Excess powder: how much is reused, when it should be removed, and what is done with it

In a laser powder bed fusion build, the laser melts only a small fraction of all the powder inside the chamber. The rest—the majority—has never melted: it has been surrounding the workpiece, serving as a bed, and is still there when the process is complete.

That powder isn't waste. It's the same material you bought, and what's done with it largely determines the economics and environmental impact of the process. A shop floor that reuses materials effectively and one that doesn't can have material costs that differ significantly.

What can be recovered and what cannot

When the chamber is emptied, the powder separates into three different destinations, and it is best not to mix them:

Where it comes from	Condition	What happens to it
From the bed, far from the part	Virtually intact	It is screened and returned to the cycle
From around the part and supports	It has undergone thermal cycles and some splashing	It is screened and returned, with increased monitoring
From the filter and the gas circuit	Splashing, condensate, fine particles	It does not return. It is hazardous waste and is managed as such

The rule that saves trouble: what comes out of the filter never goes back into the bed. It's the fastest way to ruin an entire batch and, moreover, a safety risk — that's what the ATEX article is about.

What happens to powder every time it turns around

Four things, and all four are measured.

It gets dirty from splashing. The molten bath ejects droplets that solidify in the air and fall onto the bed. These are larger particles than the original powder, sometimes fused together, and they spoil the coating. Screening exists precisely for this reason: a mesh that allows the appropriate size to pass through while retaining agglomerates and splashing.

The particle size distribution changes. Round after round, the screening removes the coarse particles, and handling loses some of the finest ones. The size distribution narrows and shifts. This changes how the powder flows and how the layer compacts, and therefore changes the density of the part even if you haven't touched a single parameter.

It absorbs oxygen. This is the mechanism that most limits the powder's lifespan, and it depends heavily on the alloy. In titanium, it's the critical factor: titanium readily absorbs oxygen, and oxygen embrittles it, so the residual oxygen content isn't just a data sheet specification; it's an acceptance criterion. In stainless steels, the margin is much wider.

Loss of sphericity. Ideal particles are smooth spheres because they flow and pack well. Thermal cycles and handling deform them and add satellites. Less sphericity means poorer flowability and less uniformly distributed layers.

What needs to be measured to decide if it's still worth it

A university laboratory isn't necessary. A well-informed decision can be made with four measurements:

Particle size distribution. Compare it to that of the virgin powder and see how much it has shifted.
Oxygen and nitrogen content. Essential for titanium and reactive alloys; recommended for all.
Flow and apparent density. This is the inexpensive measure that best anticipates coating problems.
Chemical composition. To detect cross-contamination, which is the most costly failure.

And above all, a management practice worth more than any test: batch traceability. Knowing which batch the powder for each build came from, how many cycles it has undergone, how much virgin powder has been added, and which parts were produced from it. Without this, the day a defective part appears, there's no way to know if the problem was the powder, and all parts have to be suspected.

The two strategies that are used

Continuous replenishment. After each build, all the recovered powder is sieved and a proportion of virgin powder is added to compensate for what was consumed. The batch is maintained indefinitely within certain limits and is monitored by periodic measurements. This is standard practice in industrial production and for tolerant alloys.

Closed batches with a limited number of cycles. A batch is used a fixed number of times and then discarded, without being mixed with virgin material. This is standard practice in regulated sectors, where the goal is not to extract the powder but to be able to demonstrate exactly what was inside each part.

Which one to choose isn't determined by the technique: it's determined by what you have to be able to prove. If no one is going to audit you, choose the first one. If your part is going on an airplane or a human body, choose the second one, no question about it.

When does he retire

There's no universal number of cycles, and be wary of anyone who gives you one without asking about the alloy. The criterion is always the same: it's removed when any of the measurements fall outside the window you've set, not when a specific number is reached.

Defining that window is a qualification task, and it's linked to the part, not the powder: the same powder might be perfectly acceptable for a tool holder but unacceptable for a fatigue part. This is the same reasoning explained in how to qualify an LPBF part.

What is done with the removed powder

Just because it's not suitable for your part doesn't mean it's worthless. In order of usefulness:

Other less demanding parts within the same company: tooling, prototypes, test parts. This is the most efficient and least common method.
Other additive processes, such as targeted addition, which work with much wider size distributions and tolerate powder that LPBF already rejects.
Reatomization: converting it back into spherical powder with the desired distribution. This has an energy cost, but it conserves the metal.
Conventional metallurgical recycling: returning it to the foundry as high-purity scrap. This is the final step, and it is still infinitely better than a landfill.

What must never happen is that it ends up in a general waste container. Many metal powders are combustible and some are reactive: the removed powder is stored, identified, and delivered to an authorized waste management company.

How we approach it

The GARBI sieve is specifically designed for this task: recovering powder from each build site in a condition suitable for reuse in the machine, in a closed circuit, with powder management planned from the outset, not added later. It's not an accessory: in practice, it's what determines whether the material cost of your part is what you expected or three times higher.

At SamyPowder we explain what we control in the powder we supply and why that is the information that should be demanded from any supplier.

And the next step is the powder line from waste: if the powder can come from material that is now waste, the cycle stops starting in a mine.

One last thing, and it's one of the biggest money-savers: the parameter settings for our machines are open. Qualifying a reused powder, or one from another supplier, or one derived from waste, is possible because you can write down and save its parameter set. In a machine limited to a powder certified by the manufacturer, the discussion about reuse doesn't even begin.

—

The complete environmental balance, including the benefits and drawbacks of additive manufacturing, can be found in Is 3D metal printing sustainable?. How the powder is manufactured and the different types available can be found in How is metal powder manufactured?. And handling safety can be found in What does ATEX mean?.