UV-curing adhesives, coatings and potting compounds

UV-curing adhesives, coatings and potting compounds cross-link within seconds as soon as a photoinitiator absorbs radiation in the matching wavelength range. Whether a formulation cures all the way through depends on four quantities: the wavelength, the irradiance arriving at the part, the dose and the reaction mechanism – free-radical for acrylates, cationic for epoxies. Wavelength and irradiance have to be specified and measured separately; a dose figure on its own does not describe the process.

Photopolymerization under UV light converts liquid acrylate or epoxy formulations into cross-linked solids within seconds and has become an established alternative to thermal or two-component curing in electronics, optics and medical device manufacturing. The process is governed by a small number of physico-chemical quantities: wavelength, irradiance, dose and the reaction kinetics of the binder system in use.

Free-radical curing acrylates and cationically curing epoxies differ fundamentally in oxygen sensitivity, shrinkage and post-cure behaviour, and those differences decide which material suits a given application. At the same time, the move from mercury vapour lamps to UV LED systems changes the demands on formulation and process control, while regulatory requirements on mercury and on critical photoinitiators add further constraints.

This technical report sets out the technological principles, the key process quantities and the market situation for UV adhesives, UV coatings and UV potting compounds, and compiles a referenced selection of commercially available products. The curing reaction itself is covered in detail under UV curing and photopolymerization, process control in the bond line under UV bonding, potting and encapsulation.

UV curing rests on the conversion of absorbed light energy into chemical energy by a photoinitiator. The initiator generates reactive species – free radicals or cations – and thereby starts a chain reaction that turns liquid monomers and oligomers into a cross-linked polymer network.

In free-radical photopolymerization, type I initiators such as phosphine oxides (TPO) undergo unimolecular cleavage on UV absorption into starter radicals, which add to the double bonds of acrylated monomers; the reaction proceeds through the four elementary steps of initiation, chain propagation, chain transfer and termination.

In cationic photopolymerization, onium salts – usually triarylsulfonium hexafluoroantimonate – generate a strong acid under irradiation, which protonates and opens the epoxy rings.

The two mechanisms differ fundamentally in their sensitivity to ambient conditions: free-radical systems are inhibited at the surface by atmospheric oxygen, cationic systems by humidity. What matters for process design is that the cross-link density achieved is not determined by the total radiant energy alone, but by the interplay of wavelength, irradiance, time and reaction mechanism.

Two levels of distinction matter when choosing a process: the chemistry of the binder system (free-radical versus cationic) and the radiation source (mercury vapour lamp versus UV LED), plus the decision between single-cure and dual-cure systems for geometrically demanding parts.

UV curing is described by a limited set of physical quantities, each of which has to be specified and measured separately:

• Wavelength (nm): determines whether the emission spectrum of the lamp overlaps the absorption band of the photoinitiator; without that overlap there is no initiation, however much power is applied.
• Irradiance E (mW/cm² or W/cm²): radiant power per unit area arriving at the part – not to be confused with the electrical power input of the lamp.
• Dose or radiant exposure H (mJ/cm² or J/cm²): the time integral of irradiance, H = E · t for constant irradiance.
• Photon energy (eV or J): follows directly from the wavelength via E = h·c/λ; a 365 nm photon carries 3.40 eV.
• Penetration depth and critical energy (Lambert-Beer): describe how deep a given irradiance reaches into a layer and the surface energy at which polymerization starts.
• Layer thickness and filler content: together with the absorption of the photoinitiator, they determine how far the cure progresses into the depth.
• Temperature: affects molecular mobility, final conversion and the validity of dose-time equivalences.

Which sensor measures which of these quantities reliably is covered under selecting suitable UV sensors; why two instruments can show different readings at the same source is explained under spectral mismatch.

A handful of assumptions come up again and again in process design and can lead to incomplete cure or to customer complaints:

• „The dose in mJ/cm² fully describes the process“: neither the irradiance profile nor the time can be reconstructed from a total energy figure, and different profiles can produce different material properties at the same dose. For free-radical photopolymerization the Bunsen-Roscoe reciprocity law does not hold over any relevant range of conditions either, because the polymerization rate scales with the square root of irradiance rather than linearly with dose.
• „Higher irradiance compensates for a wavelength mismatch“: if the lamp emission does not hit the absorption band of the photoinitiator, the photons pass through the formulation without reacting – regardless of the power applied.
• „Oxygen inhibition affects the whole volume“: dissolved oxygen in the bulk is consumed quickly under irradiation; the real problem is the continuously re-diffusing oxygen layer at the surface, which leaves tacky, uncured surfaces especially at low irradiance.
• „Electrical lamp power equals the optical irradiance at the part“: figures in W/cm describe the electrical power supplied, not the UV radiation arriving at the component; at the same electrical power, different system designs can deliver very different irradiances.
• „An acrylate formulated for 365 nm cures tack-free with any 365 nm LED“: for a tack-free surface, several common acrylate adhesive systems need additional radiation in the 220–260 nm range to overcome oxygen inhibition – a requirement that LED systems emitting only at 365 nm cannot meet, and one that has to be taken into account when selecting the formulation.
• „A line of sight between lamp and bond line is optional“: in areas shadowed by UV-opaque parts (metals, ceramics, filled plastics) a pure UV material stays liquid, even when directly illuminated areas next to it are fully cured.

Source ageing adds to this: irradiance and spectrum shift over operating time, so a released process window has to be re-measured regularly – see ageing of UV lamps and UV LEDs.

Filled or pigmented formulations behave fundamentally differently from clear systems. Opaque pigments such as carbon black or titanium dioxide absorb part of the UV radiation the photoinitiator needs before it reaches deeper layers; inorganic fillers such as glass beads or silica additionally scatter the light and reduce the intensity with depth exponentially.

The result is a cure gradient in which material near the surface cures completely while the core stays liquid. The Lambert-Beer law describes this quantitatively through the penetration depth D_p, at which the intensity has fallen to 1/e of its value, and the critical energy E_c needed as a minimum to initiate polymerization.

For pigmented systems it is worth moving to longer wavelengths (385 or 405 nm instead of 365 nm), because many pigments scatter and absorb these less strongly. Simply raising the dose helps only up to a point: it can over-cross-link and yellow the near-surface region without improving the cure in the core. Layer thickness also affects the shrinkage stress that builds up during cure: typical values are 1–2 % linear and 2–5 % volumetric shrinkage, which in thicker layers acts three-dimensionally and can impose mechanical stress on adjacent components when the cure is faster than the stresses can relax.

For geometrically demanding parts with undercuts or UV-opaque materials (metals, ceramics, filled plastics), dual-cure systems have become established. They combine a UV primary cure with a secondary mechanism: UV plus moisture cure for shadowed areas without additional heat input, UV plus thermal post-cure for highly loaded bonds, and UV plus anaerobic cure for inactive substrates.

Cationic epoxy systems offer an alternative rooted in the chemistry itself: after the exposure ends, their onium salt photoinitiators continue an acid-catalysed dark reaction and can therefore reach shadowed areas without a second curing mechanism – at the cost of lower initial strength and higher moisture sensitivity than acrylates. For applications with no line of sight at all between light source and bond line, research is also looking experimentally at non-optical excitation such as X-rays, which does not require a direct line of sight; that approach is at the research stage and not in broad industrial use.

• Wavelength shift from 365 nm towards 385, 395 and 405 nm: UV LEDs reach a higher wall-plug efficiency at longer wavelengths, but this calls for reformulated photoinitiator systems matched to the wavelength, because classic initiators developed for mercury lamps can lose efficiency at 405 nm.
• More dual-cure formulations: new product lines combine UV with moisture cure in order to reach shadowed areas without additional heat input; the technical consequence is an additional, time-separated post-cure step in production planning.
• Inline dose control becoming standard in regulated industries: closed control loops are increasingly replacing periodic spot checks, with the consequence that process data are available continuously for batch traceability.
• Photoinitiators without a critical classification: in response to the REACH classification of TPO, alternative aliphatic multi-functional initiators with lower cytotoxicity are being developed, which means existing formulations have to be re-assessed for cure behaviour and absorption whenever a raw material changes.
• Miniaturization in electronics and optics: smaller part geometries with tighter cavities increase the need for dual-cure systems and precise dose control, because flood curing reaches its geometric limits.

Selection starts with the question of which wavelength actually reaches the adhesive in the bond line – determined by substrate transmission, shadowed areas and the spectral match between lamp and photoinitiator – and derives from that whether a single-cure acrylate is sufficient or whether a dual-cure or cationic system is needed for unexposed zones. Only once application, technical problem, measurand and process quantity and the suitable technology have been clarified does a concrete product selection become meaningful.

What spectrum the existing source really delivers only shows in a measurement at the part; measured spectra of common lamp types are available for comparison in the spectral database for UV lamps.

The following selection is a referenced compilation of commercially available products based on publicly accessible manufacturer data sheets. It does not claim to be complete; the figures for wavelength and chemical base are taken from the respective technical data sheets and should always be verified against the current data sheet before a process is designed. Each product name links to the manufacturer document evaluated – technical data sheet, product data sheet or product page. Opsytec does not sell adhesives, coatings or potting compounds; this overview serves technical orientation only.

What is the difference between irradiance and dose?
Irradiance (mW/cm²) describes the radiant power per unit area arriving at the part at a given moment, dose (mJ/cm²) the time integral of that quantity over the exposure. Neither the irradiance profile nor the exposure time can be reconstructed from a dose reading alone, which is why both quantities should be recorded separately for process documentation.

Why does higher irradiance not compensate for the wrong wavelength range?
Initiation requires the lamp emission to overlap the absorption band of the photoinitiator. Without a spectral match the photons remain unused by the reaction, whatever power is applied – simply increasing the power produces no additional conversion.

Why do UV adhesives often stay liquid in shadowed areas?
Free-radical curing systems need a direct line of sight to the light source, because activating the photoinitiator depends on photons that actually arrive. Areas behind UV-opaque parts, behind metals or inside undercuts do not receive sufficient irradiation and do not cure, unless a second curing mechanism such as moisture or thermal post-cure is provided.

What distinguishes free-radical from cationic UV curing?
Free-radical acrylate systems cure within seconds but are sensitive to oxygen at the surface and show a linear shrinkage of 1–2 %. Cationic epoxy systems are insensitive to oxygen, continue to cure in the dark after the exposure ends and shrink less, but they react sensitively to humidity and build up initial strength more slowly.

Will UV LEDs replace the mercury vapour lamp entirely?
Not in the short term: in 2025 mercury lamps still accounted for a good half of the revenue in the market for UV curing systems, while UV LED solutions are growing considerably faster. The RoHS exemption for mercury lamps is scheduled for review in 2027, and that review does not in itself imply a ban – the state of the procedure is set out under RoHS and UV lamps.

Which standards are relevant for UV potting compounds?
For reactive resin compounds and potting systems the IEC 60455 family applies, in particular IEC 60455-2:2023 for test methods such as degree of cure and mechanical properties. In medical technology, ISO 10993 for biological evaluation applies in addition, together with industry-specific approval standards from automotive and aerospace, depending on where the potted component is used.

Why does the reciprocity law not hold without restriction for UV adhesives?
The Bunsen-Roscoe law assumes a linear dependence of the reaction rate on irradiance. Free-radical polymerization instead follows a square-root dependence because of bimolecular radical termination, so identical calculated doses can produce different conversions and material properties depending on the combination of time and irradiance – which makes empirical process validation necessary.

What role does oxygen play in UV curing?
Oxygen quenches radicals into unreactive peroxy radicals and therefore acts mainly as a surface effect: dissolved oxygen in the bulk is consumed quickly under irradiation, while oxygen continuously re-diffusing at the interface with air delays tack-free cure. Cationic systems are not affected by this, but they are sensitive to moisture.

Which irradiance does a UV adhesive need?
There is no blanket answer – the adhesive's data sheet and the absorption range of its photoinitiator are what count. Typical figures range from a few tens to a few hundred mW/cm² in the matching wavelength range, together with a minimum dose.

• ISO 10993-1:2018/2025 – biological evaluation of medical devices, part 1; the revised edition EN ISO 10993-1:2026 has been available since 15 January 2026.
• IEC 60455-2:2023 – test methods for reactive resin compounds (potting compounds), 4th edition 2023.
• ASTM D1002 / D3163-01(2023) – lap shear strength of metal and of rigid plastic adhesive joints.
• UV-Curable Adhesives Market – Mordor Intelligence, data as of 2026; market volume and segment shares.
• UV-Curable Coatings Market Report – Grand View Research, 2025; global market figures for UV coatings.
• RoHS exemption 4(f)-IV – Clarifications on 2027 Expiration, UV+EB Technology, 2026; regulatory status of mercury lamps.

An overview of the regulations relevant to UV applications is given under guidelines, norms and standards in UV.

The following scientific publications are the primary basis for the statements made in this report on reaction kinetics, the reciprocity law and measurement methodology. For full manufacturer data sheets and standards documents, please refer to the sources cited above.

• Wydra JW, Cramer NB, Stansbury JW, Bowman CN: The reciprocity law concerning light dose-relationships applied to BisGMA/TEGDMA photopolymers. Dental Materials 30(6):605–612, 2014.
• Palagummi SV, Hong T, Wang Z, Moon CK, Chiang MYM: Resin Viscosity Determines the Condition for a Valid Exposure Reciprocity Law in Dental Composites. Dental Materials, 2019/2020.
• Sangermano M, Roppolo I, Chiappone A: New Horizons in Cationic Photopolymerization. Polymers 10(2):136, 2018.
• Eibel A, Fast DE, Gescheidt G: Choosing the ideal photoinitiator for free radical photopolymerizations. Polymer Chemistry 9:5107–5115, 2018.
• Lang M, Hirner S, Wiesbrock F, Fuchs P: A Review on Modeling Cure Kinetics and Mechanisms of Photopolymerization. Polymers 14(10):2074, 2022.
• Bennett J: Measuring UV Curing Parameters of Commercial Photopolymers used in Additive Manufacturing. Additive Manufacturing 18:203–212, 2017.
• Stadler E, Eibel A, Fast D, Freißmuth H, Holly C, Wiech M, Moszner N, Gescheidt G: A versatile method for the determination of photochemical quantum yields via online UV-Vis spectroscopy. Photochemical & Photobiological Sciences 17:660–669, 2018.
• Noè C, Hakkarainen M, Sangermano M: Cationic UV-Curing of Epoxidized Biobased Resins. Polymers 13(1):89, 2021.
• Bauer F et al.: Free-Radical Photopolymerization. Polymers, 2022.

Editorial status: 2 September 2026.

Whether the released process window is still met at the source actually in use only becomes clear from a measurement at the part. We measure the spectrum, irradiance and dose of your curing process and advise on suitable measurement technology – from the flat UVpad spectroradiometer that travels through the machine, through the RMD Pro radiometer and the XT series UV curing sensors for process monitoring, to UV irradiation chambers for reproducible curing trials in the laboratory. Send us your question.