The transition from mercury arc lamps to UV LEDs in manufacturing environments is not as simple as swapping one lamp for another. The adhesive chemistry — specifically the photoinitiators embedded in the resin — was often designed around the spectral output of mercury arc lamps, and understanding how that chemistry responds differently to LED output is what separates a successful lamp migration from a process that produces unreliable bonds and unexplained failures.

The Spectral Mismatch Problem

Mercury arc lamps emit a broad spectrum of ultraviolet and visible light, with pronounced emission peaks at 254 nm, 303 nm, 313 nm, 334 nm, 365 nm, 405 nm, 436 nm, and other wavelengths. Traditional UV photoinitiators were formulated to absorb across this spectrum, often with broad absorption bands covering 300–400 nm. A single photoinitiator molecule in a mercury lamp adhesive might absorb meaningfully at 313, 334, and 365 nm simultaneously.

UV LEDs emit at a single narrow peak, typically 10–20 nm wide at half-maximum. A 365 nm LED produces almost no output at 313 nm or 405 nm. A 395 nm LED produces almost no output at 365 nm. When a mercury-lamp adhesive is exposed to a UV LED, only the portion of the photoinitiator’s absorption band that overlaps with the LED’s narrow emission contributes to cure. The rest of the absorption spectrum is effectively wasted.

Common Mercury-Era Photoinitiators and Their LED Compatibility

Benzophenone and benzophenone derivatives are classic free-radical photoinitiators that absorb primarily in the 280–330 nm range. They are poorly suited to UV LED curing at 365 nm and beyond because their absorption drops steeply above 330 nm. Formulations relying on benzophenone as the primary photoinitiator may not cure adequately under any commercially available UV LED.

Irgacure 651 (DMPA) has absorption primarily below 350 nm, with meaningful response out to about 365 nm. This photoinitiator performs reasonably under 365 nm LED lamps but shows declining efficiency at 385 nm and above. Adhesives formulated with DMPA as the primary initiator may require significantly higher dose under longer-wavelength LEDs.

Irgacure 184 (hydroxy-cyclohexyl-phenyl-ketone) is a widely used alpha-hydroxy ketone photoinitiator with good absorption in the 300–370 nm range. It functions reasonably under 365 nm LED illumination and is often used in LED-compatible formulations, though its efficiency decreases at 385 nm and higher.

Irgacure 819 (bisacylphosphine oxide, BAPO) is a photoinitiator specifically designed to absorb into the 370–420 nm range, making it well suited to UV LED curing systems operating at 385–405 nm. Its broader absorption spectrum and high molar absorption coefficient at longer wavelengths make it a frequent choice in adhesives reformulated for LED compatibility.

Irgacure TPO (diphenyl(2,4,6-trimethylbenzoyl)phosphine oxide) absorbs from approximately 350 nm to 420 nm, with good response across the typical UV LED emission range. TPO is commonly found in LED-optimized adhesive formulations and supports efficient curing at 365, 385, 395, and 405 nm.

Why Reformulation Is Often Required

An adhesive formulated for mercury lamp curing may contain photoinitiators that work reasonably well at 365 nm but show degraded performance at 395 or 405 nm. Switching to a longer-wavelength LED without verifying photoinitiator compatibility can result in dramatically longer cure times, incomplete through-cure, or surface tack that passes quick inspection but indicates incomplete polymerization.

Rather than adjusting irradiance or dose settings to compensate for a photoinitiator mismatch, the correct approach in most cases is to evaluate whether the adhesive itself should be changed to an LED-optimized formulation. Adhesive manufacturers have developed LED-specific product lines precisely because the spectral mismatch with mercury-era photoinitiators is a real limitation — not something that can be fully overcome with lamp settings.

If you are migrating a process from mercury to LED curing and are uncertain whether your current adhesive is LED-compatible, Email Us and an Incure engineer will help evaluate your photoinitiator chemistry and recommend a validation approach.

The Role of Sensitizers

One strategy that adhesive formulators use to extend the wavelength response of otherwise short-wavelength photoinitiators is the addition of photosensitizers — molecules that absorb at longer wavelengths and transfer energy to the primary photoinitiator through a process called triplet energy transfer or Dexter mechanism. This allows a photoinitiator with poor direct absorption at 395 nm to be activated indirectly through a sensitizer that absorbs there.

Thioxanthone derivatives are a common class of photosensitizers used to extend the UV LED compatibility of otherwise mercury-lamp-optimized formulations. When an adhesive datasheet lists a thioxanthone as a component or lists compatibility with wavelengths longer than the primary photoinitiator’s native absorption range, sensitization chemistry is likely contributing to that compatibility.

Depth of Cure and Wavelength

Mercury lamps, by delivering photons across a range of wavelengths simultaneously, engage the adhesive’s absorption at multiple points in the cure stack. Shorter wavelengths activate photoinitiators near the surface; longer wavelengths, which penetrate more deeply into the adhesive before being absorbed, contribute to through-cure.

UV LEDs, with their single-wavelength output, drive initiation at one depth profile. If the selected wavelength is strongly absorbed near the surface — as is often the case with 365 nm in highly loaded formulations — through-cure can be limited. This is why adhesive manufacturers for thick-section applications sometimes recommend 395 or 405 nm LEDs: the lower per-photon absorption in many photoinitiator systems at these wavelengths translates to greater penetration depth before full attenuation.

Validating LED Compatibility

Before any production process change involving a lamp technology transition, cure quality should be validated by physical testing: mechanical pull tests, lap shear tests, or other bond strength measurements relevant to the application. Visual inspection and surface tack assessment alone are insufficient because partially cured adhesive can appear fully cured while losing significant mechanical performance.

Measuring the degree of cure by differential scanning calorimetry or FTIR spectroscopy provides a more rigorous assessment of polymerization completeness and can identify under-cure conditions that physical testing alone might miss.

Contact Our Team to discuss lamp migration validation strategies and LED-compatible adhesive evaluation for your process.

Visit www.incurelab.com for more information.