Bio-based FDM filaments

Fused deposition modelling extrudes a molten polymer filament and lays it down in tracks. The part’s properties come as much from how those tracks bond as from the polymer itself.

Layer adhesion is chain interdiffusion

When a hot track is laid against a previously deposited one, the two surfaces fuse only if polymer chains diffuse across the interface and entangle. That requires the interface to be above the glass transition — for a semi-crystalline polymer, near the melt — for long enough for chains to reptate across.

The competing process is cooling, which is fast: a thin track in open air loses heat in a fraction of a second. So the weld is always partial, and strength in the build direction is lower than within a layer. Anisotropy is intrinsic to the process, not a defect of tuning, and it is why a printed part’s specification cannot be read off a resin datasheet.

Everything that helps — a heated chamber, higher nozzle temperature, thinner layers, slower printing — works by extending the time the interface spends hot enough to diffuse.

Why PLA is the default, and where it stops

PLA dominates for reasons that are all thermal and rheological. Its melt viscosity is low and its melting range modest, so it extrudes easily at accessible nozzle temperatures. Its thermal expansion coefficient is low, and it crystallises slowly, so a printed part is largely amorphous and shrinks little as it cools. Warping and lifting from the bed — the failure mode that dominates with faster-crystallising polymers — is minimal.

The cost is heat resistance. PLA’s glass transition is roughly 55–60 °C; above it the amorphous polymer softens and a part loses stiffness and creeps under its own load. A car interior in summer exceeds this. So does a dishwasher.

That number should look familiar. It is the same glass transition that makes PLA compostable at 58 °C and not in soil or seawater, discussed under biopolymers and bioplastics. A single material property explains the printing behaviour, the service limit and the end-of-life behaviour — and it means the property cannot be improved in one place without moving in the others.

Annealing raises the heat deflection temperature by letting PLA crystallise after printing, since crystallites persist above the glass transition. The trades are real: the part shrinks and distorts as it crystallises, and dimensional accuracy suffers. Nucleating agents and PLA grades containing more D-lactide or less are used to shift the crystallisation behaviour deliberately.

Moisture is a chemical problem, not a cosmetic one

Most of the interesting bio-based filaments — PLA, PHA, bio-polyamides, PETG — are polyesters or polyamides, and their backbones are hydrolysable.

At room temperature this is slow enough to ignore. At melt temperature it is not. Absorbed water in a filament hydrolyses the chain during extrusion, cutting molecular weight, and lower molecular weight means fewer entanglements and a weaker, more brittle part. The water also flashes to steam in the nozzle, giving the popping, stringing and rough surface that is the visible symptom.

The visible defect and the invisible one have the same cause, but drying fixes only the part that has not happened yet — chain scission already suffered is permanent. Filament is therefore dried before printing and stored desiccated, and this is a material requirement rather than a preference.

Fibre-filled bio-composite filaments — wood flour, hemp, flax in a PLA matrix — change the picture in a further way: the filler is itself hygroscopic and holds far more water than the polymer, and it abrades brass nozzles, so hardened nozzles are standard.