A Review on Acicular Ferrite in Fe-C-Mn-Ni Weld Metal

High-strength steel (HSS) weld metals (WMs) with Fe-C-Mn-Ni additions can suppress austenite transformation-start (T _S ) temperature to simultaneously increase both strength and low-temperature Charpy V-notch (CVN) impact toughness of WMs. Specifically, a two-step approach is useful in understanding the metallurgy of high-performance electrodes and WMs. This approach includes calculated transformation-start (T _S ) temperatures such as New A _r3 (ferrite-start), B _S (bainite-start) and M _S (martensite-start) besides Carbon content below 0.10 wt-%, controlled Yurioka’s Carbon Equivalent Number (CEN) and balanced Ti, B, Al, N, O additions that correlate identified WM chemical composition with desired high-performance microstructures to meet or exceed minimum WM tensile and CVN impact toughness property requirements.The first step uses a set of constitutive (statistical/regression) equations to control the amounts of principal alloy elements such as C, Mn, Cr, Ni, Mo, and Cu so the relevant calculated T _S temperatures such as New A _r3 , B _S , or M _S and Yurioka’s CEN stay in a desirable range relative to the base metals being joined. One also needs to ascertain that the common progression of calculated T _S temperatures wherein New A _r3 >B _S >M _S remains valid. The second step requires balanced Ti, B, Al, N, O additions to further lower the actual T _S temperature compared to the calculated T _S temperature. Both a lower T _S temperature and a narrow start-to-finish (T _S –T _F ) temperature range ensure exceptional CVN impact toughness. The balanced Ti, B, Al, N, O content can be ascertained using an Artificial Neural Network (ANN) template offered by the Japan Welding Engineering Society (JWES) at its website, built on Evans’s database. The JWES-ANN template allows one to manipulate 16 elements of WM compositions, each within a specified range and seek a lower predictive temperature (T _28J /°C) below − 80°C for achieving 28 J absorbed energy during CVN impact testing.A recent publication in Welding in the World on GMA welding of 15-mm thick, Q&T, S690QL grade (per EN 10025-6) HSLA structural steel showed that depending on weld cooling rate (CR) and calculated value of T _S temperature, one could further balance the total Ti-B-Al-N-O additions, with a Ti:B ratio at 10:1, in achieving a YS/UTS ratio at about 0.90 to provide a superior combination of high strength, excellent ductility and low-temperature CVN impact (fracture) toughness for use in demand-critical applications.Another latest research publication in Materials employed a hybrid machine learning framework combining ANN with a Genetic Algorithm (GA) to optimize chemical compositions of SMA WMs based on Evans’s database for achieving targeted mechanical properties. The inverse neural network model was enabled via Bayesian optimization but didn’t use austenite-transformation (T _S ) temperature, or New A _r3 equation. Instead, the case settings used very low targeted WM mechanical properties, with YS at 489 MPa, and UTS at 571 MPa but with YS/UTS ratio at 0.86. Application of JWES-ANN tool showed a higher error range for CVN impact toughness at T _28J /°C for weld trial # 10 that contained nearly 670 ppm Ti, 130 ppm B, 156 ppm Al, 250 ppm of N and 250 ppm O, with a total Ti-B-Al-N-O content at about 1451 ppm.