Sources and References

• Kou, S. Welding Metallurgy, 2nd ed., Wiley-Interscience.
• Lippold, J. C. Welding Metallurgy and Weldability, Wiley.
• Lippold, J. C. and Kotecki, D. J. Welding Metallurgy and Weldability of Stainless Steels, Wiley.
• Lancaster, J. F. Metallurgy of Welding, Abington.
• Messler, R. W. Principles of Welding: Processes, Physics, Chemistry, and Metallurgy, Wiley.
• American Welding Society, Welding Handbook (Vols. 1–5), AWS.
• Grong, O. Metallurgical Modelling of Welding, Institute of Materials.
• Easterling, K. E. Introduction to the Physical Metallurgy of Welding, Butterworth-Heinemann.
• Cranfield University and The Welding Institute (TWI) technical literature on arc welding and weldability.
• Ohio State University Welding Engineering program technical reports.
• MIT OpenCourseWare 3.37 Welding and Joining Processes.

Chapter 1: Framing Welding as a Metallurgical Process

A weld is a localized metallurgical event in which the base metal experiences the most severe thermal and mechanical excursion it will ever see. For a brief moment a small volume is melted and solidified under constraint, while an adjacent volume is reheated through the full sequence of solid-state transformations. Everything that subsequently governs the service performance of the joint — strength, toughness, corrosion resistance, fatigue life, residual stress — is fixed during this excursion. Welding metallurgy is the discipline that connects the heat source and shielding environment to microstructure, and microstructure to properties and defects.

Three spatial regions recur throughout the course. The fusion zone (FZ) is the volume that was fully molten and has resolidified from liquid, typically as a cast structure influenced by alloy dilution between filler and base metal. The heat-affected zone (HAZ) is the solid material that was heated high enough to alter its microstructure without melting, ranging from a narrow grain-coarsened band next to the fusion line to a remote subcritical region. The unaffected base metal is thermally undisturbed. Each region experiences a distinct thermal cycle — peak temperature, time above a critical temperature, cooling rate — and each responds with its own transformation sequence.

The instructor characterization of the thermal field comes from Rosenthal-type moving heat source solutions. For a thick plate with a point source of power \( Q \) moving at velocity \( v \), the quasi-steady temperature rise above ambient \( T_0 \) at a point behind the source is approximated by

\[ T - T_0 = \frac{Q}{2\pi k r}\,\exp\!\left[-\frac{v(r+\xi)}{2\alpha}\right], \]

where \( k \) is thermal conductivity, \( \alpha \) thermal diffusivity, \( r \) the radial distance from the source, and \( \xi \) the coordinate along the direction of travel. For a thin plate the dimensionality drops and \( r \) is replaced by a two-dimensional distance. These expressions give the cooling rate at a given temperature and the time \( \Delta t_{8/5} \) spent between 800 and 500 C, a parameter that correlates strongly with transformation hardness in steels. Heat input per unit length, \( H = \eta V I / v \), where \( \eta \) is the arc efficiency, is the practical control variable.

Chapter 2: The Fusion Zone — Solidification and Segregation

Once the liquid pool is established, solidification begins epitaxially on unmelted grains at the fusion boundary. Grains grow inward along favourable crystallographic directions that are most closely aligned with the direction of maximum heat flow. As the pool travels, the preferred growth direction rotates, producing the characteristic curved columnar grains observed in etched cross-sections. For a given alloy the balance between temperature gradient \( G \) at the solid–liquid interface and growth rate \( R \) determines morphology: high \( G/R \) favours planar or cellular growth, low \( G/R \) favours cellular-dendritic and then fully dendritic structures, and extremely low \( G/R \) with a large supercooling ahead of the front allows equiaxed grains to nucleate in the remaining liquid.

Because solute partitioning occurs at every point along the solidifying front, the last liquid to freeze is enriched in alloying elements and impurities. The resulting microsegregation is usually described by the Scheil equation,

\[ C_s = k C_0 (1 - f_s)^{k-1}, \]

where \( C_s \) is the composition of the solid formed at solid fraction \( f_s \), \( C_0 \) the nominal composition, and \( k \) the equilibrium partition coefficient. Elements with \( k < 1 \) — sulfur, phosphorus, boron, carbon in most ferrous systems — concentrate in the interdendritic liquid and can form low-melting films that are the precursors of solidification cracks. On the scale of the entire weld pool, macrosegregation arises from fluid flow driven by Marangoni (surface-tension gradient) convection, buoyancy, and electromagnetic forces. These flows control pool shape, penetration, and the distribution of inclusions and filler.

Dilution is the fraction of the fusion zone derived from melted base metal, and it ties weld metal composition directly to process parameters. A high dilution joint between dissimilar metals can push the fusion zone into a two-phase or brittle region of the phase diagram; low dilution is often desirable when cladding a corrosion-resistant overlay. Predictions of weld metal constitution in alloys where multiple phases compete, especially austenitic and duplex stainless steels, are made with constitution diagrams such as the Schaeffler, DeLong, and WRC-1992 diagrams, which correlate a chromium equivalent and a nickel equivalent to the expected ferrite content.

Chapter 3: The Heat-Affected Zone in Carbon and Low-Alloy Steels

The HAZ in a plain carbon or low-alloy steel is a natural laboratory for solid-state transformations. Moving from the fusion boundary outward, one traverses a coarse-grained HAZ where peak temperatures well above \( A_{c3} \) caused rapid austenite grain growth; a fine-grained HAZ where austenitization occurred without coarsening; an intercritical HAZ between \( A_{c1} \) and \( A_{c3} \) where only part of the original structure transformed; and a subcritical HAZ where tempering and possibly strain aging occur without austenitization. On cooling, the austenitized regions transform to some combination of ferrite, pearlite, bainite, martensite, and retained austenite, as dictated by cooling rate, prior austenite grain size, and hardenability.

Hardenability in welding is captured by the carbon equivalent. The IIW expression,

\[ \mathrm{CE}_{\mathrm{IIW}} = \mathrm{C} + \frac{\mathrm{Mn}}{6} + \frac{\mathrm{Cr}+\mathrm{Mo}+\mathrm{V}}{5} + \frac{\mathrm{Ni}+\mathrm{Cu}}{15}, \]

is used for conventional structural steels, while Ito and Bessyo’s \( P_{cm} \) expression is preferred for modern low-carbon microalloyed steels. Higher CE shifts the continuous-cooling transformation curves to longer times, allowing martensite formation at the cooling rates typical of arc welding. Coupling a CCT diagram with the \( \Delta t_{8/5} \) from a Rosenthal solution lets one estimate HAZ hardness and, through Yurioka-type empirical relations, peak hardness as a function of composition and cooling rate. Peak HAZ hardness is a standard weldability screen: above about 350 HV the joint is generally considered susceptible to hydrogen cracking unless preheat, interpass control, or postweld treatment is applied.

Microalloyed steels — HSLA grades containing small additions of Nb, Ti, and V — add a further wrinkle. Their base-metal strength comes from fine ferrite grains and carbonitride precipitates, both of which are degraded in the HAZ: grain coarsening next to the fusion line reduces toughness, while dissolution and re-precipitation of carbonitrides alters strength. Controlled heat input is essential to preserve the properties that justified the alloy choice in the first place.

Chapter 4: Hydrogen, Reheat, Lamellar, and Solidification Cracking

The dominant defect in ferritic steel welds is hydrogen-induced (cold) cracking. It requires four conditions simultaneously: a susceptible microstructure (usually untempered martensite), diffusible hydrogen above a threshold concentration, tensile stress from restraint and thermal contraction, and a temperature low enough that hydrogen becomes trapped rather than diffusing out. Hydrogen enters the weld metal from moisture in fluxes, coatings, or the atmosphere, and redistributes into the HAZ as it cools. Mitigation is a matter of removing one leg of the tetrahedron: low-hydrogen consumables and baking, preheat and interpass temperature to extend the time hydrogen has to escape, control of restraint, and post-weld hydrogen bakeout.

Reheat cracking (also called stress-relief cracking) occurs in Cr-Mo-V and similar creep-resistant steels during post-weld heat treatment, when carbide precipitation inside prior austenite grains strengthens the grain interiors faster than the boundaries relax, driving intergranular fracture. Lamellar tearing is a through-thickness failure in rolled plate loaded perpendicular to the rolling direction, caused by elongated sulfide and silicate inclusions acting as decohesion planes under welding strains.

In the fusion zone itself, solidification cracking — hot cracking along interdendritic boundaries — arises when residual liquid films coexist with tensile strain in the mushy zone. Susceptibility is governed by solidification temperature range, the amount and distribution of low-melting liquid, and the strain imposed during the vulnerable temperature window. Impurity control (especially S and P), filler chemistry chosen to promote a small amount of delta ferrite in austenitic stainless welds, and joint design that reduces restraint are the standard countermeasures. Liquation cracking in the partially melted zone of the HAZ has a similar origin but occurs in base material that was locally remelted at grain boundaries.

Chapter 5: Stainless Steels — Austenitic, Ferritic, Martensitic, Duplex

Stainless steels are classified by the room-temperature matrix, and each class presents a distinct weldability problem. Austenitic grades (300 series) solidify most safely when a few percent of primary delta ferrite forms, since ferrite dissolves S and P and breaks up continuous grain boundary liquid films. Filler metals such as 308L and 316L are formulated to land in the ferrite-containing region of the WRC-1992 diagram. Austenitic welds are susceptible to sensitization: at temperatures between roughly 500 and 850 C, chromium carbides precipitate at grain boundaries, depleting the adjacent matrix of chromium and creating sites for intergranular corrosion. Low-carbon grades (304L, 316L) and stabilized grades (321, 347) mitigate sensitization.

Ferritic stainless steels suffer from severe HAZ grain coarsening and loss of toughness, as well as embrittlement from 475 C aging and sigma phase formation on prolonged high-temperature exposure. Martensitic grades harden directly in the HAZ and behave like quenchable steels, requiring preheat and post-weld tempering. Duplex stainless steels rely on a roughly equal austenite–ferrite balance for their combination of strength and corrosion resistance; fast cooling suppresses austenite reformation, leaving a ferrite-rich HAZ that is brittle and vulnerable to pitting, while slow cooling or reheating allows sigma phase, chi phase, and secondary austenite to precipitate. Heat input must be held inside a narrow window, and nitrogen-containing fillers are used to promote austenite reformation.

Chapter 6: Cast Irons

Cast irons combine high carbon content with a microstructure that depends on graphite morphology — flake (grey), nodular (ductile), compacted, or white — and on the matrix. Welding cast iron is challenging because the HAZ is driven through the eutectoid and, locally, above the eutectic, producing massive carbide at rapid cooling rates along with shrinkage stresses that the brittle matrix cannot accommodate. Strategies include extensive preheat to reduce cooling rate and thermal gradients, nickel-based filler metals whose low carbon solubility and soft matrix absorb diffused carbon from the base metal, and cold-welding procedures that use very short beads, peening, and immediate interpass cooling to manage stress. Brazing and braze welding with copper-based fillers are frequently preferred over fusion welding for repair applications.

Chapter 7: Aluminum Alloys

Aluminum’s welding behaviour is dictated by its tenacious oxide layer, high thermal conductivity, and absence of solid-state phase transformations in most series. The oxide melts near 2050 C while the metal melts near 660 C, so cleaning and oxide disruption through cathodic action in AC TIG or through inert-gas MIG are essential. Weldability varies sharply between series. The 1xxx, 3xxx, and 5xxx alloys are solid-solution or work-hardened and are generally weldable, although cold-worked tempers lose strength in the HAZ through annealing. The 2xxx (Al-Cu), 6xxx (Al-Mg-Si), and 7xxx (Al-Zn-Mg) alloys are precipitation-hardened and present two linked problems: solidification cracking in the fusion zone and HAZ softening through dissolution and overaging of strengthening precipitates. Filler alloys are selected to move the weld metal composition out of the crack-sensitive range on curves of solidification crack susceptibility versus solute content. The 6xxx alloys respond to post-weld aging, which can partially recover strength; 7xxx alloys naturally age at room temperature and recover more completely than 6xxx.

Chapter 8: Nickel-, Cobalt-, Copper-, and Titanium-Based Alloys

Nickel alloys are austenitic throughout and span solid-solution strengthened grades such as Inconel 600 and precipitation-strengthened superalloys such as Inconel 718. The solid-solution grades weld well but are susceptible to hot cracking and ductility-dip cracking because their sluggish diffusion maintains a wide mushy zone and pins impurity segregation at grain boundaries. Precipitation-strengthened superalloys are more severe: strain-age cracking occurs when \( \gamma' \) or \( \gamma'' \) reprecipitates during post-weld heat treatment while restrained, causing intergranular failure in the HAZ. Solution-annealed starting condition, fast heating through the aging range, and modified filler compositions are used to manage this.

Cobalt-based hardfacing alloys (Stellite family) are welded as overlays for wear resistance; dilution control keeps the deposit’s carbide content in the target range. Copper and copper alloys have very high thermal conductivity, so preheat and high-power, concentrated heat sources are needed to overcome conductive losses; oxygen-bearing coppers suffer from hydrogen embrittlement when hydrogen reacts with Cu\( _2 \)O at grain boundaries. Titanium alloys are highly reactive above roughly 400 C and pick up oxygen, nitrogen, and hydrogen from the atmosphere to form brittle interstitial solutions; trailing shields, backing gas, and clean surface preparation are mandatory. Alpha-beta alloys such as Ti-6Al-4V form acicular martensitic alpha-prime in the HAZ; post-weld stress relief and, for structural welds, solution treatment and aging restore properties.

Chapter 9: Dissimilar Metal Welds and Clad Joints

Joining two different alloys introduces issues that neither alloy shows individually: differences in melting point and thermal expansion, dilution into a brittle composition, galvanic corrosion across the interface, and long-term carbon migration driven by activity gradients. A classic case is ferritic steel to austenitic stainless steel, where carbon diffuses from the ferritic side across the fusion boundary to form a decarburized band on one side and a hard carburized layer on the other during service at elevated temperature. Nickel-based buttering layers are used to place the dissimilar interface at a location of low stress and to reduce the thermal-expansion mismatch. Filler selection is guided by Schaeffler-like diagrams to avoid martensitic or fully ferritic regions in the diluted fusion zone.

Chapter 10: Brazing and Soldering

Brazing and soldering join metals using a filler that melts below the solidus of the base materials and wets the joint by capillary action. Brazing, by convention, uses fillers with liquidus above 450 C; soldering uses lower-melting fillers. Joint strength comes not from bulk filler properties but from thin-film effects: a small clearance (typically 0.05–0.15 mm for brazing) produces the highest strength because interdendritic liquid films and solidification defects are suppressed, and because constraint from the high-strength base metal raises the apparent strength of the ductile filler. Flux or reducing atmospheres disrupt oxides and control wetting. Metallurgical interactions include base-metal dissolution into the liquid filler, interdiffusion across the interface, and formation of intermetallic phases that can embrittle the joint if their thickness is uncontrolled. Silver-, copper-, nickel-, and aluminum-based brazing fillers each have characteristic service temperatures and base-metal compatibilities.

Chapter 11: Residual Stress, Distortion, and Post-Weld Heat Treatment

A weld contracts as it cools from the liquidus, but contraction is resisted by the surrounding cold material. The result is a self-equilibrated field of tensile residual stress in and near the weld, balanced by compressive stress farther out. Longitudinal residual stress in a butt weld typically reaches the room-temperature yield strength of the weld metal. Residual stresses drive distortion, reduce fatigue life by adding to applied mean stress, promote brittle and environmentally assisted fracture, and interact with buckling. Post-weld heat treatment (PWHT) at a temperature chosen to allow creep relaxation reduces residual stress, tempers martensite, and allows hydrogen to diffuse out. It must be chosen with care: in creep-resistant steels, PWHT can provoke reheat cracking, and in precipitation-hardened alloys it can trigger strain-age cracking or overage the base metal.

Chapter 12: Embrittlement Mechanisms and Property Assessment

A running theme across all alloys is that service performance depends on whether the welding cycle has moved the material into or out of an embrittled state. The mechanisms taught in this course include: hydrogen embrittlement in high-strength steels and in oxygen-containing coppers; temper embrittlement in Cr-Mo and Ni-Cr steels caused by impurity segregation to prior austenite grain boundaries during slow cooling through 350–550 C; strain aging in low-carbon steels, where dislocations introduced by plastic deformation are pinned by interstitial carbon and nitrogen; ductile-brittle transition shifts in ferritic steels, where HAZ grain coarsening elevates the transition temperature and can move the service condition onto the brittle side; 475 C embrittlement and sigma-phase embrittlement in ferritic and duplex stainless steels; and overaging of precipitation-hardened aluminum and nickel alloys. Laboratory examination in this course links each of these to optical and electron microscopy of etched cross-sections, microhardness traverses, and controlled tensile and impact testing.

Taken together, the chapters above give a framework for reasoning about any welded joint. Begin from the base alloy and its prior condition, identify the phase transformations accessible within the thermal cycle, estimate cooling rates from heat input and geometry, predict the fusion-zone and HAZ microstructures, and then ask which defect mechanisms those microstructures expose the joint to in the intended service environment. Every subsequent design choice — process, filler, preheat, interpass, restraint, PWHT — follows from that reasoning chain.