the weld metal, heat affected zone (HAZ), and base metal, but also creep strain evolution for each sub-region within the HAZ itself, including coarse-grained HAZ, fine-grained HAZ (FGHAZ), and intercritical HAZ (ICHAZ). The DIC results revealed that local creep strain in the ICHAZ reached up to 90% strain before final rupture, whereas nominal creep strain measured by the standard extensometer of the tested cross-weld specimen was below 10%, indicative of Type IV cracking of the Grade 91 weld. Microstructural analyses revealed that the faster creep degradation/deformation in the HAZ was caused mainly by accelerated matrix grain recrystallization/growth and a reduced pinning effect from the segregated and coarsened precipitates in the FGHAZ and ICHAZ. The ultimate creep rupture occurred in the ICHAZ owing to its lowest creep resistance induced by the largest recrystallized grain size, the lowest fraction of coincidence site lattice, and the lowest local strain energies/dislocation densities.

Introduction

The 9Cr creep strength–enhanced ferritic (CSEF) steels, such as Grade 91 steel or modified 9Cr-1Mo-VNb steel, are one of the most critical structural materials used in modern ultra-supercritical fossil fuel–fired power plants worldwide [[1], [2], [3]]. The structural integrity of the welded components in the steam boilers, such as headers and main steam pipes, has been one of the greatest challenges of using these 9Cr CSEF steels with increased steam temperatures (>625 °C) and pressures (>30 MPa) [2,4]. The creep deformation and rupture mechanisms of weldments have not been well understood owing to the internal varables of parent metal composition, microstructure, creep resistance, and constraint conditions across welds. The external factors, including weld geometry, residual stress, loading cycles, and service environments, cause more complexities. Creep lifetime assessment of those welded components requires local creep property data of the most vulnerable locations, for example, the heat-affected zone (HAZ) or the weld metal (WM) [5]. Uniaxial creep testing with extensometers is a standard method of evaluating the creep properties of base metals (BMs) with relatively homogeneous microstructures and creep deformation behaviors [6]. It is also widely adopted in cross-weld creep testing, in which the gauge length of the extensometer covers various regions, including the BM, HAZ, and WM, typically in the range of tens or hundreds of millimeters in gauge length. Although it is a reliable representation of the deformation behavior of a BM with uniform microstructure and thereby deformation responses, the creep strain measured by extensometer is not a faithful measure of creep deformation in individual regions of a weld in a cross-weld creep test. The reason is the highly non-uniform microstructure and associated non-uniform creep deformation behavior in different regions of a weldment. Indeed, many failures of welded 9Cr CSEF steel components are reported to be related to localized deformation in the HAZs of weldments [3,[7], [8], [9], [10]], which cannot be reliably determined by standard creep testing with an extensometer. For component design and structural integrity management, a reliable and accurate prediction of the creep lifetimes of those welded components is crucial [3]. Many studies [5,11] have attempted to simulate the creep behavior of Cr-Mo steel welds using simplified BM-HAZ-WM models. These simulations normally predict larger creep deformation in the HAZ, but the results need more refinement by consideration of more microstructural factors. It is believed that modeling of the creep deformation of individual HAZ subzones will enhance the reliability and accuracy of model prediction, which also needs further validation with experimental data.

Premature cracking caused by accelerated creep deformation in the fine-grained HAZ (FGHAZ) and intercritical HAZ (ICHAZ) has been frequently reported as the infamous Type IV cracking. Therefore, this localized creep deformation across weldments should be quantified not only for the WM, HAZ, and BM but also, more importantly, for each sub-region within the HAZ. Park and Stratford [12] attempted to estimate the localized creep strain in Type IV cracking by measuring the grain shape change in cross-weld specimens of 1.25Cr-0.5Mo steel after creep tests. These post-test estimations show higher strain in the HAZ, but the limited spatial resolution and accuracy of this method needs further improvement to obtain strain distribution data within the HAZ itself. In recent years, the nondestructive and noncontact technique of digital image correlation (DIC) has become popular as an in situ, reliable, accurate tool to measure strains in homogeneous or heterogeneous materials [5,[13], [14], [15], [16], [17]]. The DIC technique is widely used for strain measurement during tensile testing at ambient temperature, but application of DIC in long-term creep testing at elevated temperatures is quite challenging because of the oxidation and other stability issues of the critical speckle pattern during DIC image acquisition [18]. A few early studies have been attempted to measure creep deformation in 316H stainless steel welds [5] and dissimilar metal welds [16]. Their results show the high potential of DIC in creep strain measurement at elevated temperatures. Global creep strain measured by DIC shows agreement with measurements from a conventional extensometer. However, to acquire creep data for an individual coarse-grained HAZ (CGHAZ), FGHAZ, and ICHAZ remains challenging owing to the microstructure complexity across the HAZ and the limited resolution of the DIC setup for providing localized measurements.

In this work, we applied a purpose-built creep testing system with in situ full-field creep strain measurement capability by means of high-temperature DIC to evaluate and distinguish the localized creep deformation in different sub-regions of Grade 91 steel weldments. The in situ full-field DIC system has submillimeter spatial resolution, making it possible to resolve the highly non-uniform, localized creep deformation in subzones of the HAZ of a Grade 91 steel weld during cross-weld creep rupture tests. Detailed microstructural changes across the entire weldments before and after creep testing were characterized with advanced microscopy techniques and used to correlate the local creep strength degradation and underlying microstructure features.

Section snippets

Materials and experiments

In this study, Grade 91 steel (plate) (ORNL Heat No. 30176) was used as the BM. The chemical composition of the BM is tabulated in Table 1. The BM was normalized at 1050 °C for 1 hour and tempered at 760 °C for 2 hours. Gas tungsten arc welding (GTAW) with a 0.045-inch filler wire of AWS ER90S-B9, a matching filler metal for Grade 91 steel, was used to join the 1-inch (25.4 mm) thick plates with a U groove. The weld configuration is shown in Fig. 1. The weld had a dimension of 152.4 mm

Nominal creep deformation measurement

The nominal creep strain and strain rate curves over the 70-mm gauge length, measured by both LVDT extensometer and DIC, are plotted in Fig. 3. The red line in the figure is from the DIC measurement using the same 70-mm gauge length, which is essentially indistinguishable from the LVDT measurement. This cross-weld specimen failed at a creep time of 410 h with a nominal creep strain of 8%, which is consistent with reported creep failure strains of Grade 91 welds indictive of Type IV cracking [8,

Discussion

For a reliable creep lifetime assessment of these welded components, knowledge of the local creep properties across subregions in weldments, including the BM, HAZ, and WM, is essential. For Type IV cracking, creep resistance variations among the CGHAZ, FGHAZ, and ICHAZ bring more complexities. It is impossible to measure the property data for each individual region with conventional uniaxial creep tests. Ex situ deformation measurements across creep-ruptured specimens is challenging because of

Conclusions

In this study, it is demonstrated that localized creep deformation/degradation occurring in the HAZ of Grade 91 steel weldments can be monitored and measured in real time using an in situ DIC system. The correlation between the measured localized deformation and detailed microstructural analyses provided insight into creep deformation and rupture mechanisms in 9Cr creep-resistant steel weldments. The DIC creep strain measurements revealed that creep resistant of sub-regions across the weld

Author contribution

Yiyu Wang: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing - original draft, Writing - review & editing. Wei Zhang: Investigation, Writing - review & editing. Yanli Wang: Investigation, Discussion, Writing - review & editing. Yong Chae Lim: Methodology, Discussion. Xinghua Yu: Methodology. Zhili Feng: Conceptualization, Formal analysis, Writing - review & editing, Funding acquisition, Supervision.

Data availability statement

The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

This work is funded by -the US Department of Energy Crosscutting Research program (FWP-FEAA118). The research and development work was performed at Oak Ridge National Laboratory, which is managed by UT-Battelle LLC for the US Department of Energy under contract DE-AC05- 00OR22725. The authors would like to thank Mr. Doug Kyle for his help in creep testing experiments, and Mr. Roger Miller and Dr. Jian Chen for their discussion and manuscript review.

References (31)

Cited by (18)

  • Remaining lifetime assessment of 141,000-hour service-aged forge 91-pipe 91 steel header welds with novel in-situ digital image correlation

    2022, Materials and Design
    Citation Excerpt :

    Conventional extensometers can only measure nominal creep strain over a gauge section and are thus incapable of providing information relevant to local strain accumulation in the HAZ or its subregions. Recent research [21,22] shows the nondestructive and noncontact technique of digital image correlation (DIC) provides a potential solution to overcome this technical difficulty. Application of this DIC approach to welds at the component-level may enhance current or future lifetime assessment methodologies.

  • Short-term creep strain localization in 12%Cr turbine steel under biaxial stress states at 600 °C via digital image correlation

    2022, Materials Science and Engineering: A
    Citation Excerpt :

    When these microstructural changes develop into large-scale structural damage, the material fails due to formation of micro-cracks within grains and cavitation at grain boundaries (accelerated stage). Recently, digital image correlation (DIC) as an in-situ and contactless measurement approach has been successfully adopted to characterize the full field strain evolution of materials during creep deformation [11–17]. Gariboldi et al. analysed anisotropic damage mechanisms in forged Al–Cu–Mg–Si alloy and found the small scale yielding during the creep fracture process under 150 °C using DIC [12].

  • Damage behaviour in different micro-regions of P92 steel weldment under sequential low cycle fatigue and creep loading

    2021, Engineering Fracture Mechanics
    Citation Excerpt :

    However, as a result of substructures of blocks and packets, the measured average grain size of the WM is only about 1.62 μm, and the frequency of HAGBs is about 29.2%. It is important to note that the grain size discussed here is not the prior austenite grain size but the size of body-center-cubic grain that transformed from the tempered martensite [16]. The CGHAZ next to the WM also presents a coarse grain structure with a small portion of tempered recrystallized martensite laths (the average grain size: 1.76 μm; frequency of HAGBs: 34.3%), as shown in Fig. 3(b).

View all citing articles on Scopus

Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

1

Dr. Xinghua Yu is currently at Department of Materials Processing Engineering, Beijing Institute of Technology, Beijing, China.

View full text