Photoinitiators Reevaluated: PI-Free EB is About More than Just Migration

Between UV and EB, EB often has been the preferred choice for indirect food applications because it does not require a photoinitiator (PI) for free-radical polymerization, and thus, there is no concern for PI or PI-fragment migration into foodstuffs. 1-5 This abridged migration argument usually is the main takeaway from the fact that EB is PI-free; however, the discussion should encompass much more than this one advantage for food packaging. To broaden this discussion would better enable companies to make an informed decision between UV and EB technologies, be it for food packaging or any other application. As migration concerns are augmented, applications expand, regulations tighten, sustainability becomes increasingly important and cost remains an ever-looming entity, it is time to reevaluate the role of photoinitiators.

A discussion on the disuse of PIs is a tricky endeavor in an EB and UV magazine, and it is important to preface this argument with a bit of context. Though UV and EB technologies are discussed as both complementary technologies and competitive technologies, neither is wholly true. The comparison of technologies is a spectrum. On one end, there are the applications where only light is viable – dental applications, nail gels, etc.; on the other end, there is the region in which EB is the only viable option – opaque laminations, heavily pigmented and relatively thick inks/coatings, etc. In the middle of the spectrum, both technologies are functional. The advantages and disadvantages of each mechanism can be weighed and considered, and it is here, in this middle region, that the conversation about the impact of PIs is salient.

Migration Factors

While the discussion surrounding PIs should extend beyond migration, that is not to say that migration itself is not an important issue. Without too much imagination, it is possible to expand photoinitiator migration concerns from food packaging to include other sensitive areas, such as personal care, pharmaceutical and medical applications. In these types of applications, migration can be mitigated successfully in various ways, such as the synthesis of polymeric and multifunctional photoinitiators. In fact, considerable effort is put into the development of new mitigation strategies. 2,6-8 Yet, each of these strategies has the potential to be in some ways limiting. In terms of formulation, low-migration PIs may lack the variety seen in traditional PIs, demand a higher cost and be of a higher viscosity. The addition of a barrier layer is another method of achieving low migration, but it also imposes limits on the packaging construction possibilities and the sustainability of the packaging. And despite these strategies, migration remains an ever-present concern as more is learned about the effects of PIs. 6

End-of-life migration is a more recent consideration as more photopolymerized products are composted or recycled. 7, 9-11 Recycled paperboard, for example, has been shown to contain PIs, and migration of some of these PIs has been demonstrated when it is used in its second life as food packaging. PIs also have been identified in dust from electronic waste recycling facilities in China. These examples illustrate the need to take PI migration and safety into account from cradle to grave – or in a circular economy, cradle to cradle – when developing products with photopolymerized inks, coatings and/or adhesives.

Similarly, evidence suggests that PI migration into products is only a partial perspective, and attention should be paid to PI migration out of products into our environment as well. Liu, et al. has reported PIs found in indoor dust both in China and in Canada, highlighting the inclusion of dust ingestion as a potential human exposure pathway. 12,13 The concentration of PIs in Toronto, Canada, (geometric mean: 1483 ng/g) was significantly higher than that observed in Beijing, China, (geometric mean: 610 ng/g). The authors attribute this higher concentration to increased pre-packaged food consumption in North America and thus an increased amount of food packaging indoors. Currently, Liu, et al. reports estimated daily intake of PI via indoor dust ingestion in ng/kg-body weight per day, three orders of magnitude less than the established daily intake for benzophenone (30 µg/kg-bw per day). While this route of PI exposure does not pose an immediate health risk, the comparison between Canada and China suggests as photopolymerization continues to increase in popularity, PI concentrations in indoor dust also likely will increase. In addition, PIs used in non-sensitive applications are less likely to be deemed low migration than those used in food packaging, and consideration should be given as to how this affects their outward migration. Perhaps non-sensitive, household items with photopolymerized components (e.g., flooring, magazines, labels) contribute a higher percentage of PI concentrations in dust than food packaging. Overall, more studies are needed to evaluate the true impact of PIs in the environment, especially as measurable amounts have been detected in both human sera and breastmilk. 14, 15

Application Challenges

Photoinitiators are the cause of other application challenges, too. PIs can beget odor and result in the blooming of coatings. 3,7 PIs can introduce unwanted color, have solubility issues and/or not be shelf-stable for extended periods of time. Additionally, it is common for a formulation to have more than one PI and to include photosensitizers and synergists to accelerate the initiation reaction and improve depth of cure. These multicomponent initiation systems add complexity and also take up a usable fraction of a formulation, especially in optically dense inks and coatings. In some cases, PI occupation of this fraction may be of little consequence, but in others it may be significant. Consider, for example, the use of a radiation-curable binder for Li-ion batteries – an application where researchers/manufacturers continually strive to reduce the amount of inactive materials (those that do not contribute directly to energy storage). 16-19

Outdoor applications, while certainly possible with photopolymerized systems, also can present an increased challenge compared to EB-polymerized systems. “Eliminating photoinitiators in exterior coatings helps reduce their susceptibility to UV-induced degradation,” according to Jade Muns, chemist at Sherwin Williams. Residual PIs and PI fragments can contribute to poor weathering performance by negatively impacting the photooxidation stability of the ink/coating and must be chosen carefully to minimize yellowing with age. 2,3,20-22 Moreover, the addition of UV absorbers and HALS (hindered amine light stabilizer, a radical scavenger) have been shown to significantly improve the resistance of photo- and EB-polymerized coatings to solar UV radiation. In photopolymerized coatings, however, UV absorbers can impede the light required to react with the PIs, hindering polymerization, if their absorption profile overlaps with that of the PIs. This competition for light constricts the amount and the selection of UV absorbers that can be included in a formulation.

Regulations, Sustainability and Cost

In addition to more application-specific challenges, there are overarching ramifications to the use of photoinitiators – the first of which is regulations. A great many chemicals are regulated – PIs are far from unique in that way – but what has caused strife in recent years is the changing of classification and the inclusion or exclusion from various lists (e.g., Swiss, Nestlé, REACH, EuPIA, etc.) as the potential health effects of PIs and PI transformation products are brought forward. 23-25 Even non-regulatory bodies, like Nestlé, have influenced the perception and restricted the use of certain chemicals. And with insufficient toxicological data available for many PIs, the landscape is expected to remain somewhat in flux for the foreseeable future as more research is completed. For example, of the 24 photoinitiators found by Liu, et al. in household dust, only benzophenone has an established daily intake limit. 12

Each time a photoinitiator is reclassified, especially if it is widely used, it sets in motion a series of events. Often there is pushback, much discussion about the futility of fighting regulatory bodies and the motivation to generate new, compelling research to persuade regulators to reason. In the meantime, companies may be forced to reformulate and requalify products while reassuring customers there will be no performance loss. If all PIs were interchangeable, regulatory changes might be of less consequence; however, that is not the case – there are absorption spectra, solubilities, migration limits and interactions with other initiation system components to consider. 1-3 Furthermore, there is no guarantee that the replacement PI itself won’t be under scrutiny in a few years. A PI-free formulation is not free of regulation challenges, but it is free of a whole category of them.

Another ramification of using photoinitiators is their impact on the sustainability of a product. Overall, photopolymerization is a very green technology compared to traditional thermal polymerization. Yet, the sustainability of photopolymerization does not negate the fact that there is some negative contribution from the PIs and associated synergists and sensitizers. Photoinitiators (and, often, their precursors) must be synthesized in several steps, using heat and solvent and generally producing waste and side products. 26,27 The exact figures for the environmental impact of PI production are difficult to pinpoint; however, “tightened environmental regulations” in China were stated as one of the reasons for the TPO shortage in 2018. 28-30 Transportation of the PI, as well as any wasted product due to the shelf-stability of the PI, also can be tallied. No matter how eco-friendly the process for synthesizing PIs becomes, it still is an avoidable environmental cost in situations where PI-free EB inks/coatings are an alternative solution.

Last, but certainly not least, there is a cost associated with using photoinitiators. Obviously, there is the cost of the PI itself (and, likely, other initiation components), but consider, too, the other costs accompanying the choice to use PIs. These include the cost of reformulating and requalifying when regulations change, when the lamp technology changes (e.g., mercury arc to LED), when raw material costs increase and/or when a supply chain becomes unreliable, as in 2018. Also to be considered are the cost to update customers about the reformulation and the development costs for safer and more migration-resistant PIs. Some of these are supplier costs, but they inevitably are passed on to printers and coaters. 31,32 These end-users pay in a variety of ways as well, such as the cost of limiting an exterior warranty or battery performance 17, the cost of a work-around when flirting with the edge of a PI’s capability or the cost of a recall on baby formula and the resulting fallout. 33 Choosing to use PIs also is choosing a continual investment to support that choice. These real and potential costs should be included when comparing the long-term operating costs of UV and EB.


The purpose of this discussion is not to say the EB is the perfect alternative. EB also has its disadvantages and shortcomings. For instance, it requires nitrogen to overcome oxygen inhibition. It can change the color of some substrates and can have other unintended consequences on some substrates. Because EB doesn’t require a PI, there is a learning curve, too, in understanding how it reacts with various types of monomers to produce initiating radicals.4 Nor is the purpose of this discussion to say that photoinitiators aren’t capable of wonderful things. PIs are an additional, controllable variable that can be used for cascading reactions and to restrict areas of polymerization. Nevertheless, there is a complexity inherent in the use of a photoinitiator that does not exist in the same way for PI-free formulations.

In summary, if an application falls in the portion of the spectrum where UV and EB technologies compete, when weighing the decision of which to choose, be cognizant of the multitude of factors that photoinitiators influence – migration and other application challenges, regulations, sustainability and cost. The comparison of technologies for a particular application should include much more than just a discussion about the influence of photoinitiators, but recognizing a PI’s influence extends further than just migration leads to a more knowledgeable decision. And, overall, companies that make an informed decision – that have properly evaluated the risks – certainly will be better prepared, be a better ambassodor of the technology and, hopefully, be successful in their pursuits.

References for this article are available at

Sage-SchisselSage Schissel, Ph.D.
Applications Specialist
PCT Ebeam and Integration LLC



  1. Woods, R.J., Pikaev, A.K., Applied Radiation Chemistry: Radiation Processing. John Wiley & Sons, Inc.: New York, 1994.
  2. Fouassier, J., Lalevée, Photoinitiators for Polymer Synthesis: Scope, Reactivity, and Efficiency. Wiley VCH: Weinheim, 2012.
  3. Pappas, S.P., Radiation Curing: Science and Technology. Springer Science & Business Media: New York, 1992.
  4. Thiher, N.L.K., Jessop, J.L.P., Schissel, S.M., Counting Radicals: Methods to Measure Radiation Yields of Monomers in EB Polymerization, UV+EB Tech. (2), 2019, pp. 32-40.
  5. Lapin, S.C., What are the Fundamental Differences between UV and EB Technology?, UV+EB Tech. (1), 2015, pp. 10-11.
  6. Aparicio, J.L., Elizalde, M., Migration of Photoinitiators in Food Packaging: A Review, Packag. Technol. Sci., 28, 2015, pp. 181-203.
  7. Zhou, J., Allonas, X., Ibrahim, A., Liu, X., Progress in the Development of Polymeric and Multifunctional Photoinitiators, Progress in Polymer Science, 2019, 101165.
  8. Ludwig, P.E., Ossenbach, A.J., Photoinitiators: Challenges in Food Packaging Applications, UV+EB Tech. (4), 2016, pp. 42-46.
  9. Biedermann, M., Ingenhoff, J., Zurfluh, M., Richter, L., Simat, T., Harling, A., Altkofer, W., Helling, R., Grob, K., Migration of Mineral Oil, Photoinitiators and Plasticisers from Recycled Paperboard into Dry Foods: A Study Under Controlled Conditions, Food Additives & Contaminants: Part A, 30 (5), 2013, pp. 885-898.
  10. Geueke, B., Groh, K., Muncke, J., Food Packaging in the Circular Economy: Overview of Chemical Safety Aspects for Commonly Used Materials, J. Cleaner Production, 193, 2018, pp. 491-505.
  11. Li, J., Li, W., Gao, X., Liu, L., Shen, M., Chen, H., Zhu, M., Zeng, L., Zeng, E.Y., Occurrence of Multiple Classes of Emerging Photoinitiators in Indoor Dust from E-waste Recycling Facilities and Adjacent Communities in South China and Implications for Human Exposure, Environmental International, 136, 2020, 105462.
  12. Liu, R., Mabury, S.A., Identification of Photoinitiators, Including Novel Phosphine Oxides, and Their Transformation Products in Food Packaging Materials and Indoor Dust in Canada, Environ. Sci. Tech., 52 (17), 2018, pp. 10089-10096.
  13. Liu, R., Lin, Y., Hu, F., Liu, R., Ruan, T., Jiang, G., Observation of Emerging Photoinitiator Additives in Household Environment and Sewage Sludge in China, Environ. Sci. Tech., 50 (1), 2016, pp. 97-104.
  14. Liu, R., Mabury, S.A., First Detection of Photoinitiators and Metabolites in Human Sera from United States Donors, Environ. Sci. Tech., 5 (7), 2019, pp. 419-423.
  15. Liu, R., Mabury, S.A., Photoinitiators in Breast Milk from United States Donors: Occurrence and Implications for Exposure in Infants, Environ. Sci. Tech., 6 (12), 2019, pp. 702-707.
  16. Foreman, E., Zakri, W., Sanatimoghaddam, M.H., Modjtahedi, A., Pathak, S., Kashkooli, A.G., Garafolo, N.G., Farhad, S., A Review of Inactive Materials and Components of Flexible Lithium-Ion Batteries, Adv. Sustain. Sys., 1 (11), 2017, 1700061.
  17. Du, Z., Janke, C.J., Li, J., Wood, D.L., High-Speed Electron Beam Curing of Thick Electrode for High Energy Density Li-ion Batteries, Green Energy & Environ., 4, 2019, pp. 375-381.
  18. Du, Z., Schissel, S., Electrons In, Electrons Out: Electron Beam for Battery Production, UV+EB Tech. (3), 2022, pp. 46-51.
  19. Sigel, G., Baumann, D., Hamann, O., Abbott, P., Hensley, D., A New Class of UV-Curable Binders for Battery Applications, UV+EB Tech. (2), 2023, pp. 20-28.
  20. Nguyen, T.V., Le, X.H., Dao, P.H., Decker, C., Nguyen-Tri, P., Stability of Acrylic Polyurethane Coatings Under Accelerated Aging Tests and Natural Outdoor Exposure: The Critical Role of the Used Photo-stabilizers, Prog. Org. Coat., 124, 2018, pp. 137-146.
  21. Ruiz, C.S.B., Machado, L.D.B., Accelerated Weathering of UV/EB Curable Clearcoats, Nucl. Instr. and Meth. in Phys. Res. B, 236, 2005, pp. 599-605.
  22. Seubert, C.M., Nichols, M.E, Cooper, V.A., Gerlock, J.L., The Long-term Weathering Behavior of UV Curable Clearcoats I. Bulk Chemical and Physical Analysis, Polym. Degrad. Stab., 81, 2003, pp. 103-115.
  23. Postle, S.R., Photoinitiator Selection to Advance the UV Curing Industry in an Uncertain World, RadTech UV+EB Technology Conference, 2020.
  24. Van Gelder, S., Bellotti, E., Bailey, S.E., Postle, S.R., Transcending Regulatory Issues: Alternates to Omnirad 369, RadTech UV+EB Technology Conference, 2018.
  25. Gould, M., Gainey, S., New Classification for TPO has Multinational Impacts, UV+EB Tech. (2), 2023, pp. 44-45.
  26. Chu, X., Preparation Method of TPO Photoinitiator, Patent CN103333206A, 2013.
  27. Peng, G., Preparation Method of 2-ispropylthioxanthone, Patent CN101570534A, 2008.
  28. Langhammer, J., Photoinitiators: A “Run” on Substitutes, 2018,
  29. Savastano, D., Key Photoinitiator Shortage Causing Headaches for Inks and Coatings Industry, 2018,
  30. Savastano, D., Photoinitiator Concerns Continue to Mount, 2018,
  31. Cates, N., Business Leaders Project Direction for Curing in 2019 and Beyond, UV+EB Tech. (4), 2018, pp. 34-36.
  32. Sun Chemical to Increase Prices on Various Inks and Coatings, 2019,
  33. Baby Milk Scare Widens in Europe, BBC News, Nov. 22, 2015,