Argon or nitrogen: the gas that almost no one looks at when comparing LPBF machines

Inside an LPBF machine, there's a molten metal bath at over 1,500 degrees Celsius moving at meters per second. If oxygen is present, this bath oxidizes before solidifying, and the resulting part has oxides between the weld beads, is less dense, and has poorer fatigue resistance.

That's why the chamber is inertized: the air is purged and replaced with a non-reactive gas. It sounds like a minor installation detail, but decide three things in a row: what alloys you can manufacture, what your monthly consumables bill is, and what your facility needs to install the machine.

What exactly does gas do?

Three jobs at once, and it's best to keep them separate because people usually only stick with the first one:

Displace oxygen. The typical goal is to get below 1,000 ppm (0.1%) before starting, and below 100 ppm for sensitive materials. Titanium requires even less.

Sweeping away the fumes. The fusion process generates a plume of vaporized metal and spatter. If this plume remains suspended above the work area, the laser passes through it and loses power unevenly; if the spatter falls onto the powder, the next layer carries defects. The laminar flow over the platform exists to carry all of this to the filter, and its uniformity is one of the parameters that most distinguishes a good machine from a mediocre one. It's a matter of chamber design, not what gas you buy.

Cool and stabilize. The moving gas carries away some of the heat and keeps the chamber at a stable temperature.

Argon

It's the reference gas. Truly inert: it doesn't react with anything manufactured in these machines. It's denser than air, which helps it stay where it needs to be, and it works for absolutely everything: titanium, aluminium, steels, nickel, cobalt.

Its drawbacks are price and logistics. It clearly costs more than nitrogen, and if you use it a lot, you end up needing an external cryogenic tank with its periodic refills, not bottles.

Nitrogen

Cheaper, and with a major logistical advantage: it can be generated on-site. An air separation generator produces nitrogen from compressed air, and beyond a certain consumption level, the investment pays for itself in just a few years.

The problem is that it's not inert, it's just unreactive, which isn't the same thing. At bath temperature, it forms nitrides with several metals. With titanium, this is completely unacceptable: it embrittles the metal. With stainless steels and nickel, the effect is less pronounced and in many cases acceptable—some even take advantage of it—but it ceases to be a minor detail if you're qualifying against a standard, because you're changing the chemistry of the part compared to the powder you purchased.

The practical rule

Titanium and its alloys: argon, without question.
Aluminium: argon. Nitrogen is more acceptable here than with titanium, but argon is the standard and not worth arguing about given the savings.
Stainless steel and tool steels: nitrogen is usually sufficient, and this is where the real savings are.
Nickel and cobalt-chromium: argon if the part will be qualified; nitrogen only with supporting data.

And one condition that takes precedence over all others: if you are going to qualify the part, the gas is part of the frozen process. Changing it afterward requires redoing the validation, just like changing the orientation or layer thickness. We explain this in detail in how to qualify an LPBF part.

What this costs per month

Consumption has two very distinct parts. The initial inerting is a peak: filling the chamber all at once at the beginning of each build. Maintenance is a continuous trickle throughout the manufacturing process, compensating for leaks and renewing the atmosphere as the filter fills.

This leads to the counterintuitive consequence: many short builds consume more gas than a few long builds, even if they add up to the same number of machine hours. Each start-up replenishes the inerting process. This is yet another reason to fully fill the hopper before launching.

The full breakdown of this bill, along with powder, filters, maintenance and depreciation, is in how much does it cost to maintain an LPBF machine per year.

What's needed on the ship

Gas installation is also a matter of planning. You need space for the tank or cylinders, the line to the machine, and—this is what's most often forgotten—oxygen detection in the room. Argon and nitrogen aren't toxic, but they displace air, and argon is heavier, so it accumulates at the bottom. A leak in a poorly ventilated room is a real risk of asphyxiation, odorless and without warning.

What else needs to be prepared before the machine arrives, in what is needed in your shop floor to install an LPBF machine.

Questions to ask when comparing machines

What gas is each material in the library parameterized with? Not "which one it accepts": what gas are used in the parameters you'll be given?
At what oxygen level does it start, and how long does it take to reach that level?
How much gas does inerting use per cycle, and how much does hourly maintenance consume?
Can you switch between materials, and what is the cost of the transition?
Is the airflow on the platform uniform across the entire surface, or are there areas with poorer airflow? This is often overlooked, yet it's the most noticeable issue in the finished part.

The other seven that also influence a purchase are in how to choose a metal 3D printer.

—

What happens to powder when the atmosphere isn't what it should be, in porosity and density. How reactive powder is handled outside the machine, in GARBI. And the specific atmospheres, power levels, and volumes of our equipment, in ALBA 300 and ALBA 500.