Artificial silk is made by dissolving a silk protein or a cellulose pulp into a concentrated liquid dope, extruding that dope through fine spinnerets into a coagulating bath, and then drawing the forming filament to align its polymer chains. The result is a continuous fiber that mimics the hydrogen-bonded beta-sheet structure of natural silk, though its properties depend heavily on the feedstock and the spinning route.
The core problem is chemical. Natural silk is spun by silkworms at room temperature in water, using proteins that self-assemble into crystallites. Industrial processes instead force a viscous polymer solution through a tiny orifice and precipitate it into a solid thread. Every step in that chain — dissolving, extruding, coagulating, drawing, finishing — determines whether the output behaves like silk or like generic plastic.
Mechanism: turning polymer dope into a filament
The feedstock defines the process. Regenerated silk fibroin comes from cocoons that are degummed to strip sericin, then dissolved in a salt solution such as lithium bromide or a calcium chloride–ethanol mix. The dissolved protein is dialyzed against water to remove salts and concentrated into a spin dope, typically 15 to 30 percent protein by weight. Viscosity matters: too thin and the filament breaks, too thick and it clogs the spinneret.
Alternative feedstocks include cellulose (viscose and lyocell routes) and biosynthetic spider silk proteins produced in yeast, bacteria, or plants. Cellulose is dissolved in sodium hydroxide and carbon disulfide to make viscose, or in a direct solvent such as N-methylmorpholine N-oxide for lyocell. Each route changes the chemistry of the finished fiber, but the physical sequence is the same.
Wet spinning is the dominant method. The dope is pumped through a spinneret with holes measured in micrometers and extruded into a coagulation bath — often a salt solution, alcohol, or acid — where the solvent diffuses out and the polymer precipitates as a gel filament. Dry spinning evaporates the solvent in a warm air column instead. Dry-jet wet spinning places an air gap between the spinneret and the bath, which lets the polymer chains pre-align before precipitation and typically raises tenacity.
Drawing is where strength is born. Freshly coagulated filament has tangled, randomly oriented chains. Stretching it at a controlled ratio — commonly 2x to 6x, sometimes more for high-performance grades — pulls those chains into parallel alignment and increases crystallinity. The combination of coagulation chemistry and draw ratio is the single largest lever on the fiber's tensile properties. In plain terms, you are combing the molecular spaghetti into straight strands.
Finishing steps remove residual solvent, apply a spin finish for lubrication, and dry the yarn under tension. Residual solvent is a real quality issue: incomplete washing can leave salts or sulfur compounds that degrade the fiber over time.
Quantified context: what the numbers say
The global silk market is small relative to synthetic fibers, and artificial silk covers a family of products that price very differently. Viscose rayon, the oldest commercial artificial silk, sells as commodity textile fiber at low cost per kilogram. Regenerated silk fibroin and recombinant spider silk sit at the opposite end, priced as specialty materials because yields are low and processing is complex.
A comparison table shows how far the routes diverge.
| Fiber route | Feedstock | Typical spinning method | Draw ratio range | Notes |
|---|---|---|---|---|
| Viscose rayon | Wood pulp cellulose | Wet spinning | 1.5x–3x | Mature, low-cost, uses carbon disulfide |
| Lyocell | Wood pulp cellulose | Dry-jet wet spinning | 2x–5x | Direct solvent, solvent recovery is key |
| Regenerated silk fibroin | Degummed cocoons | Wet or dry-jet wet spinning | 2x–6x | Protein-based, batch-limited feedstock |
| Recombinant spider silk | Engineered yeast or bacteria | Wet spinning | 3x–8x | Highest potential tenacity, lowest yield |
| Modified cellulosic (acetate) | Cellulose acetate | Dry spinning | 1x–2x | Lower strength, used in apparel blends |
The ranges are process windows, not guarantees. A single fiber line may run narrower bands tuned to its own dope formulation and target denier.
Scale is the binding constraint. Viscose and lyocell run at industrial tonnage. Regenerated fibroin and recombinant proteins run at pilot or small commercial scale, where the cost per kilogram remains far above commodity textile fiber. That gap, not chemistry, is why artificial silk has not displaced conventional silk in high-volume apparel.
Edge cases: where the process breaks down
Protein dopes are unstable. Silk fibroin solutions can gel or aggregate during storage, changing viscosity and creating defects that survive into the fiber. Temperature and holding time are controlled tightly for that reason.
Solvent choice carries hidden costs. Viscose relies on carbon disulfide, a hazardous reagent that demands rigorous recovery and worker protection. Lyocell avoids it but requires near-complete solvent recovery to stay economical and environmentally defensible. The process is often judged on recovery rate as much as on fiber properties.
Recombinant proteins behave differently from native fibroin. Their molecular weight distribution is broader, and they may lack the exact repeat sequences that give natural silk its crystallinity. Spinning conditions developed for cocoon fibroin do not transfer directly, so each expression host needs its own dope and draw protocol.
Moisture and heat sensitivity are common failure points. Regenerated protein fibers can lose strength when wet and may yellow with heat exposure. That limits use in demanding applications even when dry tenacity looks competitive.
Misconceptions: what artificial silk is not
The biggest misconception is that artificial silk is a single material. It is a category spanning cellulose-based rayon, regenerated protein, and recombinant protein fibers with very different properties and supply chains. Treating them as interchangeable leads to bad sourcing decisions.
A second misconception is that artificial silk is a synthetic plastic. Viscose, lyocell, and regenerated fibroin all begin with a natural polymer rather than a petrochemical monomer, although the processing chemicals can be synthetic.
The third is that "silk-like" equals silk performance. Luster and drape can be matched closely, but tensile strength, elastic recovery, and moisture behavior often are not. A fabric can look like silk and behave nothing like it in wear.
Watch for these red flags when evaluating claims:
A label says "silk" but the fiber content lists viscose or rayon; these are regenerated cellulose, not protein. Marketing cites "spider silk" without stating scale; most recombinant spider silk remains pilot-scale. A product claims high strength but omits wet strength, where protein fibers often drop. Solvent recovery or wash-water data is missing from environmental claims. A specification sheet gives tenacity without a draw ratio or test standard. Price far below conventional silk for a "protein" fiber usually signals a cellulose blend.
That last point is the counterintuitive one: in textiles, the cheapest "silk" is almost never protein-based.
Practical actions: how to specify and test
If you are sourcing or evaluating artificial silk, anchor decisions to measurable fiber properties rather than marketing language. Request the fiber content by generic name and confirm the spinning route where it matters.
Concrete steps:
Verify the fiber identification on the care label against the actual generic name, such as viscose, lyocell, or silk fibroin. Ask for tenacity and elongation at break, both dry and wet, with the test standard cited. For protein-based fibers, request the draw ratio and coagulation chemistry as process evidence. Ask about residual solvent or salt content and the washing protocol used to remove it. Confirm the scale of supply; pilot-scale protein fiber cannot support tonnage orders. Test dyeability and heat sensitivity on a sample before committing to a production run.
Thresholds to reconcile explicitly: draw ratios reported by different suppliers are not directly comparable unless the starting dope concentration and coagulation bath are also disclosed. A 5x draw on a 25 percent dope is not the same material as a 5x draw on a 15 percent dope. Compare like with like, or ask for finished-fiber data instead.
Limitations: what the process cannot fix
No spinning route overcomes a poor feedstock. Low molecular weight protein produces low crystallinity, and no draw ratio restores what is not there. Process optimization has a ceiling set by polymer quality.
Scale economics remain the hard limit for protein fibers. Recombinant production yields grams to kilograms per liter of culture depending on host and construct, and downstream purification adds cost. Until those yields rise substantially, artificial protein silk will stay in specialty applications.
Environmental claims need scrutiny. Viscose and lyocell depend on solvent recovery and pulp sourcing; protein routes depend on fermentation energy and purification water. A fiber can be bio-based and still carry a heavy processing footprint.
Finally, artificial silk is a moving target. Dope formulations, spinneret designs, and draw protocols are active research areas, so published property ranges age quickly. Treat any single specification as a snapshot of one process window, not a fixed property of the category. The mechanism is stable; the numbers are not.