Do Tattoo Ink Ingredients Interact? Pigments, Metals and Particle Size Explained
A tattoo ink can change when its ingredients are mixed, exposed to light or placed in tissue. The pigment, carrier, manufacturing residues and particle surfaces all matter. The needles can add metal debris that was never in the bottle.
Particle size gets most of the attention. It deserves some. But "nano," "submicron" and "micro" describe dimensions, not safety. Chemical identity, crystal form, surface coating, clustering and dose can all matter more.
One caveat before any of it: everything below assumes the label is accurate. When researchers test inks against their own paperwork, it often isn't. That's covered separately in why you can't just read the label, and in why tattoo ink needs better testing if you want the view from the bench.
How to read the evidence
I distinguish results measured in tattoo research from mechanisms demonstrated elsewhere and ideas that haven't been tested.
When a finding comes from tattoo research, I identify the setting: an ink sample, cells, reconstructed skin, an animal or human tissue. Evidence from another use of the same material can suggest what to investigate, but it doesn't establish what happens in a tattoo. Proposed interactions remain unproven until the relevant steps are measured.
A chemically possible reaction still needs an exposure measurement and evidence of a biological effect before it can explain harm in people.
Behind any ingredient claim sit six questions. Is it present? Can it transform? Does that happen in skin? At what concentration? Does it cause a measurable effect? How often does that become a clinical problem? One study rarely answers more than two.
Nano, submicron and micro
A nanoparticle is roughly 1 to 100 nanometers. Above 100 and below 1,000 nm is submicron. At least 1 micrometer is micro. A micrometer is 1,000 nanometers. These are working bins, not thresholds where a material turns dangerous.
In a study of 58 inks, 99.94% of ink volume was particles under 100 nm. Blacks were smallest, whites largest, colors in between. That describes those products, not every ink on a shelf.
The particle and the cluster are different measurements
A pigment can have primary particles tens of nanometers across while its suspension holds clusters a thousand times bigger. An aggregate is strongly bonded; an agglomerate is loosely stuck together. Neither tells you how it behaves in tissue.
Instruments disagree too. Electron microscopy sees physical structure. Dynamic light scattering estimates size by movement in fluid, where big clusters dominate the signal. Single-particle elemental analysis finds element-bearing populations and estimates diameter from assumptions. Read them together.
For equal-density spheres, surface area per unit mass scales inversely with diameter, so a 30 nm powder has about ten times the geometric surface of a 300 nm powder. That's geometry, not a tenfold toxicity prediction. Coatings, clustering and buried internal surface change what's actually exposed.
Whether the smallest fraction can reach places other particles can't is a separate question, and one I look at in tattoo ink nanoparticles and the blood-brain barrier.
Titanium dioxide
TiO2 is the white pigment, and it's what lightens most other colors. Its main crystal forms are rutile and anatase, and they don't behave the same way despite sharing a formula.
Under UVA
Reactive oxygen species (ROS) are reactive oxygen-containing molecules. In a 2024 study using OECD Test Guideline 432, anatase was the only pigment tested that showed phototoxicity in the fibroblast assay. Crystal form decided it, not size. The suspensions were also heavily agglomerated, which muddies any size comparison.
Setting changes the answer. A separate study ran flat cell cultures alongside reconstructed tattooed skin. Phototoxicity showed up in the flat cultures, especially with anatase. The full-thickness model didn't reproduce it, and some pigments actually reduced UV-associated DNA damage markers. Tissue structure and light shielding matter. That is not a reason to treat a tattoo as sunscreen.
Unproven: that every TiO2 tattoo generates a damaging ROS dose in ordinary sun.
There's more on this pigment specifically, including what the FDA's own permanent-makeup testing found, in white tattoo ink and titanium dioxide.
Under a laser, mixed with yellow
A 2025 study fired a Q-switched Nd:YAG at 532 nm into Pigment Yellow 14, 65 and 74 and into commercial inks. TiO2 changed the degradation: different particle shape, different size, different volatile products, some of them harmful.
That's real evidence of a mixture effect under a laser. It says nothing about mixing pigments at body temperature, and it doesn't mean TiO2 always increases toxicity. Scattering, absorption, local heating and surface chemistry all need separating before anyone calls it photocatalytic synergy.
Against needle steel
A 2019 study found steel debris in human tattoo tissue and in the matching lymph nodes. In pig skin, TiO2 ink wore needles visibly harder than carbon black did. Tattoo needles run 6 to 8% nickel and 15 to 20% chromium.
For a long session with heavy white or pastel packing, pay attention to needle condition. The abrasion finding concerns the pigment's interaction with the steel.
Carbon black
Carbon black is a particulate carbon pigment, not interchangeable with every soot, graphite or engineered carbon nanomaterial. For the pigment families behind every black on the market, see black tattoo ink explained.
A study of commercial blacks found ROS potential varied by brand and by how aggregated the particles were. It did not track PAH content. So you can't pin the oxidative activity on one contaminant measurement, and you also can't credit aggregation with the whole effect.
Adsorption means molecules stick to a surface. In the 2024 phototoxicity work, free benzo[a]pyrene was phototoxic and the carbon-black-bound version wasn't. Surfaces change exposure. That isn't permanent detoxification and it isn't a human measurement.
Polycyclic aromatic hydrocarbons (PAHs) are manufacturing contaminants, not the pigment. An analysis of 19 inks found 0.14 to 201 micrograms per gram, with singlet-oxygen quantum yields of 0.18 to 0.85. A real hazard mechanism, but not a disease probability for your tattoo.
Iron oxides
"Iron oxide" covers hematite, magnetite and maghemite, which differ in chemistry, structure and surface behavior. An elemental iron reading can't tell you which one you have, or separate pigment iron from needle debris.
A 1993 report documented five cosmetic tattoos that darkened immediately and irreversibly under short-pulsed lasers. Ferric oxide in agar blackened the same way under Q-switched ruby. Reduction of ferric oxide was the proposed explanation, though the study didn't nail down every step.
Outside tattooing, 7 nm maghemite crystals were tracked in a medium mimicking the inside of a cell. Three different surface coatings dissolved at three different rates, while the undissolved crystals kept their original size and structure. Coating governs the rate, not just diameter.
Free ferrous iron can drive Fenton chemistry, forming hydroxyl radicals from hydrogen peroxide. Whether a tattoo supplies enough accessible iron and peroxide to damage tissue is a measurement nobody has made.
TiO2-derived oxidants and available iron could, in principle, feed oxidative reactions. Nobody has shown that combination damaging tissue in a human tattoo. The relevant chemical forms, accessible amounts and conditions still need to be measured.
Azo pigments
Azo covers a lot of reds, yellows and oranges with different structures, so the label alone can't tell you what a given pigment breaks into. More on this family in azo pigments, sunlight and lasers.
UVB and natural sunlight cleave a tattoo pigment in solvent. In mice, PR22 tattoos lost about 60% of their pigment over 32 days under a solar simulator, and about 51% to a laser 42 days in. Photochemical change happens. Those numbers are not a human rate.
Laser work identified hazardous breakdown products, and tested one of them, 3,3'-dichlorobenzidine, separately in human skin cells, where it damaged DNA. Together those support real concern about that compound. Neither shows cancer from laser removal in patients.
Red pigments and allergy
In 104 biopsies from chronic reactions in red tattoos, the pigments found were mainly PR22 at 35%, PR210 at 24% and PR170 at 12%. In 22% of biopsies, no red pigment turned up at all. Metals were measured, and their role couldn't be established.
That's useful clinical chemistry. It isn't a rate, because there's no denominator: nobody counted everyone who got those pigments and didn't react.
The authors proposed pigment fragments acting as haptens, small molecules that form immune-visible complexes with tissue proteins. They couldn't identify the culprit epitope. The pathway stays a hypothesis while its middle steps are unmeasured.
For artists: reds carry the strongest clinical signal in the literature. Document product and batch on red-heavy work. Don't tell a client a pigment will cause a reaction.
Quinacridones
PR122, PV19 and PR202 are structurally unlike azo pigments, and PR202 doesn't inherit findings about the other two.
A 2015 study found seven reactions across six patients with red tattoos. Five involved azo pigments, two involved quinacridones, both times PV19 and PR122. An association with those formulations, not proof the intact molecule was the allergen rather than a coating or contaminant.
A newer lead points at manufacturing residue. FDA researchers found rosin oxidation products in 32 pigments and 38 inks, averaging about three times higher in reds. Among 24 red inks containing them, PR170 appeared in 21. The authors proposed these residues act as haptens and said plainly the proposal awaits proof.
The study measured contaminants. Their proposed role in allergy still needs proving.
Metals: the element is not the exposure
This is the single idea that governs every metal below. A lab can tell you an ink contains 40 micrograms of nickel per gram. That number says nothing about how much nickel your skin can actually reach, because it depends on what chemical form the nickel is in and whether anything can dissolve it.
A 2026 study measured what actually comes out of the bottle. Nickel reached 63% extractable in saline and 59% in artificial sweat for black inks. Lead, in the same inks, exceeded regulatory limits on total content yet stayed at or below 2% available. Same bottle, opposite answers, depending on the element.
A 2020 particle analysis found that the copper and zinc in the inks tested were mostly ionic, not particles at all. Titanium in the same study measured 166 to 383 nm.
And 29 inks analysed for chromium speciation showed more than 1 mg/kg of Cr(VI) in 27.6% of samples, far above the 0.2 mg/kg limit in force at the time. A total chromium number would have hidden that completely, because chromium metal, Cr2O3 and Cr(VI) are three different exposures.
Why it matters biologically: nickel and cobalt ions activate human TLR4, an innate immune receptor, without the co-receptor bacterial endotoxin needs. And cells from chromium-allergic patients mount a mixed Th1/Th2 cytokine response on exposure. Both are ion effects. Neither tells you what any particular particle will do.
That changes how I read metal claims.
A blue ink commonly uses copper phthalocyanine, so its color doesn't identify cobalt. Alumina, metallic aluminum and soluble aluminum are different exposures, just as poorly soluble barium sulfate differs from soluble barium. A pigment coating also can't be equated with a vaccine adjuvant.
Nickel and cobalt each having an immune effect doesn't establish that their combination produces an effect beyond their individual contributions at tattoo doses. That requires a mixture experiment.
A 2025 study tested reference nanoparticles and three commercial inks in cells and in human skin explants tattooed with actual needles. CuO was toxic dose-dependently and high ZnO affected cells; Al2O3, Cr2O3, Fe2O3 and TiO2 weren't, under those conditions. The nominal nanoparticles frequently measured well over 100 nm once suspended. Reference compounds aren't the species in a real ink, and a short-term explant can't model a lifetime.
The liquid is part of the formulation
Water, glycerol, alcohols, glycols and dispersants keep pigment suspended and make ink flow. They may also influence wetting, clustering and how much dissolved material is available. Those are fair questions, unanswered by controlled tattoo studies.
Dissolved glycerol and glycols don't fit pigment-size categories. A dissolved polymer isn't automatically a nanoparticle either. The persistence of each component needs measuring; the tattoo lasting doesn't mean every ingredient stays with it.
What happens once it's in
Proteins coat the surface
In experiments on silica and polystyrene, not tattoo pigments, plasma proteins formed surface coronas within minutes, changing cell uptake and causing endothelial cell death early on. Good reason to investigate pigment surfaces. It doesn't identify what coats them in human skin.
Protein adsorption also isn't hapten formation. A protein stuck to a particle doesn't become an allergen.
Macrophages hold, release and recapture
In tattooed mice, dermal macrophages captured pigment, released it when those cells were killed, and replacement cells picked it back up. Graft tattooed skin onto another mouse and six weeks later most pigment-carrying macrophages came from the recipient, with no visible change to the tattoo. Later work looked at fibroblasts.
That explains persistence. It doesn't rank particle sizes for human retention, and uptake is not the same endpoint as inflammation, cell death or allergy.
Lymph nodes get more than nanoparticles
In human donor tissue, skin held particles up to several micrometres while the matching lymph nodes held only particles under about 200 nm. Micrometre-scale steel debris turned up in nodes in the later needle study.
So pigment and wear material leave the tattoo, and the smaller material preferentially reaches the nodes. That doesn't trace every particle's route, prove disease from deposition, or tell you what fraction moves. I go through that evidence in detail in tattoo pigment migration to lymph nodes.
An observation from my own work.
This is a personal account. It is not a case report, it was not clinically investigated, and nothing in it should be read as a research finding.
I tattooed a large back piece on my wife using a blue concentrate whose manufacturer lists its pigment as CI 74160 / CAS 147-14-8, copper phthalocyanine. After the session covering the left wing, a swelling the size of a goose egg came up in her left armpit, fast enough that we stopped and split the piece into two sessions. It resolved on its own in about a week.
Roughly 45 days later I tattooed the right wing. A swelling came up in her right armpit.
The structure is what makes it worth writing down. The first episode had fully resolved before the second session started, so the second wasn't a continuation. The side changed and the response followed. The drainage basin for that part of the upper back is the armpit on that side. My wife is a forensic pathologist, so the assessment wasn't a layman's.
What it doesn't establish. No imaging, no biopsy, nothing sampled or confirmed, so it can't identify a cause. Node reactions are reported with black ink and other pigments too, so blue isn't implicated by this alone. One person isn't a rate. The pigment dose was never quantified. And a label isn't proof of contents: for this product the manufacturer's own ingredient listing and a major retailer's don't match.
What it illustrates. Copper phthalocyanine is one of the pigment classes found in human lymph node tissue in the donor study above. The route is documented. What I watched fit it, in hours rather than years, and repeated on the opposite side under the same conditions.
I include it because transparency cuts both ways. Leaving it out for being anecdotal would be as selective as presenting it as proof.
Sunlight and lasers are different experiments
Sunlight is long and low-intensity. A removal laser is brief pulses whose effect depends on wavelength and how the pigment absorbs. Stability in one says nothing about the other.
Lasers can both fragment particles and cleave molecules. Fragmentation changes size; cleavage changes chemical identity. A study showing one isn't evidence for the other without chemical analysis.
Does mixing two colors create something harmful?
There's no validated chart of hazardous pigment pairs. Interactions can run through several routes: a pigment changing needle abrasion, particles clustering or adsorbing a molecule, light or laser changing the products, a particle supplying ions, or cells meeting particles, ions and injury together.
Only some routes have been measured for specific combinations. The list is a research agenda, not a catalogue of established harm.
Two things get overstated here. Materials scientists build TiO2 composites and dope its lattice with metals to change catalytic behavior; mixing pigments in an ink cup doesn't create that structure. And galvanic corrosion needs electrically connected phases in an electrolyte, which scattered coated particles are not.
"Synergy" also means something specific. A mixture producing an effect isn't enough. The effect has to exceed what you'd predict from the individual effects combined. Two ingredients sitting next to each other in a biopsy prove nothing.
Which ingredients most often cause reactions?
Different reactions point to different parts of the formulation.
Chronic allergic reactions are reported most often in red shades, with certain azo inks and some quinacridones implicated. The clinical papers concern selected cases, so they don't rank the risk for everyone receiving those pigments.
For metal sensitization, the available nickel, cobalt and chromium species matter, along with the person's sensitization history. In light-related lab tests, crystal form, exposed surface and associated contaminants can change the result.
Infection: a separate problem entirely. A 2024 study tested 75 tattoo and permanent-makeup inks by FDA methods and found aerobic and anaerobic bacteria in unopened, sealed bottles, some labeled sterile. Pigment chemistry is no substitute for microbiological quality.
Foreign-body reactions and irritation aren't allergies. Diagnosing the reaction is part of finding the cause.
What this means for artists
For me, the practical value is knowing what to record and what to ask the supplier.
Log the product, color, pigment index number and batch used in each session. That is especially useful with red-heavy work and white packing. In the 104-biopsy study discussed above, 22% had no detectable red pigment, illustrating how hard it can be to reconstruct what was implanted.
Ask suppliers for pigment identity and testing. Watch needle condition with whites and pastels, and treat microbiological quality as a separate issue from pigment chemistry. A sterile label doesn't replace proper handling, storage or single-use practice.
If a client reacts, provide the ink records to the treating clinician. Color alone can't diagnose the cause. Those records are useful for removal too, especially if the tattoo has already reacted.
What this means for clients
What's useful: the actual pigment identifiers, product and batch, manufacturing quality and any analytical testing. Marketing claims answer none of that.
The FDA has approved no color additives for injection into skin. European rules restrict many hazardous ingredients and require labeling, but compliance doesn't settle the long-term particle questions.
If you react, keep the ink and batch details and get a medical assessment rather than guessing the ingredient from the color. Persistent swelling, itching, raised areas or rash have several possible causes. Spreading redness, fever, severe pain or drainage needs prompt attention, and breathing difficulty is an emergency. Tell the clinician and the artist about any metal allergy. A negative patch test doesn't reliably rule out a delayed tattoo reaction.
Nothing here is a home method for testing pigment or removing a tattoo.
What's still missing
We need measurements of how real formulations change after years in tissue, which fragments and impurities drive specific immune reactions, how particle surfaces evolve, and how much material moves after UV or removal. Whether any of it reaches the level of gene regulation is a question people ask constantly and the evidence does not yet answer; I lay out what has and hasn't been shown in tattoo ink and epigenetics.
Most of all, for the size question: controlled studies using the same pigment at different sizes, with purity, phase, coating and dose held constant. Comparing unrelated "nano" and "micro" products can't isolate size, which is why the nano-versus-micro argument stays unsettled.
These interactions deserve study because some are already demonstrated. The useful approach is narrow: identify the material, identify the conditions, measure the interaction, then work out what it means biologically. That serves artists and clients better than treating every plausible reaction as a proven one.
Sources
Hogsberg et al. (2013), ROS production by black tattoo inks, Experimental Dermatology 22:464-469
Regensburger et al. (2010), PAHs in tattoo inks and singlet oxygen, Experimental Dermatology 19(8)
Engel et al. (2007), photochemical cleavage of a tattoo pigment by UVB or sunlight, JDDG 5:583-589
Raghavan et al. (2012), nickel and cobalt activate TLR4, EMBO Reports 13:1109-1115
Bocca et al. (2018), hexavalent chromium in tattoo inks, Contact Dermatitis 79:218-225
Tenzer et al. (2013), rapid formation of plasma protein corona, Nature Nanotechnology 8:772-781
Strandt et al. (2021), macrophages and fibroblasts in tattoo stability, Dermatology 237:296-302
Wang et al. (2021), analytical survey of tattoo inks, Contact Dermatitis 85:340-353
FDA: Tattoos & Permanent Makeup Fact Sheet
European Chemicals Agency: Tattoo inks and permanent make-up
This is an evidence overview, not medical advice. If you have a reaction or a health concern related to a tattoo, have it assessed by a qualified clinician.
For tattoo work in Arlington, Texas, visit the booking page.
