LED-UV Curing

Executive summary

Light-emitting-diode UV sources have gone from curiosity to mainstream on label presses, sheet-fed offset and screen lines within a decade. They cure instantly at a fraction of the electricity, contain no mercury, produce no ozone and put almost no heat into the substrate. They also change the chemistry: an ink formulated for a mercury-vapour lamp may not cure under LED at all. This paper explains the physics of the two sources, the photoinitiator chemistry that LED requires, the practical differences in press operation, and how Uniform Industries formulates LED and hybrid grades.

1. Two very different light sources

A medium-pressure mercury lamp emits a broad spectrum from 200 to 450 nm with strong lines at 254, 313 and 365 nm, plus visible light and a great deal of infra-red. An LED array emits a narrow band, typically 385, 395 or 405 nm, at high irradiance (up to 20 W/cm² at the window) with no short-wave UV and almost no infra-red. The short-wave lines of the mercury lamp are what cure the surface of a conventional ink — they are absorbed in the top micrometre and overcome oxygen inhibition. LED light penetrates deeply and cures from the bottom up; the surface is the last thing to cure and the first thing to be inhibited by oxygen.

Figure 1 – Mercury and LED emission spectra against short- and long-wave photoinitiator absorption (schematic).

2. Photoinitiators for long-wave light

LED-curable formulations need initiators that absorb at 385 – 405 nm. The candidates are the acylphosphine oxides (BAPO, and until 2025 TPO), thioxanthones (including ITX, restricted for food packaging) and newer polymeric and oligomeric long-wave initiators. TPO’s reclassification as reprotoxic category 1B under CLP in 2025 and its inclusion in the EuPIA Charter exclusion list removed the industry’s workhorse LED initiator; reformulation moved to BAPO, polymeric phosphine oxides and multifunctional acylphosphine derivatives that offer the same absorption with far lower migration and no restricted classification. Pigment competition is the second problem: yellow and white absorb or scatter near-UV strongly, so opaque white and dense black remain the limiting colours and set the lamp requirement.

3. Beating oxygen at the surface

Without short-wave UV, surface cure under LED depends on radical flux beating oxygen diffusion. Formulators raise irradiance (high-power arrays close to the substrate), use amine synergists or, for low-migration grades, polymeric amine-free co-initiators that scavenge peroxy radicals, and select oligomers with inherently fast surface cure. Nitrogen inerting works but is rarely practical on a label press. The right combination gives a tack-free, rub-resistant surface at production speed; the wrong one gives a print that is hard underneath and tacky on top — the classic LED failure.

4. Hybrid inks for a mixed pressroom

Most converters run mercury and LED side by side during a transition, and many presses carry both lamp types. Uniform Industries formulates its current UV flexo and coating grades as hybrids: a balanced initiator package with a long-wave initiator for LED depth cure and a short-wave initiator for mercury surface cure, without TPO, BP or ITX. One ink runs under either source, which removes duplicate inventory and the risk of loading the wrong ink on the wrong press.

Figure 2 – Output decay with lamp hours: mercury loses short-wave UV first; LED is stable (illustrative).

5. What changes on press

Energy: an LED array draws roughly a third to a fifth of the power of an equivalent mercury system, with no standby consumption and no warm-up.

Heat: substrate temperature rises by a few degrees rather than tens; thin films, shrink sleeves and heat-sensitive labels print flat, and chill rolls can often be switched off.

Lamp life: 20 000 hours or more with slow output decay, against 1000 – 2000 hours and rapid short-wave decay for mercury; but LED output still degrades and must be measured.

Measurement: a radiometer matched to the LED wavelength is essential — mercury-band radiometers read LED output wrongly. Record dose at the substrate plane at set-up and use it as the release parameter.

Safety: no mercury, no ozone, no hot quartz; but intense near-UV and visible light require the same eye and skin shielding.

Cure verification: thumb-twist and solvent rub immediately after the lamp and again after five minutes, because LED-cured surfaces can post-cure slightly in ambient light.

Figure 3 – Electrical power and substrate heating, mercury versus LED, on a 330 mm label press (illustrative).

6. Where LED is ahead, and where it is not

LED is the clear choice for narrow-web labels, shrink sleeves and any heat-sensitive film, and for screen and sheet-fed applications where energy and heat matter. It is still more demanding for thick, opaque white screen deposits and for coatings where a mirror-gloss surface must cure tack-free at very high speed; there, hybrid systems on presses with an end-of-line mercury or LED-plus-mercury station remain common. The gap is closing with each generation of arrays and initiators.

Key takeaways

LED emits narrow-band long-wave UV; it cures from the bottom up and needs its own initiator chemistry.

TPO is gone; BAPO, polymeric phosphine oxides and amine-free co-initiators replace it.

Surface cure under LED is an oxygen problem solved by irradiance and formulation.

Hybrid inks cure under either source and make the transition manageable.

References: CLP Regulation (EC) 1272/2008 (TPO harmonised classification); EuPIA Charter 2026 Annex 1; RadTech LED-UV measurement guidelines; IEC 62471 (photobiological safety).