🖨️ Metal Additive Manufacturing & Powder Metallurgy

Additive Manufacturing Enters Metal Engineering

For decades, metal parts were made by removing material (machining), forcing it into shape (forging, casting), or pressing and sintering powder into simple shapes. Metal additive manufacturing (AM) — often called metal 3D printing — adds a fourth path: building a part up layer by layer directly from metal powder or wire, guided by a digital model. It enables geometries impossible with any subtractive or forming process — internal cooling channels, lattice infill, and consolidated assemblies that used to require multiple parts and joints — and it shares deep roots with traditional powder metallurgy, the century-old craft of turning metal powder into finished components. Understanding both the process physics and the powder science behind them is essential for any engineer specifying or qualifying an AM part.

The Two Dominant Metal AM Process Families

Powder Bed Fusion (SLM/DMLS)

In powder bed fusion — commonly called selective laser melting (SLM) or direct metal laser sintering (DMLS) despite that second name, since the powder is actually fully melted rather than sintered — a recoater blade or roller spreads a thin layer of metal powder, typically 20 to 60 microns thick, across the build plate inside an inert-gas (argon or nitrogen) or vacuum chamber. A focused laser (or, in electron beam melting, an electron beam under vacuum) then scans across the layer following the part's cross-section, fully melting the powder along its path and fusing it to the layer beneath. The build plate lowers by one layer thickness, a fresh layer of powder is spread, and the cycle repeats — thousands of times for a tall part. Unfused powder around the part acts as support and is later removed and, within limits, recycled for the next build.

Directed Energy Deposition (DED)

In directed energy deposition, metal powder or wire is fed directly into a melt pool created at a moving nozzle by a laser, electron beam, or electric arc, building up material only along the path the nozzle travels — much like a robotic welding torch that also adds new material where directed. DED deposits material far faster than powder bed fusion and is not confined to an enclosed build chamber, making it well suited to large structural parts, cladding a wear- or corrosion-resistant layer onto a cheaper substrate, and repairing worn or damaged high-value components (turbine blade tips, for example) by rebuilding lost material directly onto the existing part. The trade-off is coarser feature resolution, a rougher as-deposited surface, and less capability for fine internal detail than powder bed fusion.

Powder Characteristics That Matter

The feedstock powder is not an incidental commodity — its physical characteristics directly control build quality and final part properties.

• Particle size distribution (PSD): most AM powders are gas- or plasma-atomized into a narrow size band, commonly 15–45 microns for powder bed fusion and somewhat coarser for DED. A tight, well-controlled PSD lets the recoater spread a thin, uniform layer and lets the laser fully melt every particle; too broad a distribution, with a tail of very fine or very coarse particles, leads to uneven layers, incomplete melting of coarse particles, or excessive fine-particle spatter and oxidation.
• Flowability: powder must flow freely and pack consistently to spread into a smooth, gap-free layer. Flowability is typically quantified with a Hall flow meter (time for a fixed powder mass to flow through a standard funnel) or an angle-of-repose test. Poor flowability — from irregular particle shape, moisture, or fine-particle agglomeration — produces streaks, gaps, and local density variation in the spread layer, which becomes porosity in the finished part.
• Morphology and sphericity: gas-atomized powders are nearly spherical, which packs and flows far better than the irregular, angular particles from cheaper water atomization. Sphericity is a major reason gas-atomized powder dominates powder bed fusion despite its higher cost.
• Moisture and oxygen content: reactive metals like titanium and aluminum alloys are especially sensitive to absorbed moisture and surface oxide, both of which can be trapped in the melt pool as gas porosity or oxide inclusions. Powder is stored, handled, and often re-dried under controlled, low-humidity or inert conditions for this reason.
• Reuse degradation: unfused powder recovered from a build is commonly sieved and reused, but repeated exposure to the build atmosphere and thermal cycling gradually degrades sphericity and raises oxygen content, so producers track powder reuse cycles and blend or retire powder once it falls outside qualified limits.

How AM Microstructure Differs From Wrought and Cast Metal

The physics of AM solidification is unlike either casting or wrought processing, and the resulting microstructure reflects it.

• Rapid, repeated solidification: each melt pool solidifies in milliseconds and is then reheated by the next several passing layers, producing very fine, non-equilibrium microstructures — quite different from the coarser, slowly solidified grains of a casting or the recrystallized, worked grains of wrought material.
• Columnar grains and anisotropy: because each layer solidifies epitaxially onto the one beneath, along the steep thermal gradient pointing back toward the already-solid material, grains tend to grow long and columnar in the build direction rather than equiaxed. The result is anisotropy — mechanical properties, especially ductility, fracture toughness, and fatigue life, can differ significantly depending on whether the load is applied parallel or perpendicular to the build direction, something wrought metal and most castings do not exhibit to the same degree.
• Porosity: AM parts can contain several distinct porosity types — gas porosity from dissolved or entrapped gas in the powder or melt pool, appearing as small, spherical voids; lack-of-fusion porosity, irregular voids where successive layers or scan tracks fail to fully melt and bond together, often the most damaging type because of its sharp, crack-like geometry; and keyhole porosity, formed when excessive laser energy vaporizes metal and creates an unstable, collapsing vapor cavity. Each type points to a different process parameter needing correction — laser power, scan speed, hatch spacing, or layer thickness.
• Residual stress: the intense, localized, repeated heating and cooling of each melt pool leaves substantial residual stress in the as-built part, which can distort thin sections during or after the build and, if left untreated, reduce fatigue performance.

Post-Processing to Reach Comparable Properties

Because of these as-built characteristics, few metal AM parts go straight into service without further processing:

• Stress relief: a moderate-temperature heat treatment, often performed with the part still attached to the build plate, that relaxes residual stress before the part is cut free and distorts.
• Hot isostatic pressing (HIP): the part is heated to roughly 60–80% of its melting temperature while simultaneously subjected to high inert-gas pressure (100 MPa or more) from all directions. This combination collapses internal gas and lack-of-fusion pores and diffusion-bonds them shut, closing internal voids that would otherwise act as fatigue-crack initiation sites. HIP is close to mandatory for AM parts destined for fatigue-critical aerospace or medical applications, since closing porosity is usually the single largest factor separating AM fatigue life from wrought material.
• Heat treatment: a full solution treatment and aging (for aluminum and nickel alloys) or an anneal, normalize, or quench-and-temper cycle (for steels and titanium alloys) breaks down the columnar, non-equilibrium as-built grain structure toward something closer to the equiaxed, more isotropic structure of wrought material, restoring more balanced strength, ductility, and toughness.
• Surface finishing: the as-built surface of both powder bed fusion and DED parts is rough — partially melted powder particles adhere to external surfaces — and that roughness is itself a significant fatigue-crack initiation site. Critical surfaces are machined, media-blasted, or shot-peened (which also imparts beneficial compressive residual stress) before the part is considered finished.

Even after all of this, many qualification programs still report a fatigue-life debit for AM parts relative to wrought equivalents unless every one of these steps is verified, which is why aerospace and medical device qualification of AM parts is unusually process-control-intensive: the powder lot, the exact machine parameters, the build orientation, and every post-process step are locked into a qualified procedure, much like a welding procedure specification.

Powder Metallurgy Beyond AM

Traditional powder metallurgy (PM) predates AM by generations and remains a distinct, high-volume manufacturing route in its own right: metal powder is pressed in a rigid die under high pressure to form a "green" compact close to final shape, then sintered — heated below the melting point so the particles bond by solid-state diffusion at their contact points — to reach final strength. Press-and-sinter PM is prized for near-net-shape production of small, geometrically simple, high-volume parts (gears, bearings, cams) with minimal material waste and tight dimensional control, and it can produce self-lubricating bearings by leaving controlled internal porosity that is impregnated with oil — the opposite goal from AM, where porosity is almost always the enemy.

Where Metal AM Earns Its Keep

Metal AM is not a wholesale replacement for casting, forging, or machining — it is most valuable where its unique capabilities outweigh its higher per-part cost and heavier qualification burden: low-volume or highly customized parts (patient-specific medical implants), geometries impossible any other way (conformal cooling channels in injection-mold tooling, lattice-infilled brackets that cut weight while meeting a stiffness target), and rapid repair or rebuild of high-value components via DED cladding. As process monitoring, powder quality control, and post-processing standards mature, the gap between AM and wrought mechanical properties continues to narrow, but an engineer specifying an AM part today should still treat porosity control, build-direction anisotropy, and a defined post-processing route as first-order design decisions, not afterthoughts.