This paper investigates the cause of premature failures in certain Grade 91 steel components used in UK power plants. The failures were linked to both low material hardness and specific chemical compositions that fell within ASTM specifications but had a low nitrogen-to-aluminum ratio (N:Al). The investigators examined eight material batches, including those involved in failures, new stock, and in-service components with similar properties. Testing confirmed these materials had lower creep resistance compared to standard Grade 91 steel. Microscopic analysis revealed the presence of large aluminum nitride precipitates, which limited the formation of beneficial vanadium nitride precipitates, leading to reduced creep strength. These findings suggest that even within the ASTM specification limits, a low N:Al ratio can negatively impact the performance of Grade 91 steel.

Dissimilar metal welds between T91 ferritic steels and TP347H austenitic alloys are commonly used in fossil power plants in China. Premature failure of such dissimilar welds can occur, resulting in unplanned plant outages that can cause huge economic losses. In this article, microstructural evolution of T91/TP347H dissimilar welds after different service conditions were studied, mechanical properties before and after service were also analyzed, a full investigation into the failure cause was carried out. The results show, the dissimilar metal welds in the as-welded condition consists of a sharp chemical concentration gradient across the fusion line, failure is attributed to the steep microstructural and mechanical properties gradients, formation of interfacial carbides that promote creep cavity formation.

A Japanese national project has been undertaken since Aug. 2008 with the objective of developing an advanced ultra-supercritical power plant (A-USC) with a steam temperature of 700°C. Fe-Ni and Ni-based alloys, namely HR6W, HR35, Alloy617, Alloy740, Alloy263 and Alloy141, were taken as candidate materials for piping and superheater/reheater tubes in an A-USC boiler. Weldments of these alloys were manufactured by GTAW, after which long term creep rupture tests were conducted at 700°C, 750°C and 800°C. Weldments of HR6W, HR35 and Alloy617 showed similar creep strength as compared with these base metals. Weldments of Alloy740 tended to fail in the HAZ, and it is considered that voids and cracks preferentially formed in the small precipitation zone along the grain boundary in the HAZ. The creep strength of Alloy263 in weldments exhibited the highest level among all the alloys, although HAZ failure occurred in the low stress test condition. A weld strength reduction factor will be needed to avoid HAZ failure in Alloy740 and Alloy263. Also, to prevent premature failure in weld metal, optimization of the chemical composition of weld filler materials will be required.

Fossil fuels continue to be the primary source of energy in the U.S and worldwide. In order to improve the efficiency of fossil power plants, advanced structural materials need to be developed and deployed to meet the need of high temperature creep resistance and corrosion resistance. Examples include creep strength enhanced ferritic (CSEF) steels, austenitic stainless steels, nickel-based superalloys, and oxide dispersion strengthened alloys. Welding is extensively used in construction of fossil power plants. The performance of the weld region can be critical to the safe and economical operation of fossil power plants. Degradations in performance such as reduced creep strength and premature failure in the weld region (e.g. Type IV failure in ferritic steels) are examples of longstanding welding and weldability problems for boiler and other components. In the past, extensive studies have been carried out to characterize the different microstructures in different regions of a weld, and to a certain extent, to establish the correlations between the microstructure and the creep strength. However, the metallurgical or microstructural induced local stress/strain variations have been seldom quantified. In addition, it has been long recognized that, due to the sharp microstructure and property gradients in the weld and HAZ, the standard creep testing procedure for the base metal can produce erroneous results when used for weld testing.

Cold working and bending during boiler manufacturing can induce strain hardening in austenitic stainless steel, potentially compromising creep ductility and leading to premature failures during operation. While design codes like ASME I, PG 19 provide guidelines for maximum strain levels before solution treating is required, industry concerns suggest these limits may be too high, prompting some boiler manufacturers to implement more conservative thresholds. This study examined the creep ductility of four austenitic stainless steels (TP310HCbN, XA704, TX304HB, and Sanicro 25) at prior strain levels of 12% and 15%, with Sanicro 25 demonstrating the highest ductility, followed by TX304HB, XA704, and TP310HCbN. Solution annealing successfully restored creep ductility to exceed 10% elongation in all materials, though this treatment may be necessary at strains of 12% and 15% for all materials except Sanicro 25 to ensure adequate creep ductility. The findings suggest that ASME I PG 19 guidelines for austenitic stainless steels containing Cb, V, and N should be reviewed, as lower strain limits could help reduce strain-induced precipitation hardening failures.

An innovative additively manufactured gradient composite transition joint (AM-GCTJ) has been designed to join dissimilar metals, to address the pressing issue of premature failure observed in conventional dissimilar metal welds (DMWs) when subjected to increased cyclic operating conditions of fossil fuel power plants. The transition design, guided by computational modeling, developed a gradient composite material distribution, facilitating a smooth transition in material volume fraction and physical properties between different alloys. This innovative design seeks to alleviate structural challenges arising from distinct material properties, including high thermal stress and potential cracking issues resulting from the thermal expansion mismatch typically observed in conventional DMWs. In this study, we investigated the creep properties of transition joints comprising Grade 91 steel and 304 stainless steel through a combination of simulations and creep testing experiments. The implementation of a gradient composite design in the plate transition joint resulted in a significant enhancement of creep resistance when compared to the baseline conventional DMW. For instance, the creep rupture life of the transition joint was improved by > 400% in a wide range of temperature and stress testing conditions. Meanwhile, the failure location shifted to the base material of Grade 91 steel. Such enhancement can be primarily attributed to the strong mechanical constraint facilitated by the gradient composite design, which effectively reduced the stresses on the less creep-resistant alloy in the transition zone. Beyond examining plate joints, it is crucial to assess the deformation response of tubular transition joints under pressure loading and transient temperature conditions to substantiate and demonstrate the effectiveness of the design. The simulation results affirm that the tubular transition joint demonstrates superior resistance compared to its counterpart DMW when subjected to multiaxial stresses in tubular structures. In addition, optimization of the transition joint’s geometry dimensions has been conducted to diminish the accumulated deformation and enhance the service life. Lastly, the scalability and potential of the innovative transition joints for large-diameter pipe applications are addressed.

There is a constant need for improved knowledge of the influence of non-standard processing on the expected performance of creep strength enhanced ferritic (CSEF) materials as the total installed tonnage of these materials is rapidly increasing across the power generation industry. Cr-Mo-V steel grades micro-alloyed with niobium and titanium designed for pressurized equipment operating in the supercritical steam range proved to be very sensitive to relative minor variations in the principal heat treatment parameters time and temperature, when compared to the traditional Cr-Mo-V grades. A key component for successful welds is optimised post weld heat treatment (PWHT). Under certain conditions premature failures of welds can occur when incorrect weld and heat treatment performance result in a reduction of specified mechanical properties and high temperature creep performance, it is therefore of significant importance to have a good understanding of actual material properties for effective operation and plant life studies. This study investigated the effect and impact variations of post weld heat treatment time and temperature on mechanical properties of tungsten inert gas (TIG) and manual metal arc (MMA) welds on Grade 91 pipes from a set of reference samples. This is in preparation of establishing a benchmark set of tests to determine the integrity and expected long-term performance of butt-welds from limited site sample volumes, providing a non-intrusive methodology to identify welds suspected to have received non-standard PWHT cycles on Grade 91 pipework systems.

Critical sections of steam plants and heat-recovery steam generators require materials with enhanced properties such as 9Cr-1Mo steel. Ensuring compliance with specifications for heat treatment, chemical composition, contamination limits, and joint design is crucial to prevent premature failures. This study describes the development of a user-friendly, multi-property nondestructive sensor arrangement to qualify heat-treated 9Cr-1Mo steel. Experimental results demonstrate that correlations between thermal heat treatment and electronic, magnetic, and elastic measurements can determine if T91 steel achieves the necessary microstructure and properties for service. Additionally, rejected parts can be assessed for microstructural issues causing unacceptable properties. The techniques utilize a common electronic setup with different sensors, requiring calibration for specific NDE systems and sensor setups, high-speed data acquisition, and frequency analysis (FFT). Further development on crept and welded samples is recommended to enhance NDE practices for in-service T91 steel conditions.

Solid particle erosion (SPE) and liquid droplet erosion (LDE) cause severe damage to turbine components and lead to premature failures, business loss and rapier costs to power plant owners and operators. Under a program funded by the Electric Power Research Institute (EPRI), nano­coatings are under development for application in steam and gas turbines to mitigate the adverse effects of PE and LPE on rotating blades and stationary vanes. Based on a thorough study of the available information, most promising coatings such as nano-structured titanium silicon carbo-nitride (TiSiCN), titanium nitride (TiN) and multilayered nano coatings were selected. TurboMet International (TurboMet) teamed with Southwest Research Institute (SwRI) with state-of-the-art nano-technology coating facilities with plasma enhanced magnetron sputtering (PEMS) method to apply these coatings on various substrates. Ti-6V-4Al, 12Cr, 17-4PH, and Custom 450 stainless steel substrates were selected based on the current alloys used in gas turbine compressors and steam turbine blades and vanes. Coatings with up to 30 micron thickness have been deposited on small test coupons. These are extremely hard coatings with good adhesion strength and optimum toughness. Tests conducted on coated coupons by solid particle erosion (SPE) and liquid droplet erosion (LDE) testing indicate that these coatings have excellent erosion resistance. The erosion resistance under both SPE and LDE test conditions showed the nano-structured coatings have high erosion resistance compared to other commercially produced erosion resistance coatings. Tension and high-cycle fatigue test results revealed that the hard nano-coatings do not have any adverse effects on these properties but may provide positive contribution.

Dissimilar metal welds (DMWs) between ferritic and austenitic materials at elevated temperatures have long posed challenges for boiler manufacturers and operators due to their potential for premature failure. As the industry moves towards higher pressures and temperatures to enhance boiler efficiencies, there is a need for superior weld metals and joint designs that optimize the economy of modern boilers and reduce reliance on austenitic materials for steam headers and piping. EPRI has developed a new filler metal, EPRI P87, to enhance the performance of DMWs. Previous work has detailed the development of EPRI P87 for shielded metal arc welding electrodes, gas-tungsten arc welding fine-wire, and its application in an ultra-supercritical steam boiler by B&W. This study examines the weldability of EPRI P87 consumables through various test methods, including Varestraint testing (both trans and spot), long-term creep testing (approximately 10,000-hour running tests), procedure qualification records for tube-to-tube weldments between traditional/advanced austenitic steels and creep-strength enhanced ferritic steels, and elevated temperature tensile testing. Macroscopic examinations from procedure qualification records using light microscopy confirmed the weldability and absence of cracking across all material combinations. The findings demonstrate that EPRI P87 is a weldable alloy with several advantages for DMW applications and highlight that specific weld joint configurations may necessitate the use of high-temperature tensile data for procedure qualifications.

Olefin furnaces contain gravity cast U-bend fittings from Fe-Ni-Cr alloys that can experience premature failures due to a combination of harsh service conditions. The fittings undergo steep temperature variations during startup and shutdown, outer diameter (OD) oxidation from furnace flue gases, and inner diameter (ID) carburization from process fluids. As a result, cracking often occurs along large solidification grain boundaries from interconnected networks of carbides and secondary phases. To address these degradation concerns, Wire Arc Additive Manufacturing (WAAM) is being used to produce a functionally graded fitting that provides increased oxidation, carburization, creep, and thermal fatigue resistance. Three welding wire compositions have been designed based on thermodynamic and kinetic modeling techniques to address the appropriate corrosion resistance and mechanical properties needed in the OD, Core, and ID regions of the U- bend fitting cross-section. A Fe-35Cr-45Ni-0.7Nb solid welding wire is being used for the Core section, and metal-cored welding wires based around this composition with additions of Si or Al are being used for the OD and ID sections, respectively. This study involved weldability evaluation focused on understanding the microstructures and potential additive manufacturing printability challenges associated with graded WAAM structures using these welding wires. To achieve this, Cast Pin Tear Testing (CPTT) was performed to evaluate solidification cracking susceptibility of the welding wires. Additionally, Scheil calculations were performed in Thermo-Calc software to predict solidification microstructures. To validate the results, SEM characterization was conducted on cast buttons of each welding wire to identify phases in the respective microstructures. These unique data will help inform WAAM design parameters needed to produce a Functionally Graded Material (FGM) that improves the lifetime of Fe-Ni-Cr U-bend fittings in olefin furnaces.?

This paper summarizes recent efforts to improve creep performance in Grade 91 (Mod. 9Cr-1Mo, ASTM A387) steel weldments via non-standard heat treatments prior to welding. Such heat treatments offer a potential solution for minimizing Type IV failures in creep strength enhanced ferritic (CSEF) steels. A lower temperature tempering (LTT, 650°C) of the 9Cr steels prior to gas tungsten arc welding (GTAW) resulted in improved creep-rupture life at 650°C compared to the samples tempered at a standard condition (HTT, 760°C) before welding. From detailed characterization of precipitation kinetics in the heat affected zone, it was hypothesized that M 23 C 6 carbides in the fine-grain heat-affected zone (FGHAZ) in the LTT sample were fully dissolved, resulting in re-precipitation of strengthening carbides during post weld heat treatment (PWHT). This was not the case in the HTT sample since M 23 C 6 in the FGHAZ was only partially dissolved prior to welding, which caused coarsening of existing M 23 C 6 after PWHT and premature creep failure in the FGHAZ. However, it was also found that the LTT raised the ductile-brittle transition temperature above room temperature (RT). Two different thermo-mechanical treatments (TMTs); two-step tempering and aus-forging/aus-aging, of the modified 9Cr-1Mo steels were attempted, in order to control the balance between creep properties and RT ductility, through control of precipitation kinetics of the M 23 C 6 carbides and/or MX carbo-nitrides. The hardness map of the TMT samples after GTAW and PWHT were evaluated.

For safe operation of thick-walled components for Advanced Ultra Super Critical (A-USC) power plants, detailed knowledge of the creep crack initiation and growth behavior is essential. The high loading and high temperature conditions in an A-USC power plant require, in many cases, the employment of nickel base super alloys. Unfortunately, both manufacturing and nondestructive evaluation (NDE) of thick-walled components (> 50 mm) made of nickel base super alloys are quite challenging. In this paper, one candidate material for such applications, Alloy C-263, was tested for creep and creep crack behavior at 700 °C. Objective of the study was to determine a critical flaw size. In order to establish this size, the duration to achieve the 1%-strain limit at a given load is compared with the time to grow the initial flaw for Δa = 0.5 mm when the component was loaded with the same given load. It will be shown that manufacturing parameters, e. g. heat treatment procedures, have a significant influence on the creep crack initiation and growth behavior and thus on component life. Decoration of grain boundaries with precipitates, for instance caused by the manufacturing process, can reduce the creep crack resistance and thus increase the risk for premature component failure.

The use of creep strength-enhanced ferritic alloys like Grade 91 has become popular in fossil power plants for applications at temperatures above 566°C (1050°F). Compared to Grades 11 and 22, Grade 91 offers higher stress allowables, better ramp rate tolerance, weight reduction, and lower thermal expansion coefficients at operating temperatures. However, Grade 91's superior elevated temperature strength requires specific microstructure and metallurgical considerations. This paper highlights concerns that warrant further investigation. Initial operating stresses in Grade 91 piping systems may exceed 262 MPa (38 ksi), and lack of creep relaxation below 593°C (1050°F) could lead to weldment failures within years, especially above 159 MPa (23 ksi) after one year. While cold spring can reduce initial stresses for systems below 593°C (1050°F), creep relaxation rates up to 206 MPa (30 ksi) need study. Above 593°C (1050°F) and below 103 MPa (15 ksi), weldments may fail prematurely by Type IV creep mechanism. Long-term creep rupture studies on cross-weld and multiaxially loaded thick-walled specimens should evaluate deteriorated weldment properties, particularly below 103 MPa (15 ksi).

Modified 9Cr-1Mo steel (ASTM Gr.91) is widely used in components of fossil fueled power plants around the world today. This grade of steel has however been shown to exhibit significant variations in creep life and creep ductility, which has led to premature in-service failures. The aim of this work is to define potential metallurgical risk factors that lead to this variation in performance. To achieve this, a set of creep test samples that represent a wide range in this variation of creep behavior in this steel grade have been studied in detail. As a first stage in this characterization the macro-scale chemical homogeneity of the materials were mapped using micro-XRF. Understanding the segregation behavior also allows quantification of microstructural parameters in both segregated and non-segregated areas enabling the variations to be determined. For example this showed a significant increase in the number per unit area of Laves phase particles in high compared with low Mo content areas. To study the effect of MX particles on segregation a methodology combining SEM and TEM was employed. This involved chemically mapping the larger V containing particles using EDS in the SEM in segregated and unsegregated areas and then comparing the results to site-specific TEM analysis. This analysis showed that although the average size of the V containing samples is in the expected 0-50 nm size range, these particles in some samples had a wide size distribution range, which significantly overlaps with the M 23 C 6 size distribution range. This together with the segregation characteristics has important implications for determining meaningful quantitative microstructural data from these microstructurally complex materials.

Ferritic 9-12 wt.% chromium steels are commonly used for thick-walled high-temperature components in thermal power plants, but they face two major limitations in high-temperature service. Firstly, a reduction in creep strength occurs after approximately 10,000 hours at service temperatures around 600°C, due to the dissolution of finely dispersed V-rich nitrides and the precipitation of coarse particles of the modified Z-phase, [(Cr,V,Nb)N]. Secondly, welded joints of nearly all ferritic steel grades are prone to premature creep failures in the fine-grained heat-affected zone, known as Type IV cracking, which results from a strength loss of up to 50% compared to the base material. This study describes the development of a 9Cr3W3CoVNb steel with added boron and controlled nitrogen content. Preliminary creep testing results up to 24,000 hours at 650°C show a significant improvement in creep strength compared to established ferritic 9Cr grades like P91 and P92, attributed to a reduced driving force for the precipitation of modified Z-phase particles. Crosswelds of the new 9Cr3W3CoVNbBN steel also demonstrate improved creep behavior at 650°C, with creep rupture strength comparable to the mean base material creep strength of the best commercially available grade P92.

This study investigates a novel approach to addressing the persistent Type IV cracking issue in Grade 91 steel weldments, which has remained problematic despite decades of service history and various mitigation attempts through chemical composition and procedural modifications. Rather than further attempting to prevent heat-affected zone (HAZ) softening, we propose eliminating the vulnerable base metal entirely by replacing critical sections with additively manufactured (AM) weld metal deposits using ASME SFA “B91” consumables. The approach employs weld metal designed for stress-relieved conditions rather than traditional normalizing and tempering treatments. Our findings demonstrate that the reheat cycles during AM buildup do not produce the substantial softening characteristic of Type IV zones, thereby reducing the risk of premature creep failure. The study presents comprehensive properties of the AM-built weld metal after post-weld heat treatment (PWHT), examines factors influencing deposit quality and performance, and explores the practical benefits for procurement and field construction, supported by in-service data and application cases.

In this study, creep rupture behaviors and rupture mechanisms of dissimilar welded joint between Inconel 617B and COST E martensitic steel were investigated. Creep tests were conducted at 600 ℃ in the stress range 140-240 MPa. Scanning electron microscopy (SEM) and micro-hardness were used to examine the creep rupture behaviors and microstructure characteristics of the joint. The results indicated that the rupture positions of crept joints shifted as stress changed. At higher stress level, the rupture position was located in the base metal (BM) of COST E martensitic steel with much plastic deformation and necking. At relatively lower stress level, the rupture positions were located in the fine-grained heat affected zone (FGHAZ) of COST E or at the interface between COST E and WM both identified to be brittle fracture. Rupture in the FGHAZ was caused by type Ⅳ crack due to matrix softening and lack of sufficient precipitates pinning at the grain boundaries (GBs). Rupture at the interface was related to oxide notch forming at the interface.

In this paper, the dissimilar metal welds (DMWs) between 617B nickel-based alloy and 10%Cr martensitic heat-resistant steel filled by 617 filler metal was studied, focused on the high temperature creep rupture properties. The high temperature creep rupture properties of welded joints with different welding processes were tested, and the microstructure of welded joints before and after the creep rupture test was observed by OM and SEM. The results showed that, there were three failure modes: base metal failure, type W failure and interface failure, among which interface failure caused the most serious life reduction. The welded joints using ER NiCr-3 filler metal reduced the strain concentration at the interface, so the fracture location shifted from the interface to HAZ of 10%Cr martensitic heat-resistant steel under high temperature and low stress conditions, and creep rupture life was improved. Similarly, weld cap shifted the creep crack propagation path by changing the groove form, so as to altered the stress state of joint and prolong the creep rupture life.

At present there is no recognized standard test method that can be used for the measurement of the tensile properties of additively manufactured lattice structures. The aim of this work was to develop and validate a methodology that would enable this material property to be measured for these geometrically and microstructurally complex material structures. A novel test piece has been designed and trialed to enable lattice struts and substructures to be manufactured and tested in standard bench top universal testing machines and in small scale in-situ SEM loading jigs (not reported in this paper). In conjunction with the mechanical tests, a finite element (FEA) modelling approach has been used to help cross validate the methodology and results, and to enable larger lattice structures to be modelled with confidence. The specimen design and testing approach developed, is described and the results reviewed for AlSi10Mg.