You know what's funny? I wrote my name on a plastic container with a Sharpie three months ago. It's gone. Meanwhile, there's a piece of papyrus in the Egyptian Museum in Turin that someone wrote on four thousand years ago, and you can still read every character. So what gives? How did ancient Egyptians nail permanence with lamp soot and tree sap, while a twenty-first-century marker can't survive a few dishwasher cycles?
Here's the thing — this isn't just a curiosity. The archival ink market is projected to hit four point two billion dollars in the next two years. Museums are digitizing and preserving, forensics labs need tamper-proof marking, industrial supply chains demand tracking that survives heat and solvents. And yet, most people walking through the stationery aisle have no idea there's a difference between "permanent" and "archival." They're not the same thing at all.
Daniel sent us this one, and I think it gets at something deeper than just "which pen should I buy." Humans have been trying to make lasting marks since — well, since we were human. Cave paintings, clay tablets, stone inscriptions. The question is: can a surface coating ever truly compete with etching or engraving for durability? What are the most permanent marking technologies we've ever invented?
That's a five-thousand-year story, and the chemistry is genuinely fascinating. You've got pigment-based marks on one side — inks, paints, coatings that sit on top of a surface — and substrate modification on the other, where you physically alter the material itself. The central tension is whether any coating can match a structural change in pure staying power.
I think the answer turns out to be more interesting than just "no, etching wins." There are pigments that get fired into ceramics at eight hundred degrees Celsius and essentially become part of the glaze. There are industrial paint markers that crosslink on contact and laugh off acetone. The line isn't as clean as you'd think.
Right — and before we get into the modern stuff, we've got to understand what "permanent" actually means, because it's not one thing. Are we talking about UV resistance? A mark that survives a thousand years in a dark archive might disintegrate in six months of direct sunlight. A mark that handles industrial solvents might flake off under mechanical wear. Different problems, different chemistries.
Let's trace it from the beginning. What actually worked, what didn't, and where we are now — because the answer to "can ink beat engraving" depends entirely on which ink and which engraving and what kind of forever you're actually asking for.
That's exactly what we need to break apart. When a conservator at the Smithsonian says "permanent," she means something completely different from what a factory floor manager in Cleveland means. They're not even speaking the same language.
The conservator wants the ink to not eat the paper over three centuries. The factory manager wants a label that survives a degreasing bath. Same word, opposite problems.
And that's why you've got to think about permanence on three axes. First, UV resistance — how fast does the mark break down under light? Second, chemical resistance — can it handle solvents, acids, cleaning agents? Third, abrasion resistance — if you rub it, scrape it, drag something across it, does it stay put? A mark can be brilliant on one axis and useless on the other two.
Like a sunscreen that's waterproof but doesn't block UV.
That's actually a perfect analogy. And what's wild is how many products labeled "permanent" optimize for exactly one of those three — usually the one the marketing department thinks sells markers — and ignore the rest.
Before we even ask whether ink can beat engraving, we've got to ask: permanent against what?
This is where the fundamental divide kicks in. Pigment-based marks — ink, paint, toner — they're a coating. They sit on top of the substrate. Their durability depends on how well they adhere, how chemically stable the pigment particles are, and how resistant the binder is to whatever's attacking it. Substrate-modifying marks — etching, engraving, laser ablation — they don't add anything. They remove material. The mark is a physical void, a shape carved into the object itself.
The coating is a guest. The engraving is a scar.
Beautiful way to put it. The guest can be evicted. The scar is part of the body.
Which makes the central question almost unfair on its face. How can a layer of something that's merely sitting there ever outlast a hole in the material?
Yet — here's where it gets interesting — sometimes it can. Ceramic inks fired at eight hundred Celsius vitrify. They literally turn into glass, fused with the substrate at a molecular level. At that point, is it still a coating, or has it become a surface modification? The boundary blurs.
The framing isn't "can ink beat engraving" — it's "under what conditions does a coating become indistinguishable from a structural change." And the answer changes depending on whether you care about UV, chemicals, abrasion, or just the sheer brute passage of millennia.
Which means before we dive into the five-thousand-year history of what humans have actually tried, we need to hold onto that question. Because every technology we're about to talk about — from Egyptian lampblack to laser-induced graphene — is really just a different answer to the same problem: how do you make a mark that outlasts the person who made it? And that brings us to the very first answer.
Twenty-five hundred BCE. Egyptian scribes are writing on papyrus with an ink formula that is, in its essentials, almost absurdly simple. You take lampblack — that's the soot collected from burning oil lamps — you mix it with gum arabic, which is just hardened acacia tree sap, and you add water. That's it.
Which sounds like something I'd cook up in a cave and call "leaf medicine.
Except this one actually worked. Carbon black is elemental carbon in a finely divided form — particles maybe ten to fifty nanometers across. And here's why it's still, after four and a half millennia, the gold standard for lightfastness. Carbon is chemically inert. It doesn't oxidize, it doesn't react with acids or bases under normal conditions, and UV light — which shreds almost every organic pigment — essentially bounces off it. The ASTM D6901 standard for lightfastness? Carbon black inks score a Category One, the highest rating, and they've held that rating since the standard was created.
The Egyptians stumbled onto the one pigment that physics can't really touch.
They didn't stumble — they iterated. There's evidence of earlier ink experiments using charcoal and plant gums going back to thirty-two hundred BCE in China. But the Egyptian formula was the one that got refined and spread across the Mediterranean. And the reason it works so well on papyrus is that the gum arabic acts as a binder — it holds the carbon particles in suspension and then dries into a thin film that lightly adheres to the plant fibers. The carbon sits on the surface, physically trapped. It's not chemically bonded — there's no covalent link between the carbon and the cellulose. It's just...
Which makes it vulnerable to abrasion, I'd imagine. You could flake it off.
Carbon black ink on papyrus is a coating, to use your framing. It's a guest. If you rub it hard enough, it'll come off. But nobody was scrubbing their tax records, and in a dark, dry tomb, the ink just... The Egyptian Museum in Turin has documents where the ink is still legible — not faintly visible, but actually readable — after four thousand years of doing absolutely nothing but sitting there.
The first great permanence technology was essentially: use the most chemically boring substance you can find, and keep it out of the rain.
That worked brilliantly for about two thousand years. Then Europe decided to get clever, and things got both much better and much worse at the same time.
This sounds like the setup for iron gall ink.
Iron gall ink is one of those technologies that's so elegant it's almost tragic. The recipe appears around the fourth century, but it really takes over Europe between the fall of Rome and the Renaissance. You take oak galls — those knobbly growths on oak trees caused by wasp larvae — and you crush them and boil them in water. That extracts tannic acid. Then you add iron sulfate, which was originally made by pouring sulfuric acid over old nails. The iron ions react with the tannic acid to form a dark, intensely black iron-tannate complex. You add a little gum arabic for body, and you've got ink.
Unlike the Egyptian stuff, this one actually bites into the paper.
That's the key. Iron gall ink doesn't just sit on the surface — it chemically bonds with the cellulose fibers. The iron-tannate complex forms coordination bonds with the hydroxyl groups in the cellulose. It literally becomes part of the paper. You cannot wash it off. You can't flake it off without destroying the fiber itself. It's not a guest — it's somewhere between a tenant and a structural modification.
Which is why it dominated for fifteen hundred years. Constitution was written with it. Bach's manuscripts, Da Vinci's notebooks, the Magna Carta — all iron gall.
Yet — and this is where the tragedy comes in — the same chemistry that makes it permanent also makes it destructive. The iron ions continue to catalyze oxidation over centuries. The ink eats through the paper. Constitution has acid burn-through where the iron gall has literally corroded holes in the parchment. Conservators look at iron gall manuscripts and see a ticking clock.
You get a mark that's chemically bonded and physically unremovable, but it's also slowly destroying its host. Like a parasite that needs the host alive.
That's not even an exaggeration. There are medieval manuscripts where the ink has eaten clean through the vellum, leaving the text as a kind of stencil — legible only as holes.
Which raises the obvious question. If they knew — and by the Renaissance, people definitely noticed the degradation — why did it stick around for fifteen centuries?
Because the alternatives were worse. Carbon black ink doesn't bond — you can wash it off parchment with a damp cloth. That's fine for a scroll in a tomb, but if you're a medieval scribe and someone spills wine on your manuscript, you lose everything. Iron gall was the only ink that gave you chemical permanence — the mark survived the apocalypse even if the paper didn't. And for legal documents, land records, religious texts — things where erasure was literally a sin — that mattered more than longevity.
Which brings us to the Archimedes Palimpsest, which is the perfect case study in exactly this tradeoff.
This story is incredible. In the tenth century, a scribe in Constantinople copies the works of Archimedes — including two treatises that exist nowhere else — onto parchment using iron gall ink. Fast forward two hundred years. Parchment is expensive. A monk needs material for a prayer book. So he scrapes the Archimedes text off — physically abrades the surface to remove the ink — and writes over it. The iron gall had bonded so deeply into the collagen fibers that even scraping couldn't remove it entirely. Ghosts of the original text remained.
In 2008, multispectral imaging pulled the whole thing back.
They used ultraviolet, infrared, and X-ray fluorescence to detect the iron atoms still bonded in the parchment fibers. The text was physically gone from the surface, but the chemical signature was still there, embedded at the molecular level. That's how deep iron gall penetrates. It's not a coating you can strip — it's a chemical alteration of the substrate.
Iron gall wins on chemical permanence but fails catastrophically on substrate preservation. The Egyptian ink wins on substrate preservation but fails on mechanical permanence. Neither one solves the whole problem. And then the nineteenth century arrives and makes everything worse.
The aniline dye revolution. An eighteen-year-old chemist named William Henry Perkin is trying to synthesize quinine from coal tar. Instead, he accidentally produces a vivid purple compound — mauveine — and essentially invents the synthetic dye industry. Within thirty years, you've got hundreds of aniline-based colors, and for the first time in history, brightly colored inks are cheap enough for mass production.
Catastrophically non-archival, I assume.
It was discovered in 1893 — there was a study published in the Journal of the Society of Arts — that aniline dyes fade within decades under direct sunlight. Sometimes within years. The dye molecules are small, highly conjugated organic structures that absorb visible light beautifully, which is why they're so vibrant. But those same conjugated systems are vulnerable to photodegradation. UV photons smash the double bonds, the molecule fragments, and the color disappears.
The synthetic dye era gave us color at the cost of permanence. And that's the tradeoff embedded in basically every consumer marker sold today.
It got baked into the modern marker's DNA from the very beginning. Sidney Rosenthal invents the Magic Marker in 1953 — it's a glass bottle of ink with a felt wick. You press the felt tip to the surface and the ink flows through. Then in 1964, the Sharpie arrives — a felt-tip pen with an aluminum barrel, using xylene and glycol ether solvents to carry dye-based ink. The solvents evaporate fast, leaving the dye on the surface. Fast-drying, convenient, cheap.
On paper, it works fine because the paper fibers give the dye something to grab onto. But on plastic?
This is where the chemistry gets specific and unforgiving. Polyethylene — the stuff Ziploc bags are made of — has a surface energy of about thirty-one millinewtons per meter. That's extremely low. It means the surface is non-polar and doesn't want to interact with anything. Most solvents and dyes are moderately polar. They bead up rather than wetting the surface. And even if the solvent does deposit the dye, the dye molecules just sit there — no chemical bond, no mechanical interlock, nothing.
Because polyethylene is smooth at the microscopic level. There are no pores, no fibers, no micro-abrasions for the dye to get caught in.
The dye is a guest on a surface that doesn't want guests. Give it a few weeks, a little humidity, some handling, and it flakes or rubs right off. The Sharpie-on-a-Ziploc problem isn't a Sharpie problem — it's a fundamental mismatch between dye chemistry and substrate physics.
The modern permanent marker is permanent only on porous surfaces where the dye can wick into the fibers and get mechanically trapped. On anything non-porous — plastic, glass, metal — it's basically temporary.
Sharpie has sold two hundred million markers as of 2020, most of them dye-based. That's two hundred million writing instruments whose "permanence" is conditional on what you're writing on, and almost nobody reads the fine print.
We've got four and a half thousand years of ink technology, and the scorecard is: carbon black lasts forever but doesn't stick. Iron gall sticks forever but destroys what it's stuck to. Synthetic dyes look great and last about as long as a mayfly. And the modern marker combines the worst of all worlds — dye-based, no chemical bond
Which brings us to the question of whether a surface coating can ever actually compete with etching for real permanence. And the answer is: sometimes, but only when the coating stops being a coating.
That's the ceramic ink category, right? The stuff that gets fired in a kiln.
Ceramic inks are this fascinating middle ground. You print or paint a glass-based pigment mixture onto a surface, then fire it at around eight hundred degrees Celsius. At that temperature, the glass frit in the ink melts and vitrifies — it fuses into the substrate itself. The mark isn't sitting on the ceramic or the metal anymore. It's become part of the glaze. You can't peel it off because there's no interface to peel.
It's a coating that becomes a structural modification through heat. Like a caterpillar turning into a butterfly, except the butterfly is fused glass and it'll outlive your grandchildren.
This is where the ASTM D6901 revision from last year gets really interesting. They introduced a new Category Five rating for what they're calling "extreme permanence." Only two technologies qualify: ceramic inks fired to vitrification, and laser engraving. Not paint markers, not archival pigment inks, not even iron gall. The bar is essentially: can this mark survive a thousand hours of accelerated weathering and come out unchanged?
Which is an almost absurd standard. A thousand hours in a QUV weathering chamber simulates something like five to ten years of brutal outdoor exposure.
Laser engraving sails through it because there's nothing to degrade. A laser engraver creates a cavity — typically zero point one to zero point five millimeters deep — in the surface of metal or glass. The mark is literally a void. It's the absence of material. UV light has nothing to attack. Solvents have nothing to dissolve. You can sandblast it and the mark is still there because you're just making the void deeper.
Chemical etching works on the same principle, just through a different mechanism. Ferric chloride on copper, for example — it's removing atoms from the surface, creating a recessed pattern that's physically part of the object.
The Voyager Golden Record is the ultimate expression of this. Launched in 1977, it's a nickel-plated copper disc with analog data engraved directly into the metal. No ink, no pigment, no coating of any kind. The information is stored as physical topography. NASA expects it to remain readable for roughly one billion years in interstellar space.
A billion years. That's such an absurd number it almost stops meaning anything. The sun will have gone through about ten orbits of the galaxy by then. The continents on Earth will be unrecognizable. And this engraved copper disc will still play its little greeting from humanity.
There's a tradeoff, and it's the one you've been circling since the beginning. Etching and engraving are destructive by definition. You're removing material. You can't undo it. You can't erase it and start over. If you make a typo on a laser-engraved serial number, that part is scrap.
Versus a paint marker, where you can wipe it off with the right solvent and try again. The reversibility is a feature, not just a limitation.
Industrial paint markers have gotten remarkably good at threading this needle. Take the Markal B — it uses an alkyd resin binder loaded with metallic pigments. When you apply it, the solvents flash off and the resin crosslinks, forming a hard thermoset coating. On rough or oxidized metal, the pigment particles physically lock into micro-abrasions. It's not a chemical bond, but it's a mechanical interlock that's surprisingly tenacious.
It's the same principle as the dye wicking into paper fibers, but engineered for non-porous surfaces.
You're creating the porosity by choosing a pigment particle size that matches the surface roughness. And on the archival side, the pigment-based inks in high-end printers — Epson's UltraChrome K3 is the classic example — take a different approach. They encapsulate pigment particles in resin, so when the ink dries, each particle is sealed in a microscopic polymer shell. The Wilhelm Imaging Research Institute rates these at over two hundred years of dark storage permanence, and eighty-plus years under glass.
Which is impressive, but it's still not a billion years on a copper disc.
No, and it never will be. A coating is always more vulnerable than a void. But there's a new technology that's blurring the line in a way I find exciting. In 2024, a group at Rice University published a technique called laser-induced graphene marking. You take a polyimide surface — Kapton tape is the common substrate — and you run a carbon dioxide laser over it at specific power settings. The laser doesn't ablate the surface. It photothermally converts the polymer into porous graphene.
Wait — it converts the substrate itself into a different material?
Right at the surface layer. The polymer's carbon atoms reorganize into a graphene lattice. The mark is physically part of the substrate — it's not a coating — but it's also not a void. It's a structural transformation. And because it's graphene, it's electrically conductive, chemically stable, and bonded at the molecular level. In 2025, the first commercial application went live for aerospace part tracking, where you need a mark that survives extreme temperatures and solvents but can't compromise the structural integrity of the part by engraving a cavity into it.
It's the best of both worlds. Permanent like an engraving, but non-destructive like a coating.
That's the promise. And it raises an interesting question about whether the whole coating-versus-etching dichotomy is even the right framework anymore. If you can transform the surface without removing material and without adding material, you're in a third category entirely.
Which makes me wonder about the five-hundred-year outdoor exposure question. Is there any pigment-based mark that can survive that?
Not without help. The only pigment-based marks that approach that timescale are ceramic inks on architectural glass and certain enamel formulations on metal, and both require firing to achieve it. An unfired pigment coating — even an archival one — will eventually succumb to UV, thermal cycling, and moisture. The binder degrades. The pigment particles lose adhesion. The coating fails.
The honest answer to the prompt's central question — can pigment-based solutions compete with etching for durability — is: only when they stop being pigment-based solutions and become fused glass or transformed substrate.
That's the through-line of the whole five-thousand-year story. Every time we've tried to make a surface coating truly permanent, we've ended up reinventing it as a structural change. Iron gall bonding to cellulose. Ceramic ink vitrifying into glaze. Laser-induced graphene reorganizing the polymer itself. The quest for the permanent mark keeps leading us away from marking and toward transformation. And that transformation is exactly what we need to keep in mind when we're choosing a tool for a practical job.
If we're going to give people something they can actually use, the first thing is: throw away your Sharpie if you're labeling anything non-porous. Get a paint marker — oil-based enamel, the kind you have to shake until you hear the ball bearing rattle. The pigment particles are orders of magnitude larger than dye molecules, and they physically lock into whatever microscopic roughness exists on the surface.
Which is why the paint marker on a plastic storage bin works fine for years, but the Sharpie label smears off the first time it rubs against something in the cabinet.
For anything you actually care about keeping — documents, photographs, the family recipe cards — the rule is brutally simple. Pigment-based ink on acid-free paper. Carbon black if you can get it. Avoid dye-based inks entirely. They'll look great for five years and then they'll feather into illegibility because the dye molecules are a thousand times smaller than pigment particles and they diffuse right into the paper fibers.
The acid-free paper part matters as much as the ink. You can use the best archival ink in the world, but if the paper has lignin in it, it's going to yellow and embrittle, and your writing goes with it.
Then there's the third tier — the one that applies when you need immortality. If something has to survive a hundred years outdoors, or a thousand, the only answer is etching or laser engraving. No coating, no matter how well-formulated, can match the durability of a physical void in the substrate. You're not protecting the mark from the elements. The mark is the absence of material, and the elements can't destroy what isn't there.
The decision tree is: indoors on paper, pigment ink. Indoors on plastic or metal, paint marker. Outdoors or multi-generational, engraving.
There are quick ways to verify what you're buying. For art supplies, check the ASTM lightfastness rating — Category One is what you want for anything you expect to last. For archival markers, look for ISO 11798 compliance on the label. And if you want to test permanence yourself, the rubbing alcohol wipe test takes five seconds. A truly permanent mark won't budge. A dye-based one will come right off on the cotton swab.
Which is a humbling moment the first time you do it on something you thought was permanent.
That humbling feeling brings me to a question that's been rattling around in my head since we started this. There's some wild research coming out of MIT right now on bio-inspired adhesives — specifically mussel-foot proteins. Mussels secrete these proteins that let them stick to anything. Rocks, metal, Teflon, you name it. The proteins contain a particular amino acid called DOPA that forms covalent bonds with whatever surface it touches.
Instead of hoping pigment particles find a rough spot to grab onto, you engineer the binder to chemically bond like a mussel gluing itself to a pier.
That's the idea. And the 2025 study showed something remarkable — they achieved three times better adhesion on PTFE, which is Teflon, than any existing coating. PTFE has surface energy so low that nothing sticks to it. That's why it's on non-stick pans. And these mussel-inspired adhesives just...
The question is whether you could load those proteins with pigment and create an ink that covalently bonds to literally any surface. That would be a genuine competitor to engraving, because you'd have a coating that's chemically inseparable from the substrate.
It's early-stage, but it's the first approach I've seen that might actually close the gap between coating and etching without requiring an eight-hundred-degree kiln.
The other thing we haven't talked about is where industry is pushing this. Supply chains are moving toward mandatory permanent tracking — ISO 17367 for RFID tagging, for example. You need a mark on an aerospace bracket or a medical implant that survives sterilization, solvents, and decades of use, and that mark has to be machine-readable.
That's where laser-induced graphene becomes the obvious answer. It's a mark that is the part. You're not adding a tag that can fall off. You're not engraving a stress riser that could cause a fatigue crack. You're transforming the surface into something that's both permanent and functional — it's conductive, so it can double as a sensor or an antenna.
The line between a mark and a circuit starts to disappear.
And I think in ten years we'll look back at the coating-versus-etching debate the way we now look at the debate about whether telephones need wires. The answer was always going to be: neither. Something else entirely.
That brings me to a final thought about the Voyager record. It's the most permanent mark humans have ever made — a billion years in deep space. But it's also the least accessible. If nobody ever finds it, if no civilization ever plays it, then what is it? Permanence without readability is just damage. A scratch on a copper disc floating through nothing.
The mark only matters if there's someone to read it.
Which is why the five-thousand-year story of permanent marking isn't really about chemistry or engineering. It's about the hope that someone, someday, will see what we left behind.
Now: Hilbert's daily fun fact.
Hilbert: In the 1860s, prospectors in the Yukon observed a rare halo phenomenon where ice crystals in high cirrus clouds refracted moonlight into a perfectly circular ring with two bright spots on either side. They called it a "moon dog," but the scientific name — parhelion — comes from the Greek "para helios," meaning "beside the sun." The name stuck even for the lunar version, which technically should be called a paraselene. Nobody bothered to fix it.
We've been calling it the wrong thing for a hundred and sixty years and just...
That feels about right for humanity. This has been My Weird Prompts. If you enjoyed this episode, leave us a review wherever you listen — it helps more people find the show. For our producer Hilbert Flumingtop, I'm Herman Poppleberry.
I'm Corn. We'll be back next week.
