TL;DR — Only one of the six filaments spun into commercial floss is not a plastic: silk. PLA is a plant-derived plastic, not an absence of one. Measured as finished retail floss rather than raw polymer, tensile strength runs from 194.18 ± 24.61 MPa (UHMWPE) through 49.28 ± 9.07 MPa (nylon) down to 11.78 ± 0.77 MPa (PTFE tape), with silk statistically indistinguishable from nylon. And every one of them, PLA included, belongs in household trash under the municipal guidance we could find — the disposal route any label claim has to be measured against.
Six filament materials are actually in commercial dental floss production: nylon, PTFE, polyester, UHMWPE, silk and PLA. Only one of them — silk — is not a plastic. PLA is a plant-derived plastic, not an absence of plastic. And in the single study that measured four of these materials as finished retail floss rather than as raw polymer, tensile strength ranged from 194.18 ± 24.61 MPa for UHMWPE down to 11.78 ± 0.77 MPa for PTFE, with nylon at 49.28 ± 9.07 MPa and silk not significantly different from nylon.
Figure 1 — Tensile strength, measured as finished retail floss
I am an industrial designer, not a dentist. What follows is a materials description, not clinical advice. If your gums bleed, hurt, or your interdental spaces have changed, that is a conversation with a dental professional, not with an article.
This page is the materials reference for the cluster. If what you want is a buying decision, the category ranking scores whole product categories — holders, picks, spooled floss — on certification, disposal and cost per meter, and ends in a recommendation. This one stays with the fiber.
The five criteria, applied identically to each material
Every section below answers the same five questions: what the fiber is made of, how it behaves passing a tight contact, how it holds up mechanically, what happens to it after use, and what to watch for. Two of those five are far better documented than the other three, and I will say so where the data thins out.
One structural distinction governs most of the rest. Nylon, silk and UHMWPE floss are multifilament — many fine filaments twisted into a thread. PTFE floss is not a thread at all: it is described as a broad, sheet-like, layered structure. That single difference explains most of what people notice in the mouth.
Nylon (multifilament)
Made of polyamide filaments, spun and twisted, then usually wax-coated. Patent literature from the category names ceresin, ozokerite and beeswax at 5–30 % by weight, applied for glide alone.
How it flosses. In a single-blind crossover trial measuring insertion and removal force through 14 interproximal contacts in 27 people, PTFE flosses passed strong proximal contacts with less force than nylon. Nylon needs more push.
Durability. Mid-range on the numbers, 49.28 ± 9.07 MPa. But scanning electron microscopy of commercial floss before and after real use found the opposite of what the category assumes: the sample with clearly fractured filaments was a flat, compacted floss, while the two genuinely twisted multifilament threads showed only minor visible structural damage. One flat sample went from roughly 50 % elongation at break when unused to about 150 % after use — the internal structure was destroyed even though breaking force fell by only about 10 N. Fraying is filament fracture, not surface wear. That study identified its samples by product, not by polymer, so read the link to nylon as inference from the category.
Worth the honesty: in the same user-preference study that produced the tensile numbers, a large majority of responses disagreed that any of the four floss types frayed or shredded in use. The widely repeated claim that nylon shreds is more marketing than measurement.
After use. Household trash. Not compostable under any scheme.
Watch for. Nylon was the least preferred of four materials by the 16 participants in that preference test — a small, subjective sample.
PTFE (monofilament tape)
Made of polytetrafluoroethylene, extruded and expanded into a flat, layered band.
How it flosses. Best in class, and measurably so: lowest surface roughness of the four materials tested (Ra 0.0478 ± 0.0034 µm, against 0.10 ± 0.022 µm for nylon), lowest insertion force, most preferred by users.
Durability. The 11.78 MPa figure needs an explanation, or it misleads. Tensile strength there is calculated over the whole cross-section, and a flat tape has a large one. It does not mean PTFE tape snaps easily in the mouth.
After use. Trash. PTFE is a fluoropolymer, and a life-cycle review concluded there is no scientific rationale for treating fluoropolymers as low concern or as separate from other PFAS — an argument about manufacturing, processing aids and disposal, not about what happens in your mouth.
Watch for. PTFE is itself a PFAS, and the evidence on what that means for your body runs in both directions: a small cross-sectional study associated PTFE-floss use with higher serum PFHxS, while a far larger NHANES analysis found floss users had lower overall serum PFAS. The PFAS evidence review in this cluster works through both studies — designs, sample sizes, confidence intervals, and the reformulation of the most-studied PTFE product — in full.
Polyester
Made of PET or similar polyester filaments, often the actual filament behind products marketed on some other material entirely.
How it flosses, durability, after use. I found no peer-reviewed measurement of polyester floss specifically — it was not among the four materials mechanically tested. Contract manufacturers list it as a standard filament; consumer round-ups almost never mention it. Treat any performance claim about it as unverified.
Watch for. Polyester is the most common answer to "what is this actually made of" when a plastic-free-looking product does not say.
UHMWPE
Made of ultra-high-molecular-weight polyethylene, gel-spun into fiber. This is the strongest material in the category by a wide margin: 194.18 MPa as finished floss, roughly four times nylon.
How it flosses. Surface roughness not significantly different from nylon. Strength is the point, not slip: it is what you use when the filament has to survive a very tight contact without breaking.
After use. Polyethylene. Trash.
Watch for. No published data on its abrasion resistance in a floss context. It is also rarely labeled as such at retail.
Silk
Made of degummed Bombyx mori fibroin, twisted into a multifilament thread and — this matters — coated in wax, usually beeswax or candelilla.
How it flosses. Highest surface roughness of the four tested materials, Ra 0.3035 ± 0.0254 µm, roughly six times PTFE. That grip is either the appeal or the problem, depending on how tight your contacts are.
Durability. Statistically indistinguishable from nylon as finished floss. The 740 MPa figure that circulates for silk is raw fiber, not a coated retail thread.
After use. This is where the marketing outruns the evidence. Silk is a protein and is enzymatically degradable in principle, but I found no published measurement of degradation time for a waxed, flavored commercial floss. Meanwhile one of the largest US municipal organics programs lists dental floss among items not accepted, and rejects wax and waxed products unless certified — waxed silk floss is excluded twice over.
Watch for. Silk is not vegan. And on the abrasion test against human enamel, all four materials showed no significant difference — silk's roughness is a feel difference, not a documented enamel risk.
PLA (corn-derived)
Made of polylactic acid, melt-spun, wax-coated. Sold as compostable floss.
Figure 2 — What actually happened to PLA fibre at three disposal temperatures
How it flosses. No peer-reviewed study of PLA in floss form exists that I could find. Category sources consistently describe it as the least resistant material, poorly suited to tight contacts. That is consensus, not measurement.
After use. Here the numbers are good, and they are unforgiving. Melt-spun PLA fibers disintegrated after three weeks at 58 °C under ISO 20200, while home-composting samples at 28 °C showed no disintegration up to 14 weeks; under ambient weathering at 25 °C the yarn kept its mechanical properties over 80 days. PLA's glass transition sits near 55–60 °C, above any garden heap. Those were lab yarns, not floss, so treat the direction as reliable and the timings as indicative.
Watch for. The label language. Under US rules it is deceptive to make an unqualified degradable claim for items entering the solid waste stream if they do not completely decompose within one year after customary disposal — and customary disposal for floss is the trash can. Any "compostable" claim on floss should name the facility type, the temperature and the standard. Most do not.
"Bamboo charcoal" floss
There is no verifiable filament here. Mechanical bamboo fiber is too short and stiff to spin into a floss filament, so the thread is something else — polyester or nylon — carrying charcoal powder. Competing sources give flatly incompatible compositions, none with a primary reference. Treat the category as unspecified until a manufacturer publishes the polymer.
Verdict table
| Material | Tensile as retail floss | Surface roughness Ra | Passes tight contacts | End of life | Main caveat |
|---|---|---|---|---|---|
| UHMWPE | 194.18 ± 24.61 MPa | ≈ nylon | Strongest option | Trash | Rarely labeled; no abrasion data |
| Nylon | 49.28 ± 9.07 MPa | 0.10 ± 0.022 µm | Needs more force | Trash | Shredding claim poorly supported; least preferred in a small user test |
| Silk | ≈ nylon | 0.3035 ± 0.0254 µm | Grippy, not slippery | Trash; wax blocks composting | Not vegan; no degradation data on coated floss |
| PTFE | 11.78 ± 0.77 MPa (flat tape) | 0.0478 ± 0.0034 µm | Best glide, least force | Trash | PFAS association contested both ways |
| PLA | Not measured as floss | Not measured | Reported weakest | Needs ≈58 °C industrial composting | Label claims routinely unqualified |
| Polyester | Not measured | Not measured | Unknown | Trash | Often the hidden filament in other categories |
What the table cannot decide for you
Notice what none of the six materials fixes: the handle. Disposable picks and dispensers are the other half of the floss-waste question, and it is exactly the half reusable holders are sold on — in the one life-cycle study of interdental aids, the polypropylene handle alone accounted for 49 % of a floss pick's carbon footprint. That trade-off, and the full ranking of holders, picks and spooled floss, lives in the category ranking — go there to choose a product; come back here when you want to know what a filament actually is.
If you want the raw consumption number for your own math: the ADA's consumer guidance of about 18 inches of floss, once daily works out to roughly 167 m — about 550 feet — per person per year. That is an upper bound on an ideal-compliance assumption, and it is the honest input for any cost or plastic calculation. From here, Silk vs Nylon vs PLA puts the three most-marketed filaments head to head, the label checklist ranks what the package terms legally mean, and the PFAS evidence review covers the fluoropolymer question at length.
