Electron beam welding (EBW) is a promising technology for the joining of large steel components such as pressure vessels. Notably, EBW does not use a filler metal, meaning that the weld and parent material (PM) should have the same composition. Accordingly, it may be possible to homogenise microstructure and properties across the weld given a suitable post-weld heat treatment (PWHT), essentially making the weld ‘disappear’. To investigate this possibility, three different PWHTs were applied to an SA508 Grade 4N weld. A stress-relief PWHT only had a relatively minor impact on the variations in microstructure and hardness across the fusion zone (FZ), heat-affected zone (HAZ) and PM versus the as-welded state, and compositional microsegregation in the FZ was unchanged. Hardness was successfully homogenised across the parent and weld in heat treatments involving a traditional 860 °C austenitisation, quench and temper, but compositional microsegregation persisted in the FZ. The addition of a homogenisation step at 1200 °C eliminated this microsegregation, with the weld area only remaining visible owing to the presence of compositional banding in the PM. However, homogenisation also led to the formation of undesirable prior austenite grain structures following subsequent re-austenitisation at 860 °C. Nevertheless, the application of PWHTs involving a full re-austenitisation seems promising for future applications.

Recently, iron-based alloys with a π-ferrosilicide phase have emerged as potential alternatives to cobalt-based hardfacing alloys. Here, we present the development of two π-ferrosilicide containing alloys: one with a ferritic matrix and the other with a ferritic-austenitic matrix. In the as-cast condition, both alloys revealed fine Ni- and Si-rich coherent cubic shaped D03 precipitates in the BCC matrix. The π-ferrosilicide phase was found to have an orientation relationship with the ferrite phase, nucleating within ferrite matrix and from ferrite grain boundaries. In contrast to carbide-strengthened hardfacing Fe-alloys, here the dissolution of the π-ferrosilicide phase at 1200°C enables easy thermomechanical processing of these alloys, which results in refinement of the π-ferrosilicide and additional formation of χ-phase precipitates in the ferrite. Nano-scratch tests provided evidence of a resilient silicide-ferrite interface, likely to due to it possessing some coherency. Both alloys also displayed compressive strengths approaching 2 GPa and ductility in compression of approximately 25%. The combination of processability and attractive mechanical properties suggests that these alloys have the potential to serve as alternatives to carbide-reinforced hardfacing Fe-alloys.

Underclad cracking in nuclear pressure vessels was of significant concern in the 1970s and 1980s before mitigating adjustments were made both to steel compositions and manufacturing practice. Unfortunately, the cracking mechanisms are still not well understood, and this can undermine confidence when changes to cladding operations are under consideration. In this work, Charpy-sized test coupons of 18MND5 steel were subject to a range of thermal cycles that can be experienced by the substrate immediately adjacent to the interface with the overlay. The Charpy test results are considered in combination with analysis of the samples through field emission gun (FEG) scanning electron microscopy (SEM) and wavelength dispersive X-ray spectroscopy (WDXS). The findings suggest that the heat affected zone subject to the coarse grained plus intercritical thermal cycle, before post weld heat treatment, has a peculiar and potentially brittle microstructure. However, in a steel with no obvious macrosegregation, no clear evidence of intergranular cracking in the early stages of the post weld heat treatment could be found. The second part of the work focuses on the effects of segregation regions in large forgings, more specifically ghost lines, on the tendency for underclad cracking. Ghost line regions that subsequently form a coarse grained heat affected zone (CGHAZ), and are then heated to just under the A1 temperature, seem to be the most prone to intergranular cracking. Cracking susceptibility appears to be associated with elevated concentrations of alloying and impurity elements at grain boundaries.

Directed energy deposition via electric arc (DED-Arc) and wire-feed system can offer a low-cost, high deposition rate method to produce large-scale, near-net-shape structural components from Inconel 718 (IN718). In this work, DED-Arc with cold metal transfer (CMT) has been compared with pulsed-spray droplet transfer (GMAW-P) using different build strategies by correlating the local cyclic thermal histories and the position of deposition interruptions to the resulting tensile properties. The measured heat input (0.18 - 0.59 kJ/mm) and inter-pass temperature (80 - 650 degrees C) affected the thermal field and residence times in the 900 - 600 degrees C temperature range, where & gamma;'/& gamma;" phases are more likely to form, thereby impacting the yield strength along the build direction (370 - 460 MPa). Furthermore, CMT deposition with a lower travel speed (0.4 m/min vs. 1.0 m/min) prevented the onset of solidification defects, contributing to more consistent tensile ductility. Tensile properties of material surrounding build interruptions developed strain localisation and failed prematurely as compared to regions without, due to localised cracking and transient thermal fields generated once the deposition resumed.

Directed energy deposition via electric arc (DED-Arc) and wire-feed system can offer a low-cost, high deposition rate method to produce large-scale, near-net-shape structural components from Inconel 718 (IN718). In this work, DED-Arc with cold metal transfer (CMT) has been compared with pulsed-spray droplet transfer (GMAW-P) using different build strategies by correlating the local cyclic thermal histories and the position of deposition interruptions to the resulting tensile properties. The measured heat input (0.18 -- 0.59 kJ/mm) and inter-pass temperature (80 – 650 °C) affected the thermal field and residence times in the 900 -- 600 °C temperature range, where γ'/γ" phases are more likely to form, thereby impacting the yield strength along the build direction (370 – 460 MPa). Furthermore, CMT deposition with a lower travel speed (0.4 m/min vs. 1.0 m/min) prevented the onset of solidification defects, contributing to more consistent tensile ductility. Tensile properties of material surrounding build interruptions developed strain localisation and failed prematurely as compared to regions without, due to localised cracking and transient thermal fields generated once the deposition resumed.

The oxidation state and surface properties of powder particles play a major role in the final properties of powder manufactured components. In the present study, the coating of a non-stainless low alloy (SA508 Grade 3) steel powder was explored to protect it from progressive oxidation while also studying the effects on powder flowability and electrical charging. The protective coating was applied by magnetron sputtering of chromium. The surface chemistries of both as-received and Cr coated powders were studied using X-ray photo electron spectroscopy (XPS). Accelerated oxidation tests were carried out on both uncoated and Cr coated powders to study the effects of coating on oxidation resistance. Hard X-ray photoelectron spectroscopy (HAXPES) analysis was used to measure oxygen pick up near the surface, showing significant reductions for the case of the Cr coated powder. The conductivity of the powder was found to increase with Cr coating. The flowability of the powder was characterised by the tapped density, the angle of repose (AOR) and a powder rheometer, and it was found to improve with a Cr coating, which can be attributed to reduced tribo-electrical charging and reduced cohesivity of the powder particles.

This article describes three areas that, in the opinion of the author, should be priorities for development in the nuclear sector. The application of electron-beam (or laser in vacuo) welding to pressure vessel fabrication, the optimisation of filler metal compositions for multipass steel welds, and the pre-fabrication of dissimilar metal transition pieces are each highlighted and discussed. While conceived with the nuclear sector in mind, these priority areas are relevant more generally to the welding of pressure vessels, piping, and similar structures. The intention is to stimulate ideas, to provoke debate, and to encourage the welding community to rise to the challenges that must be overcome as we transition to a low-carbon future.

In this work a novel mathematical framework, that fully describes the fusion and vapourisation state transitions in multi-component systems, has been applied to assist in understanding the fundamental mechanisms of defect formation and chemical homogenisation in the laser powder bed fusion process (L-PBF). Specifically, the role of vapourisation and condensation of the multi-component metallic substrate is investigated to determine the importance of properly capturing the state transitions when understanding the substrate evolution. The framework is applied to a ternary metallic system; it is revealed that entrained vapour bubbles in chemically dissimilar flows promote greater homogenisation during the condensation and collapse of these bubbles when compared to non-condensing phases. It is further shown that as the laser power density is increased, there is a greater tendency for preferential element evaporation of the lighter elements; this preferential element evaporation is quantified numerically for the first time, and shown to be a non-linear function of power density.

This study investigates the extent to which the build orientation and heat treatment schedule affect the micro-structure and mechanical properties for thin-walled and additively manufactured IN718 components produced with the cold metal transfer process (CMT-WAAM). Uniaxial tensile tests using digital image correlation (DIC), microhardness analysis and fractography were used to characterise the mechanical behaviour, both in the as-deposited condition and after heat treatments used in the aerospace and oil and gas industries. Wrought material was also tested to benchmark the measured properties. The solution treatment of 1040 degrees C for 1 h in the oilfield specification reduced the area fraction of Laves phases significantly (similar to 80%) and promoted higher ho-mogenisation of ageing constituents. However, grain growth near interlayer boundaries resulted in localised low hardness (similar to 50 HV0.2 below the average) after age hardening. In the as-deposited condition, the yield strength was similar to 10% lower along the build direction and changed to being 8-13% higher along the same direction in heat-treated samples, while the elastic modulus relative to deposition orientation was unaffected. Furthermore, solidification defects, such as porosity and hot cracking, caused strain localisation during tensile testing and substantial scatter in macroscopic strain. The ductility was improved with oil and gas heat treatment, but it was substantially lower along the build direction due to the longer axis of defects being perpendicular to loading direction. This study highlights the importance of optimising process parameters to minimise defects and tailoring heat treatments to achieve a higher ductility in IN718 processed by WAAM.

A potential connection between the residual stresses generated by a cladding operation, the thermal stresses generated during post weld heat treatment (PWHT), and the propensity for underclad cracking was investigated. As-deposited residual stresses were measured at ambient temperature, using the contour method and neutron diffraction. Measurements were performed in 20mm thick steel substrates that were partially clad with AISI grade 309 (layer 1) and grade 308 (layer 2) stainless steels, using the submerged-arc welding process. Neutron diffraction measurements were also performed after heating the clad samples to 200 and 325°C. The mechanical properties in each region within the steel heat affected zone (HAZ) were determined, at those same temperatures, by subjecting coupons to simulated cladding thermal cycles and then to tension tests. Substantial tensile residual stresses in the HAZ were measured after cladding. These persisted during subsequent heating to 325°C: a temperature that is approximately at the mid-point of the heating ramp in a typical PWHT cycle. The tensile results, however, indicated that each microstructural zone within the HAZ is likely to retain sufficient ductility to prevent fracture, and that underclad cracking is unlikely to occur during the early stages of the PWHT ramp.

Reduced-pressure electron beam (EB) plate butt welds were manufactured in two low-alloy pressure-vessel steels, SA508 Gr 3 Cl 1 and SA508 Gr 2, at two thicknesses in both steels, 30 mm and 130 mm. Transient temperatures during welding were recorded using thermocouple arrays. Residual stresses in the as-welded condition and after post-weld heat treatment were measured using diverse methods: neutron diffraction and the contour method at 30 mm thickness; and deep hole drilling and the contour method at 130 mm. Incremental centre hole drilling measurements were performed at 130 mm thickness to better understand near-surface stresses. Weld and heat-affected zone microstructures and microconstituents were evaluated using a combination of hardness mapping, optical microscopy and electron microscopy. The as-welded residual stresses exhibit the characteristic M-shaped distribution for hardenable steels, reaching 500–600 MPa in tension in both steels at both thicknesses. However, the modest changes to the chemical composition and the change in plate thickness both significantly influenced microstructures, mechanical properties and residual stress distributions. These sensitivities underline the need for physically faithful models. This extensive characterisation study enables the development and validation of models that predict the development of microstructures and residual stresses in EB welds in low alloy pressure vessel steels.

Data from contour method cut surfaces of low alloy steel plates clad in stainless steel. Two plates, each measuring 300 mm (length) x 200 mm (width) x 20 mm (thickness), were extracted from the outer cylindrical structure of a nuclear steam generator. The material was forged 18MND5 (French designation equivalent to A 508 Gr.3 Cl. 1). Stainless steel beads were then deposited, by submerged arc strip cladding, on the plates. One plate was clad in a single layer of AISI 309L, the second one was clad with a double layer, 309L followed by 308L. The datasets are in the form of lists of x, y, z coordinates, with one point per line, whitespace delimited in millimetres. Each cut has four files associated to it, two for each cut surface. For each surface, there is an outline file identifying the cut surface periphery and a points file containing the points lying on the surface. Two .mat files have also been uploaded, with the results from the analyses on the single and double layer clad plates. These measurements are part of a broader experimental investigation to better understand the role of residual stresses in underclad cracking. The contour method was used to characterise residual stresses in conjunction with neutron diffraction measurements. The details of the experimental procedure and other information will be found in the paper “Internal stresses in a clad pressure vessel steel during post-weld heat treatment and their relevance to underclad cracking" Cattivelli et al., soon to be published.

Ferritic steels experience solid-state phase transformation (SSPT), which causes volumetric changes due to differences in the atomic packing density of different phases in the steel. The importance of the prior austenite grain size (PAGS) as an input physical variable is assessed, for adequately modelling the anisothermal SSPT during welding of ferritic steels. The knowledge of the PAGS value pre-requires a thorough microstructural study of each particular weld, information that might be difficult to acquire. A relationship between hardness, PAGS and phase fractions is proposed to be used to feed in weld models. The case of a single-pass, autogenous, reduced-pressure electron beam weld is used for this study. The adequacy of the finite-element weld model in predicting the micro-constituents, the hardness and the residual stress is demonstrated via comparing the predicted results of the thermo-metallurgical and stress analyses with the set of corresponding experimental data. This work aims at providing a better understanding of the impact of PAGS on transformation kinetics and best practice guidelines for modelling, using an extensively validated electron beam weld model as baseline.

A thermal-metallurgical-mechanical model was developed to study the effects of dilution in each weld pass for multipass gas tungsten arc and submerged arc welding in low alloy steel (i.e. SA508) plates. Hardness distributions and residual stresses were measured on the transverse sections perpendicular to the welding direction of the manufactured weldments. The predicted hardness and residual stresses were compared with the measurement data and shown to be reasonably accurate. The results showed that dilution can significantly affect both the hardness and the residual stress field in the weld metal. It was found that, for the base and filler materials used, increased dilution led to greater weld-metal hardness and reduced the magnitude of tensile stress or promoted compressive stress in the as-deposited and reheated weld metals. This mechanical behaviour is associated with the tendency for diluted weld metal to experience delayed austenite decomposition, owing to the high hard-enability of SA508 steel relative to the filler materials used. Although dilution is irrelevant for the hardness of the base material and its transformation products adjacent to the weld metal, it affected the full-field residual stresses via the equilibrium interaction between the stresses in the base and weld metals.

In this study we aim to determine how the choice of welding process might impact on the through-life performance of critical nuclear components such as the reactor pressure vessel, steam generators and pressuriser in a pressurised water reactor. Attention is devoted to technologies that are currently employed in the fabrication of such components, i.e. narrow-gap variants of gas-tungsten arc welding (GTAW) and submerged arc welding (SAW), as well as a technology that might be applied in the future (electron beam welding). The residual stresses that are introduced by welding operations will have an influence on the integrity of critical components over a design lifetime that exceeds 60 years. With a view to making an assessment based on residual stress as pertinent as possible, weld test pieces were manufactured with each process at a thickness that is representative for such components, i.e. 130 mm. Residual stress measurements were made in the as-welded state using both incremental deep hole drilling and the contour method, and after post-weld heat treatment using deep hole drilling. Part 1 of this study documents weld manufacture, while Part 2 presents the residual stress measurements and discusses their significance.

The accurate prediction of transient temperature fields, induced in alloy systems during advanced manufacturing processes, is critical. These fields influence the magnitude and distribution of residual stresses, the evolution of material microstructures, and material properties such as fracture toughness. Such predictions in the vicinity of a concentrated heat source require precise knowledge of the associated heat flux as a function of position and time. If the applied thermal load is time-dependent this can have a significant effect on the resulting temperature fields and microstructures. In this work a novel three-dimensional heat source is proposed to represent the friction stir welding process along with the semi-analytical solution for the temperature field. The volumetric heat source model has a nontrivial spatial distribution constructed from physical arguments and may account for complex mass transfer, and the associated thermal effects, without explicitly solving the flow equations. A method for incorporating a time-dependent heating scenario into analytical solutions generated by this heat source is also presented. Predicted temperatures are compared with those measured experimentally for two cases reported in the literature and good agreement is observed. Example solutions for various time-dependent heat inputs are also presented.

A three-pass groove weld made by gas-tungsten arc welding in a 20-mm thick SA508 steel plate is modeled to predict the thermal, metallurgical, and mechanical behavior. The dilution for each pass is estimated as the proportion of base material in the weld metal, based on the predicted cross-sectional areas for the fusion zone (FZ) associated with each individual pass. The temperature predictions are consistent with the thermocouple measurement data and cross-weld macrographs. The predicted microstructures are qualitatively compared with the observed microstructures in cross-weld optical micrographs. The measured hardness is then used to quantitatively validate the predictions for postweld microconstituents (e.g., the ferrite, bainite, and martensite fractions), based on a hardness-microstructure correlation. The predicted residual stresses are compared with those measured by neutron diffraction. The results show that the dilution significantly affects the metallurgical and mechanical properties of weld metal (either as-deposited or reheated), and its consideration notably improves the predictions for microstructure and residual stress in the multipass steel weldment. Furthermore, the increase in dilution promotes the formation of martensite, which enhances the hardness, and leads to lower tensile stresses (or higher compressive stresses) in the weld metal. Such behavior arises due to the higher hardenability of the base material employed in this study, coupled with delayed austenite decomposition on cooling.

Grain boundary migration in the presence of concentrated sources of heat is a complex process that has a considerable impact on resultant material properties. The large thermal gradients generated during welding cause grain boundaries to migrate in order to minimise the total free energy of the system. It is important to consider both the thermal gradient driving force, as well as the local curvature driving force of the grain boundaries which both play a significant role in the evolution of the micro-structure in the weld region. In this work a multi-phase field model is used to predict the grain boundary evolution in a Ti6Al4V substrate subjected to a heat source representative of the electron beam (EB) welding process. While numerical simulations incorporating the mass transfer and complex flow dynamics associated with high energy density welding processes are favourable in that they consider the physical processes occurring in the weld explicitly, they are also extremely computationally expensive. As such, the thermal field, on which the phase field model is dependent, is computed using a semi-analytical solution technique. In this approach the complicated flow dynamics of the EB process are represented as a four-quadrant volumetric heat source, the recently published DEC heat source which has shown to be a good thermal representation of EB processes. Using a Green’s function approach, the time and position dependent thermal field is obtained for this DEC heat source in motion is found, free from numerical errors. Predicted grain size distributions are presented for various energy inputs and conclusions drawn based on the applied driving forces, captured in the phase field model

Vacuum laser welding was employed to manufacture 80 mm thick welds in SA508 Grade 3 steel in two weld passes, using a 16 kW laser, while travelling at 150 mm/min. The motivation was to explore the potential for the application of the process to the joining of large, safety-critical nuclear components, such as the steam generators or the pressuriser in a pressurised water reactor (PWR). The advantages of vacuum laser welding are first reviewed, and compared to those of electron beam welding, in terms of the process physics. Preliminary development work is then summarised, together with an evaluation of weld quality, mechanical properties and residual stresses. Vacuum laser welding warrants further development, as it offers significant promise for future nuclear build programmes.

Weld-induced in-process tempering of martensite/bainite was studied through characterisation and modelling. Three-pass gas tungsten arc (GTA) and submerged arc (SA) welds were produced in grooved plates made from a low-alloy ferritic (SA508) steel. A thermal-metallurgical-mechanical model was developed to simulate multi-pass welding while accounting for tempering kinetics. Significant tempering of martensite, in the heat affected zone that was produced by the first pass, occurred during the second and third passes, resulting in a coarsened lath structure, increased carbide precipitation and reduced hardness. The tempering effect was more extensive in the SA weldments than in the GTA weldments, since the tempering mainly occurred in a martensite-dominant region without re-austenitisation for the former while in a partially re-austenitised region for the latter. The predictions for tempered microstructures were consistent with microscopic observations, and the predicted micro-hardness agreed well with measurements when tempering was considered in modelling. The peaks in predicted tensile and compressive residual stresses were reduced by considering tempering effects, since the local yield strength reduced as a consequence of the tempering, thereby limiting the stresses that could be sustained.