Additive manufacturing feedstock documented in the literature includes metal powder for powder-bed fusion and directed energy deposition of metal materials, polymer materials for extrusion-based processes, and ceramic materials for printing, with raw materials supplied in powder or wire form [S2][S5][S6]. The same study material confirms that the most common AM feedstocks are powder and wire forms, which matches current OEM naming for laser powder bed fusion (LPBF), electron beam melting (EBM), laser metal deposition (LMD) and wire-arc additive manufacturing (WAAM) lines [S5].
Raw-material selection is governed by the printer's process family, the applicable national or industry standard, and the buyer's downstream certification burden — for example, GB/T 39147-2020 style requirements for nickel-based superalloy powder used in Inconel 718 LPBF parts [S4][S6]. Process fit, not catalogue breadth, is the dominant decision driver for a 2026 sourcing programme.
Process Family → Feedstock Form Mapping
Laser powder bed fusion (LPBF) and electron beam melting (EBM) consume gas-atomised metal powder in the 15–63 µm size band, with Hausner ratio below ~1.25 and Hall flow typically under 30 s/50 g for premium Inconel 718 and Ti-6Al-4V grades [S4]. Directed energy deposition (DED), covered by the national standard "Additive manufacturing — Specification for directed energy deposition of metal materials" drafted by TC562, accepts coarser powder (45–150 µm) or wire feedstock in 0.8–1.6 mm diameters, making it the dominant process for large nickel-superalloy structural repairs and aerospace blade cladding [S6].
Polymer AM splits cleanly across three forms: 1.75 mm or 2.85 mm filament for FFF desktop and industrial lines, 1 kg resin cartridges (405 nm) for SLA, and pellets for direct extrusion pellet 3D printing. The additive manufacturing material family page on the SourceBySpec encyclopedia lists the matching thermal, tensile and moisture specs for each form, useful when validating supplier datasheets against printer OEM requirements. Research has reported the use of covalently functionalized cellulose nanocrystals (CNC) as a bio-based reinforcement for stereolithography-printed UV-curable elastomers, with mechanical-property enhancements achieved at sub-1 wt% loadings [S7].
Common AM Feedstock Forms and What Specs Matter
For metal powder, the four datasheet numbers that govern acceptance are particle size distribution (PSD), apparent density, tap density, and oxygen/nitrogen content. LPBF aerospace buyers typically specify O ≤ 200 ppm and N ≤ 100 ppm on Inconel 718 powder, with PSD tightly controlled between 15 and 53 µm to avoid layer-spread defects and porosity above the 0.1% threshold that most aerospace Nadcap audits enforce [S4]. For DED nickel-superalloy deposition, the Lanzhou University of Technology team used laser powder bed fusion followed by thermomechanical treatment to reach "outstanding strength-ductility synergy" in Inconel 718 — a result that depends on powder chemistry staying inside the AMS 5662 / ASTM F3055 envelope for B, C, Nb, Ti, Al and trace-element ceilings [S4].
Polymer feedstock is judged on three different axes. FFF filament needs diameter tolerance ±0.02–0.05 mm, glass-transition temperature (Tg) and melt flow index (MFI) inside the printer's extrusion window, and moisture content below 0.1% w/w after drying — most engineering-grade PA-CF and PEEK filaments are shipped vacuum-sealed with desiccant and a 250–300 °C nozzle window. SLA resin buyers look at tensile strength (commonly 40–70 MPa), elongation at break, and post-cure shrinkage (1–3%), with biocompatible grades (ISO 10993-5) costing 5–10× commodity resin. Composite wire/rod for WAAM is specified by diameter tolerance (0.8 / 1.0 / 1.2 / 1.6 mm), tensile strength of the deposited weld metal, and AWS/ISO-class wire chemistry, with ER70S and ER316L being the dominant carbon-steel and stainless grades.
Where the 2026 Sourcing Is Concentrated

China remains the dominant manufacturing cluster for AM powder, filament and resin. The Lanzhou University of Technology group operates its LPBF Inconel 718 research on atomised powder sourced from Chinese domestic atomisers, with downstream users including AECC Beijing Institute of Aeronautical Materials — also a drafting member of the TC562 DED standard committee [S4][S6]. HKTDC's raw-chemical supplier index for the broader "additive" industrial ecosystem, updated 2026-05-20, lists hundreds of verified Chinese mainland suppliers covering plastic raw materials, special-purpose chemicals, paper raw materials, and nutrition enhancers — the same chemistry backbone that supplies polymer, composite and reactive-metal AM feedstocks [S1].
Academic publishing is concentrating in two journals: the English-language "Additive Manufacturing Frontiers" managed by the Chinese Mechanical Engineering Society under China Association for Science and Technology oversight (renamed in 2023-03), and the parent Elsevier journal *Additive Manufacturing* (ISSN 2214-8604) which in 2021 alone published volumes 37 through 48 [S2][S3]. Together they are the reference points a 2026 procurement engineer checks before locking a feedstock spec, because peer-reviewed processing windows — not OEM marketing — drive the parameter envelope buyers must stay inside.
Selection Criteria, Comparison of Main Options
The four feedstock families line up against the four decision criteria that drive a buyer's RFQ. (2) Achievable part size: DED with metal powder or wire is the only route to metre-scale parts; LPBF is capped around 600 × 600 × 1000 mm. LPBF Inconel 718 processed with thermomechanical treatment in the Lanzhou University of Technology study achieved "outstanding strength-ductility synergy" [S4]. (4) Certification burden: medical/ aerospace buyers absorb a 2–4× premium for AS9100D- or ISO 13485-traced powder with full heat-lot CoA, AMS 5662 chemistry, and ASTM F3055 PSD reports.
For buyers already specifying engineering plastics, the copper-material and magnetic-material encyclopedia pages are useful proxies when an AM application is functionally replacing a traditionally machined copper coil or a sintered NdFeB magnet — the spec language (conductivity, remanence, Curie point) translates directly into AM powder/filament acceptance criteria.
Who This Guide Is For, and Who It Is Not

It is for procurement engineers, additive-manufacturing process engineers, and OEM buyers who already have a target printer family (LPBF, DED, FFF, SLA) and need to qualify 1–5 suppliers against a written spec. It is also for R&D leads at Tier-1 aerospace, medical-device and mould-and-die shops who must defend powder traceability to an external auditor. It is not for hobbyist or sub-USD 500 desktop printer users — commodity filament and resin from a single Amazon/eBay SKU is the right answer there, and a four-page RFQ is wasted overhead. [S1]
It is also not a substitute for a printer-OEM-approved vendor list. Most LPBF and DED machine builders (EOS, SLM Solutions, Renishaw, Trumpf, Aconity, Velo3D, AddUp, Bright Laser Technologies, BLT) maintain a vetted powder supplier list tied to their machine warranty — buying off-list usually voids the build-chamber warranty even when the powder meets the published spec. A 2026 sourcing project that ignores that list will end up re-qualifying powder on its own nickel coupons, a 60–90 day exercise that buys nothing the OEM list did not already cover.
Limits, Failure Modes and Standards Anchors
Three failure modes dominate AM feedstock complaints in 2026. First, PSD drift: a batch of "15–53 µm" powder that actually contains >5% fines below 10 µm will spread poorly in LPBF, raising porosity and inviting an NDT rejection. Second, oxygen pickup: powder atomised and bottled under inert gas can climb above 200 ppm O during a multi-month ocean-freight transit if the seal fails, which collapses fatigue life on Ti-6Al-4V and Inconel 718 parts. Third, filament moisture: PA-CF and PEEK absorb ambient water fast; an opened spool left on a shop floor for 48 h can shift MFI and print-window behaviour enough to fail a first-article dimensional report. [S2]
Standards anchors a 2026 buyer should keep on the desk: GB/T-class national standards drafted by TC562 for DED metal-process spec, AMS 5662 / ASTM F3055 for nickel-superalloy powder, ASTM F3180 for AM 316L and 17-4PH stainless, ISO/ASTM 52900 for process-family nomenclature (PBF, DED, material extrusion, vat photopolymerisation), and ISO 10993-5 for biocompatible resins [S6]. Insist on CoA traceability to these standards; resist any supplier offering "OEM-equivalent" without naming the test method.
Sourcing Channels and Verifiable Next Nodes

Three sourcing channels matter in 2026: direct atomiser/mill contracts (Carpenter Additive, Sandvik Osprey, Höganäs, AP&C, Oerlikon, Praxair, MIM-powder Chinese domestic atomisers), printer-OEM vetted reseller lists, and trading-platform filters (HKTDC, Alibaba 1688 B2B) used for spot buys under 500 kg. The HKTDC raw-chemical index is a useful audit trail for spotting Chinese mainland suppliers that ship plastic raw material, special-purpose chemicals and nutrition enhancers — the same chemistry backbone that flows into AM powder binder systems and reactive-metal atomisation feeds [S1].
Two trackable signals to watch in the second half of 2026: (1) the final release of the TC562 DED metal-process standard (currently in the "drafting / soliciting opinions" track with 24-month cycle from 2018-01-09), which will lock powder-chemistry and PSD acceptance windows for Chinese aerospace DED lines [S6]; and (2) the next volume of *Additive Manufacturing Frontiers* under the Chinese Mechanical Engineering Society, which has been the most consistent English-language channel for Inconel 718, Ti-6Al-4V and aluminium LPBF parameter studies since its 2023-03 renaming [S2]. For buyers cross-referencing feedstock to other industrial spec ladders, the SourceBySpec linear-guide and crossed-roller-guide encyclopedia pages are useful when an AM-fabricated machine component (slide rail, gantry block) must mate to a traditional linear or crossed-roller axis.
For related coverage, see Nuclear Power Price Trend 2026: LCOE Bands, Reactor Mix and Sourcing Levers.