Microstructure and Mechanical Properties of SUS316L-15CrMo Dissimilar Steel Joints Welded with Different Filler Wires

Literature Overview

This paper, published in Materials in Mechanical Engineering (Volume 47, Issue 10, 2023), investigates the microstructural evolution and mechanical performance of dissimilar steel weldments joining SUS316L austenitic stainless steel to 15CrMo low-alloy steel using TIG welding. Two filler wires were compared: ERNiCrMo-3 (a nickel-based alloy) and ER309 (austenitic stainless steel). The study was supported by the National Natural Science Foundation of China (Grant 51601078) and the Jiangsu Provincial Government Overseas Scholarship Program (JS-2019-322). The research team from Jiangsu University of Science and Technology, Jiangsu Better Pipe Fittings Co., Ltd., and the Taizhou Branch of Jiangsu Special Equipment Safety Supervision and Inspection Research Institute conducted systematic metallographic and mechanical evaluations of both weld configurations.

Core Technical Findings

The fundamental challenge in dissimilar steel welding lies in the metallurgical incompatibility between the two parent materials. SUS316L is an austenitic stainless steel with high corrosion resistance and non-magnetic properties, while 15CrMo is a creep-resistant low-alloy steel containing approximately 1.25% Cr and 0.30% Mo, typically used in high-temperature service conditions. When these materials are joined, the welding process introduces a complex gradient of chemical composition, microstructure, and mechanical properties across the joint.

The study reveals that the two filler wires produce distinctly different solidification modes in the weld metal. ER309 produces an austenite-ferrite mixed solidification mode, resulting in a dual-phase microstructure of austenite and delta ferrite. This is consistent with the Schaeffler diagram predictions for ER309 composition, where the iron dilution from the 15CrMo side shifts the weld composition toward a region that favors delta ferrite formation. In contrast, ERNiCrMo-3 produces a completely austenitic solidification mode, yielding a fully austenitic weld microstructure. This difference is significant because the nickel-based filler wire introduces sufficient nickel content to maintain the austenite stability even with iron dilution from the 15CrMo side.

Weld Microstructure Analysis

The microstructural observations provide critical insights into the metallurgical behavior of each joint. The ER309 weld shows the expected austenite-ferrite mixed structure, where delta ferrite forms as the primary phase during solidification and partially transforms to austenite during cooling. The ferrite content in ER309 welds typically ranges from 5% to 15%, depending on the dilution ratio and welding parameters. This delta ferrite serves a dual role: it mitigates hot cracking susceptibility by absorbing sulfur and phosphor impurities into the ferrite phase, but excessive ferrite can reduce corrosion resistance and increase magnetic permeability.

The ERNiCrMo-3 weld presents a fully austenitic microstructure without delta ferrite. However, the study identifies the formation of Laves phase in this weld metal. The Laves phase is a hard, brittle intermetallic compound with a topological structure of M2C (where M represents Fe, Cr, Ni, or Mo). Its formation is attributed to the interaction between the high nickel content of the filler wire and the carbon and chromium from the 15CrMo base metal during the solidification and cooling process. The presence of Laves phase is a critical concern for weld ductility and fracture toughness.

Mechanical Property Comparison

The most striking finding is that ERNiCrMo-3 produces weldments with higher tensile strength and greater elongation than ER309, despite the presence of Laves phase. This counterintuitive result can be explained by the suppression of carbon diffusion into the 15CrMo heat-affected zone. During welding, carbon atoms from the 15CrMo side tend to diffuse toward the weld pool, forming brittle carbides in the HAZ that reduce ductility. The nickel-based filler wire creates a diffusion barrier effect, limiting carbon migration and preserving the ductility of the 15CrMo HAZ. This results in improved overall joint ductility.

The peak hardness for both joints occurs in the coarse-grained heat-affected zone (CGHAZ) on the 15CrMo side, which is the expected location of maximum hardness due to the formation of hard martensitic or bainitic microstructures in the high-temperature HAZ of 15CrMo steel. The ERNiCrMo-3 weld metal exhibits approximately 30 HV higher hardness than the ER309 weld metal, directly attributable to the Laves phase precipitation.

Engineering Practice Implications

In industrial applications involving dissimilar joints between austenitic stainless steel and low-alloy creep-resistant steels, the selection of filler metal must balance competing requirements. ER309 is the conventional choice for such joints due to its established track record, adequate ductility, and resistance to hot cracking. However, for applications where carbon diffusion into the low-alloy steel HAZ is a critical concern—such as high-temperature service conditions where carbon migration can embrittle the 15CrMo side—the ERNiCrMo-3 option warrants serious consideration.

The Laves phase issue in ERNiCrMo-3 welds represents a potential weakness for applications requiring high fracture toughness. The Laves phase is inherently brittle and can act as crack initiation sites under cyclic or impact loading. For pressure vessel or piping applications governed by ASME Section VIII or Section IX, the presence of Laves phase may require additional qualification testing, including fracture toughness evaluation and fatigue testing.

Recommended Process Controls

For ERNiCrMo-3 weldments, the following process controls are recommended to mitigate Laves phase formation:

1. Minimize heat input to reduce the time spent in the temperature range where Laves phase precipitation is thermodynamically favorable (approximately 800–1000°C).
2. Apply post-weld heat treatment (PWHT) at 1050–1100°C followed by rapid cooling to dissolve and re-distribute the Laves phase.
3. Consider multi-pass welding with interpass temperature control to prevent excessive carbon diffusion between passes.
4. Perform post-weld metallographic examination specifically targeting Laves phase identification using optical microscopy with appropriate etchants.

The study also highlights the importance of considering the entire joint cross-section rather than focusing solely on the weld metal. The 15CrMo CGHAZ remains the weakest link in both configurations, and any filler wire selection must be evaluated in the context of the complete joint performance.

Key Reflections

This research contributes valuable data to the engineering community's understanding of dissimilar steel welding with nickel-based fillers. The finding that ERNiCrMo-3 can suppress carbon diffusion while maintaining adequate mechanical properties opens new possibilities for optimizing dissimilar joints in high-temperature service. However, the Laves phase issue must not be overlooked. Engineers should carefully evaluate the service conditions—particularly temperature cycling, stress levels, and corrosion exposure—before selecting a nickel-based filler for such applications.

The study's methodology, which compares two fundamentally different filler wire systems under identical welding conditions, provides a clean experimental basis for decision-making. Future work should extend this comparison to include other nickel-based alloys such as ERNiCr-3 or ERNiClad-Mo-1, as well as evaluate the long-term creep and stress-rupture behavior of these joints. For engineers working in power generation, petrochemical, or nuclear industries where SUS316L-to-low-alloy-steel transitions are common, this research offers practical guidance on filler selection that goes beyond the conventional ER309 approach.