Sustainability and life-cycle assessment
📍 Where we are: Part V · Industrial Reality — Chapter 14 of 19. The last two chapters asked whether AM parts can be trusted and tracked; this one asks whether making them this way is actually better for the planet.
Imagine comparing two ways to get a sculpture: carve it out of a huge marble block, or build it up from clay. Clay wastes less stone, but the clay oven burns far more fuel per kilogram than the chisel does. Which one is "greener" depends on how much stone you throw away, how hungry the oven is, and, if the sculpture is a part of an aeroplane, how much lighter it makes the plane for the next 30 years. Life-cycle assessment (LCA) is the bookkeeping that tries to add all of this up, and the answer changes with where you draw the line around the books.
What this chapter covers
We follow a metal part from ore to scrap: feedstock, powder, printing, post-processing, use and end of life. For each stage we say what has been measured, what has only been modelled, and where published LCAs contradict each other. Evidence is labelled as lab-scale, production or contested. Cost figures are in the next chapter; here the currency is energy and emissions.
Problem
Advocates say AM saves material and enables lighter parts; critics say the machines are energy-hungry. Both are partly right [1], and a headline such as "AM is greener" is not a testable claim until the part, process, grid, volume and system boundary are fixed. The engineering problem is to know per application whether AM lowers impact, and by how much.
The foundational review by Kellens and colleagues found that reported specific energy (energy per kilogram of part) of AM systems is one to two orders of magnitude above conventional processes, mainly because deposition rates are low; they also concluded that AM can still be beneficial for very small batches and where redesign gives large use-phase advantages such as lightweighting [2]. That framing, a high manufacturing burden that must be repaid elsewhere, still organises the field.
State of the art
Printing energy (evidence: production-scale measurements, wide spread)
Reported specific energy for metal AM is one to two orders of magnitude above conventional processes [2], and individual L-PBF studies differ widely because machine, material, how full the build plate is, and what the study counts all change the result. We therefore quote no single kWh/kg value. What the evidence does support: in a life-cycle study of selective laser melting, electricity during printing dominated per-part impact in nearly all scenarios, maximising capacity utilisation cut impacts per part by a factor of 14 to 18, and auxiliary equipment was flagged as a reduction target [3]. Two 2025 studies built on measured primary data from industrial AlSi10Mg production describe the literature as sometimes contradictory because of variability among systems and how efficiently they are used, and report that job scheduling and job design (number and size of parts) measurably change the energy of a job [4] [5]. Evidence level: production-scale measurement; I could not verify the single-value energy figures those papers quote and omit them.
Feedstock: powder is not free (evidence: production data, grid-dependent)
Gas atomization turns melt into powder. A 2025 LCA using the average German electricity mix gave 4.61-16.71 kg CO2-eq per kg for closed-coupled atomization and 5.58-24.81 kg CO2-eq per kg for free-fall atomization; inert-gas demand was a major contributor, and argon carries roughly six times the global-warming potential and cumulative energy demand of nitrogen [6]. Those figures are for German electricity; a cleaner grid would lower them.
Buy-to-fly and material savings (evidence: modelled, alloy-specific)
Buy-to-fly (BTF) is the mass of raw stock bought divided by the mass of finished part flown. When a conventional route throws away most of an expensive billet, AM can save material and the energy embedded in it; when the part is simple or the stock cheap, it cannot. A systematic review of comparative LCAs and cost studies found AM preferable where parts are small and traditional manufacture would waste much material, but not in high-volume production [7], and an LCA of AM against CNC machining found the outcome depended primarily on how fully each machine was utilised [8]. The saving is therefore a property of the pair "alloy and part", not of AM. Evidence level: modelled, case-specific; I give no numeric BTF saving because the sources I could open do not support one.
Lightweighting in use (evidence: modelled, not measured in service)
A fleet model combining five aircraft components with stock and fuel-use data found 35-65% mass reductions and, for rapid adoption by 2050, fleet-wide life-cycle primary-energy savings of about 70-174 million GJ per year (sources quote the upper bound as 173 or 174), with about 95% of the saving from reduced aircraft fuel burn [9]. A 2023 Rochester Institute of Technology LCA commissioned by an AM trade association modelled a turbine bracket: cradle-to-gate GWP100 was 55.83 kg CO2-eq per additive bracket against 49.77 for the machined one, yet a flight model for 24 installed brackets over 125,000 flight hours gave a use-phase saving of 21,565 kg CO2-eq, against the AM total of 1,340. The authors state that they assumed equal performance and did not validate it, and that the local energy mix can override other factors [10]. Note the sponsor: the report describes itself as ISO 14040 compliant with review by a three-expert panel if released publicly, but it is still a model of one part.
Powder reuse (evidence: lab-scale to production, mixed)
Reusing unmelted powder is central to AM's material efficiency. A 2025 study of reused Ti-6Al-4V over four builds without top-up found minor morphology change and higher oxygen content, mechanical properties within acceptable limits, a 33% cost drop after one reuse cycle and lower waste, energy and carbon in its LCA, while noting that aeronautical use still faces certification hurdles [11]. That is four builds; the ceiling on reuse is set by oxygen and nitrogen pickup, and different studies disagree on how fast it rises.
Where the burden and the savings sit: powder and printing carry the manufacturing impact; buy-to-fly and use-phase lightweighting are where it can be repaid.
Original diagram by the authors, created with AI assistance.
Why it is hard
- System boundaries. Cradle-to-gate, gate-to-gate and cradle-to-grave studies answer different questions. A study that stops at the factory gate cannot see lightweighting; one that includes it can be dominated by an assumed service life and fuel-burn coefficient.
- Utilisation. Specific energy depends strongly on how full the build plate is, so the same machine yields very different per-kilogram figures [3].
- Grid and location. The RIT study and the atomization LCA both show that electricity mix can override process choice [10] [6].
- Volume. A broad review concluded AM tends to have lower environmental impact only at very low volumes (about 1,000 parts per year or less) and for low solid-to-envelope-volume ratios, unless use-phase gains offset production impact [7]. That is a review conclusion for the cases it examined, not a universal threshold.
- Data quality. Machine-level energy data are scarce and vendor-specific; even recent authors describe literature results as sometimes contradictory because of high variability among systems and efficiency of use [4].
- Assumed equivalence. Use-phase credits assume the AM part is qualified and performs equally, which the qualification chapter shows is itself unresolved.
Open questions
- How should LCAs treat powder reuse and recycling: closed-loop revert versus open-loop downcycling? The reuse evidence cited here covers only a few cycles [11], so credits for long-term recycling remain unsupported.
- Can LCA uncertainty be reported by default? Many studies give single values for processes whose measured energy varies several-fold (contested).
- Does the use-phase saving survive qualification, fleet reality and rebound effects? No in-service, fleet-scale measurement of AM lightweighting savings was found for this chapter; the figures above are models.
- How much of the burden is fixable? The Jung review argues that cheaper and less energy-intensive feedstock, support-free builds, faster production and lower machine cost need fundamental advances rather than scale alone [7].
- Is a standard method needed? A systematic review of 77 LCA studies of AM found that some AM technologies are under-studied, part quality and mechanical performance are often left out, and life-cycle stages after manufacture get little attention [12]; a 2025 cost-and-sustainability review stresses running LCAs with actual yields and the local energy mix [13], which implies current practice often does not.
Industry impact
For buyers, the practical rule from the evidence is conditional: AM is most likely to be greener for expensive alloys where the conventional route wastes much material, for low volumes [7], for geometries where redesign removes substantial mass in a use phase that burns fuel, and on lower-carbon electricity. It is least likely to help for high-volume, simple, solid parts. Suppliers increasingly publish environmental claims; the sponsor, boundary and assumptions determine whether they are comparable. Powder producers face pressure on argon use and atomization energy, and qualification rules (not just chemistry) limit how much reuse savings can be banked in regulated sectors [11].
Why it matters
Environmental claims are becoming procurement criteria. An engineer who cannot say which boundary a number came from will either over-promise or reject a worthwhile application.
In the real world
- Aerospace lightweight brackets are the best-documented case for use-phase savings, yet the quantified results are models [9] [10].
- Auxiliary equipment is part of the L-PBF energy burden [3], and primary data from an industrial cell show that job scheduling and design change the energy per job [5].
- Powder producers' choice of argon versus nitrogen and the electricity source change the feedstock footprint measurably [6].
Key terms
- Life-cycle assessment (LCA) — a standardized accounting of environmental impacts across a product's life, from raw materials to disposal.
- System boundary — the line that decides which stages are counted in an LCA.
- Cradle-to-gate — an LCA covering raw material extraction up to the factory exit.
- Specific energy — energy used per kilogram of product, often in kWh/kg.
- Buy-to-fly ratio — mass of raw material purchased divided by mass of the finished part.
- Global-warming potential (GWP100) — greenhouse-gas impact expressed as CO2-equivalent over 100 years.
- Cumulative energy demand (CED) — total primary energy used across the counted life-cycle stages.
- Gas atomization — producing metal powder by breaking a molten stream with a jet of inert gas.
- Powder reuse — feeding unmelted powder from a build back into later builds.
- Use-phase savings — emissions avoided while a part is in service, such as fuel saved by a lighter aircraft part.
- Rebound effect — extra consumption triggered by an efficiency gain that offsets part of the saving.
Where this leads
Whether the environmental case holds often depends on cost and throughput, the subject of Cost, throughput and scaling to production.