Atlas of Polymers

Tools

Molecule LookupDraw and measure any small moleculeProperty PredictorFour models, from structure aloneCompare PolymersUp to three, side by side

Presentation

Reading

About

About the AtlasWhat this is, and how to get in touch

The Smart Materials Era (2000-2015)

2014

Recyclable Thermosets

The Polymer That Finally Learned to Let Go

thermoset·vitrimer-crosslinked-network·Jeannette Garcia

On 20 January 2014, a room full of engineers at the European Space Operations Centre in Darmstadt punched the air and hugged each other as a signal finally arrived from a spacecraft that had gone silent on purpose. Rosetta had spent the coldest, most distant leg of its ten-year chase after a comet in hibernation (two and a half years with its systems powered down and nobody on Earth able to confirm it had survived), and its internal alarm clock had just done exactly what it was set to do.

Plate I

A group of engineers in a space mission control room, several with fists raised and mouths open in celebration, standing at rows of monitors below screens reading 'ESA'.
Mission control at ESOC, Darmstadt, on 20 January 2014, the moment Rosetta's signal confirmed it had woken from hibernation on schedule.Wikimedia Commons

Ten months of precision followed, and then, on 12 November, the part of the plan nobody could fully engineer around: Rosetta’s little lander, Philae, made first contact with comet 67P/Churyumov–Gerasimenko right on its intended target, and its anchoring harpoons failed to fire. It bounced twice, drifted for the better part of two hours in the comet’s near-nonexistent gravity, and came to rest wedged sideways in the shadow of a cliff, far from the sunlit spot the mission had spent a decade planning for. The landing succeeded. It also went nothing like the plan.

Plate II

A grainy black-and-white image of an irregularly shaped, cratered comet nucleus made of two lobes joined by a narrower neck, photographed against the blackness of space.
Comet 67P/Churyumov–Gerasimenko, photographed by Rosetta days before Philae's landing attempt, a decade of navigation ending in a bounce nobody had planned for.Wikimedia Commons

That same year, an accident of the opposite kind happened in a chemistry lab in California, and it stuck the landing completely. In her first week at IBM Research–Almaden, the polymer chemist Jeannette Garcia was given a routine task: combine two reagents, heat the mixture, and weigh out a third. This was part of an attempt to build a stronger material around a solution of broken-down plastic bottles. She never got to the third ingredient. By the time she came back to the flask, the first two alone (paraformaldehyde and an aromatic diamine) had already set into a white solid so hard she needed a hammer to knock it loose from the glass.

Plate III

A steel-headed claw hammer with a black and yellow rubber-gripped handle, resting on a concrete floor.
An ordinary claw hammer, the tool Garcia actually needed to free her accidental discovery from its flask.Wikimedia Commons

A Failure Worth Not Throwing Away

Most chemists would have washed the flask out and started again. Garcia broke the solid free and kept it, because whatever had formed was not supposed to be possible from just those two ingredients, and she wanted to know what it was. Working with colleagues back at Almaden, she and her group identified the material as a new kind of thermoset network, built from hemiaminal and hexahydrotriazine linkages rather than the epoxy or phenolic chemistry that had defined the category for a century. They published the discovery in Science in May 2014, and gave the new family a name that undersold nothing: Titan.

Plate IV

An aerial view of a low, dark, multi-winged office building set among dry golden hills, reached by a single winding access road, with a small town visible in the hazy valley beyond.
IBM Research–Almaden, San Jose, where Garcia's group turned an accident in a flask into a published discovery.Wikimedia Commons

The Trick Was in the Kind of Bond, Not the Ingredients

Every earlier thermoset (Bakelite in 1907, the epoxies and unsaturated polyesters that followed it) gets its permanence from crosslinks that, once formed, are simply irreversible: there is no chemical switch that can be thrown afterward to take them apart again cleanly. Titan’s hexahydrotriazine crosslinks are just as strong under everyday conditions, which is what makes the material rigid and solvent-resistant enough to behave like any other engineering thermoset. But they share a property nothing before them in this category had: soak the cured material in a sufficiently strong acid, and those same linkages come apart, unzipping the network back down into its original small molecules rather than merely swelling or dissolving. Recover those monomers, and you can build the network again from scratch. Nothing about the chemistry needed a mould, a catalyst discovered specifically for this purpose, or years of targeted synthesis; it needed a mistake that somebody was curious enough not to discard.

What the Data Actually Says

Because this page represents one specific, well-characterised chemistry rather than a whole category, its numbers are real rather than placeholders, so they are worth citing precisely: Titan holds its rigidity to well past the boiling point of water, and does not begin to break down chemically until it is hotter still, comfortably beyond the working range of most engineering thermoplastics. Its stiffness sits closer to a fibre-reinforced composite than to an ordinary cast resin, which is part of why the earliest press coverage reached for comparisons to bone. Under normal handling it shrugs off solvents and mild acids the way any good thermoset should; only a strongly acidic soak, well outside anything the material would meet in service, triggers the controlled unzipping that makes it recyclable.

One Material, Not a Whole Category

It is worth being clear about what this page is and is not. Titan is a single, specific chemistry: paraformaldehyde and an aromatic diamine, built into a hexahydrotriazine network, discovered essentially by accident at one company in one year. It sits in the same broad family as vitrimers, the covalently adaptable networks covered elsewhere in this Atlas, but the two arrived by opposite roads: vitrimers came from a research group asking what a network with catalysed, heat-driven bond exchange would do, while Titan came from a chemist noticing that an ordinary bench reaction had gone somewhere nobody had sent it.

Two Landings, One Year

Rosetta’s team spent a decade engineering a landing that still went sideways into a shadowed crevice, and turned it into a scientific success anyway by refusing to write the mission off. Garcia spent a few unsupervised minutes on a bench reaction that went somewhere nobody had planned, and turned it into a new category of material by refusing to pour it down the sink. Both are 2014 stories about what happens when careful preparation meets an outcome nobody drew up in advance, and both ended the same way: with someone deciding the accident was worth keeping.

values with [n] cite the numbered references·estimates are flagged·“not yet available” and “N/A” are honest states, not gaps

Abbreviation
—
Type
polymer family (hub)
CAS number
None (heterogeneous class or not assigned)
Resin ID code
none assigned
Formula
Represented by IBM's polyhexahydrotriazine (PHT, nicknamed 'Titan'), discovered accidentally in 2014. This is a specific dynamic-covalent-network chemistry rather than a fixed general repeat unit for 'recyclable thermosets' as a category, which is closely related to (but distinct from) vitrimers.
Repeat unit (BigSMILES)
Represented by IBM's polyhexahydrotriazine (PHT, nicknamed 'Titan'), discovered accidentally in 2014. This is a specific dynamic-covalent-network chemistry rather than a fixed general repeat unit for 'recyclable thermosets' as a category, which is closely related to (but distinct from) vitrimers.
IUPAC name
—
Synonyms
polyhexahydrotriazine; PHT; Titan
Also known as
polyhexahydrotriazinePHTTitan

Chemical family
vitrimer-crosslinked-network
Backbone class
heterochain
Polymerization mechanism
step-growth-condensation
Constitutional monomer
Paraformaldehyde4,4'-Oxydianiline
Polymer class
thermoset

Year of origin
2014
Era
The Smart Materials Era (2000-2015)
Key figures
Jeannette Garcia
Events referenced
Rosetta wakes from hibernation, European Space Operations Centre (20 January 2014) · Philae becomes the first spacecraft to land on a comet nucleus, 67P/Churyumov–Gerasimenko (12 November 2014)

Polymerization type
step-growth condensation
Common monomers (feedstocks)
paraformaldehyde, 4,4'-oxydianiline
Catalysts
not yet available

IBM Research chemist Jeannette Garcia accidentally discovered this new thermoset family in her first week on the job in 2014, when a routine 3-ingredient mixture solidified into an unexpectedly hard material she needed a hammer to break free. She and colleagues published the discovery in Science in May 2014, nicknaming the material 'Titan'. Unlike conventional thermosets, which resist remolding, this polymer can be reprocessed via a chemical (acid-triggered) reaction that reverses the network back to its monomers, enabling true recycling.

Tacticity
not yet available
Crystal structure
not yet available
Typical crystallinity
0 %estimate[1]Amorphous crosslinked network.

Molecular weight

Number average (Mn)
not yet available
Mass average (Mw)
not yet available
Dispersity (Mw/Mn)
not yet available

Mark-Houwink constants

not yet available

Rigid, densely crosslinked network like a conventional thermoset, but the hemiaminal/hexahydrotriazine linkages can be selectively reversed under acidic conditions, unzipping the network back to monomers; this is the key innovation over conventional irreversible thermoset crosslinks.

Density
not yet available
Melt flow index
Not applicable
Refractive index
not yet available
Transmittance
not yet available
Haze
not yet available
Gloss
not yet available
Water absorption
not yet available
Dielectric constant
not yet available
Dielectric strength
not yet available
Electrical conductivity
Not applicable

Glass transition (Tg)
190 °C[2]polyhexahydrotriazine (PHT/"Titan"), the specific chemistry this page represents
Melting temperature (Tm)
Not applicableThermoset; does not melt.
Crystallization (Tc)
Not applicable
Heat deflection (HDT)
not yet available
Decomposition onset
300 °C[2]polyhexahydrotriazine (PHT/"Titan"), the specific chemistry this page represents
Thermal conductivity
not yet available

Tensile modulus
10000 MPa[2]Young's modulus, polyhexahydrotriazine (PHT/"Titan"), the specific chemistry this page represents
Yield strength
not yet available
Tensile strength at break
not yet availableDescribed qualitatively as an 'exceptionally strong plastic' comparable in application scope to conventional industrial thermosets; no specific number sourced.
Elongation at break
not yet available
Impact strength (Izod)
not yet available
Impact strength (Charpy)
not yet available
Hardness
not yet available
Flexural modulus
not yet available
Poisson's ratio
not yet available
Coefficient of friction
not yet available

Weathering / UV
not yet available
Hydrolysis resistance
stable to solvents at pH > 3; depolymerizes back toward a viscous, remoldable state under strongly acidic conditions (e.g. sulfuric acid), the basis of its chemical recyclability[2]polyhexahydrotriazine (PHT/"Titan"), the specific chemistry this page represents
Flammability (UL94)
not yet available
Limiting oxygen index
not yet available
Solubility parameter (δ)
not yet available

Gas permeability

not yet available

Polymer-solvent interaction parameter (χ)

not yet available

Processing methods
casting/molding (initial cure)acid-triggered depolymerization (recycling step)
Drying required
not yet determined
Processing temperature
not yet available
Shrinkage rate
not yet available

  • General industrialsmartphone components · aircraft partsConventional thermosets (the category this material targets replacing) make up roughly one-third of global polymer production and are historically very difficult to recycle.

Recyclable
Yes
Biodegradable
No
Degradation pathway
Deliberately reversible network chemistry: an acid treatment breaks the crosslinks back down to monomers/oligomers for genuine chemical recycling, unlike conventional thermosets.

The whole point of this material class: giving thermosets (historically near-impossible to recycle) a real end-of-life recycling pathway.

LD50 (oral, rat)
not yet available
NFPA health
not yet available
NFPA flammability
not yet available
NFPA reactivity
not yet available
Carcinogenic classification
not yet available

  1. [1]Jeannette Garcia's accidental discovery of a recyclable thermoset (IBM, 2014)Web search summary (Scientific American, Jeannette Garcia Wikipedia)Accessed 2026-07-14https://en.wikipedia.org/wiki/Jeannette_Garcia[search-recyclable-thermoset]
  2. [2]Handbook of PolymersChemTec Publishinghttps://www.worldcat.org/isbn/9781895198928[handbook-wypych-2016]

Illustrations

  1. Plate IMission control at ESOC, Darmstadt, on 20 January 2014, the moment Rosetta's signal confirmed it had woken from hibernation on schedule.ESA - Jürgen Mai · CC BY-SA 3.0 igoWikimedia Commons
  2. Plate IIComet 67P/Churyumov–Gerasimenko, photographed by Rosetta days before Philae's landing attempt, a decade of navigation ending in a bounce nobody had planned for.ESA/Rosetta/NAVCAM · CC BY-SA 3.0 igoWikimedia Commons
  3. Plate IIIAn ordinary claw hammer, the tool Garcia actually needed to free her accidental discovery from its flask.J.C. Fields · CC BY-SA 3.0Wikimedia Commons
  4. Plate IVIBM Research–Almaden, San Jose, where Garcia's group turned an accident in a flask into a published discovery.Dicklyon · CC BY-SA 4.0Wikimedia Commons