Using 3D Printing Technology to Design Carbon Dioxide Capture Filters
A direct look at how additive builds change carbon dioxide capture filters: channel geometry, wall thickness, material choice, and where the approach stops paying off. Written for engineers who must decide between printing a filter, machining it, or doing both.
Why additive building fits carbon dioxide capture filters
A carbon dioxide capture filter is a surface-area problem. Gas must touch amine, zeolite or a metal-organic framework long enough to transfer. Every extra square meter of wetted surface raises capture rate per unit volume.
Conventional manufacturing reaches that surface with stacked corrugated sheets, rolled mesh or extruded honeycomb. Those routes fix geometry first: constant channel pitch, constant wall thickness, few options for mixing features inside the duct.
3D printing technology removes that constraint. The part is built layer by layer, so channel width, wall thickness and internal ribs can all change along the gas path.
That freedom is what makes additive attractive for sorbents. A printed monolith can widen channels at the inlet, where dust loads are highest, then narrow them downstream where contact time matters more. Extrusion cannot do that in one piece.
The catch is that printing also changes the surface. Layer lines add roughness, which helps wetting but traps fines. We treat roughness as a design variable, not a defect.
Channel and lattice geometry that actually works
Start with the smallest channel the process can clear. Powder-bed fusion holds roughly 0.4–0.6 mm channels once you account for trapped powder. Material extrusion needs 0.8 mm or wider, because the nozzle drags and the walls slump.
Wall thickness sets two things at once: stiffness and print time. A 0.6–1.0 mm wall survives handling and moderate back pressure. Anything below 0.4 mm tends to warp during cooling or delaminate under load.
Lattice walls give better strength per gram than straight fins. A gyroid or diamond cell spreads load in three directions, so a 0.5 mm wall can carry the same pressure drop as a 0.9 mm straight wall.
Pressure drop is the number that kills designs. Pack too much surface into a duct and the blower eats the benefit. We size channels so the clean-filter drop stays under about 500 Pa at design flow, then check the loaded case separately.
For lab-scale test rigs, printed coupons let you sweep three or four channel widths in a single build. That is faster than machining each variant, and the comparison stays honest because the material batch is the same.
Material choice for printed capture media
The printed part is usually a support structure, not the sorbent itself. It holds the active material in place, so chemical resistance and thermal stability matter more than strength. A filter that swells or softens after 200 hours of amine exposure is scrap.
PA12 and PA11 handle humid gas well and print with good resolution. They are the default for benchtop rigs. PEEK survives higher temperatures and steam regeneration, but it needs a heated chamber and costs far more per part.
Photopolymer resin prints the finest channels available, down to 0.2 mm in SLA. It is fine for flow-visualization models. It is a poor choice for real capture work because most resins creep and embrittle under UV and amine contact.
When the housing carries load, print the lattice and machine the frame. We routinely run 6061-T6 or 316L frames on our 5-axis centers and bond or bolt them to printed cores. The frame holds the flange flat; the print does the surface-area work.
Filter media can also be coated after printing. Dip-coating a printed lattice in a slurry leaves the active phase on the walls while the polymer takes the mechanical load. Washcoat thickness of 50–200 μm is typical.
Where each printing route stops working
Powder-bed fusion gives the best geometry freedom and the best as-built tolerance, around ±0.1 mm on small features. It also traps powder in closed cells. Any internal volume you cannot blow out will stay filled, so design drain paths from the start.
Material extrusion is cheap and fast for large ducts. It cannot hold thin walls over tall spans, and the seam between layers is the weak point under vibration. Use it for prototypes and ducting, not for a 4,000 mm production stack.
SLA and DLP win on feature size. They lose on part size and on long-term material stability. Good for a 150 mm test coupon, poor for a full-scale module.
Metal binder jetting and DMLS open the door to stainless and Inconel lattices that survive 400 °C regeneration. Cost per part is high, and surface finish as-built is Ra 8–12 μm, which needs post-processing if you care about fouling.
Whichever route you pick, the first article should be inspected. We measure wall thickness, channel width and flange flatness on a CMM and report the numbers before the run continues.
Comparing build routes for carbon dioxide capture filters
Numbers are typical for small to medium filter cores, not guarantees.
| Route | Min channel | Wall range | Best fit |
|---|---|---|---|
| Powder-bed fusion (PA12) | 0.4–0.6 mm | 0.6–1.0 mm | Complex lattices with drain paths |
| Material extrusion | 0.8 mm | 1.0–1.6 mm | Large ducting and early prototypes |
| SLA / DLP resin | 0.2 mm | 0.4–0.8 mm | Flow-visualization coupons |
| Metal DMLS | 0.5 mm | 0.5–1.2 mm | High-temperature regeneration duty |
| Printed core + machined frame | 0.4 mm | 0.6 mm | Production modules with flat flanges |
Pick the route before you pick the geometry
If you need the most surface area in a small volume and can tolerate polymer, print the lattice in PA12 with drain paths and a machined 6061-T6 frame. If the filter sees steam regeneration above 200 °C, go to a metal lattice and accept the higher cost per part. Do not print thin walls in resin and expect them to survive a real gas stream.
Questions engineers ask before printing
Can a printed filter core be sealed to a metal housing?
Yes. The usual approach is a machined flange bolted to the printed core with an O-ring or gasket in between. We keep the flange flat within 0.05 mm so the seal does not leak.
If the gas is aggressive, add a PTFE gasket and keep the printed polymer out of the sealing face.
How do I stop powder staying inside the lattice?
Design every internal channel so it opens to at least two external faces. Add a 2–3 mm purge hole at the lowest point of each cell.
After printing, we bead-blast and blow out the core, then weigh it. A weight above the CAD estimate means trapped powder.
Does surface roughness help or hurt capture?
Roughness increases wetted area and helps liquid distribution in a washcoated core. It also traps fine dust and raises pressure drop over time.
For clean gas, keep the as-built surface. For dust-loaded flue gas, plan a post-process polish on the inlet channels.
What lead time should I expect for a printed core plus machined frame?
We quote and return a DFM analysis within 12 hours. Production can start within 24 hours of approval, and parts ship in 3–5 days for standard runs.
Metal lattice work takes longer because of post-processing and inspection.
Can you print one prototype and then scale to 10,000 parts?
Yes. There is no minimum order quantity, so a single prototype and a 10,000-part run use the same process window.
For volume, we usually switch the frame to die casting or machining and keep the lattice printed if the geometry still needs it.
How is the finished core checked?
We inspect 100% of parts before shipment: raw material check, in-process monitoring, and final inspection. Reports are available on request.
For filter cores we measure channel width, wall thickness and flange flatness, and record the numbers against the drawing.
Send the core geometry and we will flag the print risks
Upload your STEP file and get a quotation with a free DFM analysis within 12 hours. We will tell you which features will not print and what to change.
12-hour quote100% inspectionNDA on request
More process notes
We publish setup notes, tooling trials and inspection data from the factory floor.
