Bonnie Bedolla, Flint Ink10.15.09
Editor’s Note: “Improving Properties of UV Flexo Inks By Curing Under Nitrogen” received a Carroll Scientific honorarium at the National Printing Ink Research Institute’s Annual Technical Conference in 2002.
Oxygen inhibits the free-radical curing of UV flexo inks. When UV flexo inks are cured in an environment with abundant oxygen, such as ambient air, the oxygen reacts with free radicals from the photoinitiator (PI) and growing polymer chains to form peroxy radicals. (Walia, 2001) Peroxy radicals are more stable than the radicals formed by the photoinitiator or propagating monomer, so the curing reaction does not continue. (Walia, 2001) This results in the ink being undercured for a given energy dose and photoinitiator level.
Flooding the curing environment with nitrogen dramatically reduces the oxygen concentration and minimizes the inhibition of the reaction. Curing printing ink under a nitrogen blanket provides several advantages. Test results demonstrate that UV flexo inks cured under nitrogen have higher crosslink density, even with a reduced level of photoinitiator. This provides more flexibility in optimizing the formulation since less of the formula needs to be reserved for photoinitiator. Inks with higher crosslink density also have increased resistance properties.
Photoinitiators and their fragments can cause many undesirable side effects, such as odor and migration, so reducing the level of photoinitiator required to cure the inks is very beneficial. Levels of extractable materials and residual odor are also lower for prints cured under nitrogen. Low odor and extractables levels can make UV flexo a more attractive option for a variety of markets, such as food and pharmaceutical packaging.
In this study, we looked at how the crosslink density and residual odor and extractable materials were affected by the atmosphere in which the ink was cured. We formulated a model ink containing some extractable materials. This ink was prepared with a photoinitiator compound at 4 percent, 7 percent, 10 percent and 13 percent. We also made the ink with two different single photoinitiators. Color strength of the inks was held constant so it wouldn’t affect curing.
We cured these inks under air and under a nitrogen blanket. A comparison of extractables, odor and cure response for the various combinations was then done. For these tests, prints were made on foil with a 600 line, 2.52 BCM anilox and cured with a 400 W/in mercury-vapor bulb at 120 m/min. For prints cured under nitrogen inerting, an oxygen level of 100-400 ppm was maintained.
Procedure: Degree of cure was tested using an Ink Cure Analyzer. This apparatus uses a test solution made of two solvents that penetrates the ink film. The test method is based on the premise that a well-crosslinked coating will not allow a solvent to penetrate it as much as a coating with a lower crosslink density. The first solvent is a volatile solvent that is deposited on the ink surface and penetrates it to varying degrees depending on the degree of crosslinking. The volatile solvent evaporates quickly, leaving only the second solvent, which has a low vapor pressure and contains a small amount of C14 marker. A Geiger-Mueller counter detects how much of the C14 marker is on the surface after various increments of time. An attached computer performs calculations to determine how deeply the film was penetrated. (ICA Technical Manual, 2003)
The instrument generates indices to correspond to the degree of crosslinking at different depths in the film. Indices I, II, III, and IV correspond to the crosslink density at different depths in the film. Index I reflects the degree of surface cure, while Index IV represents the degree of through-cure. Higher numbers correspond to a higher degree of crosslinking.
Results: The degree of crosslinking was much greater when the inks were cured under nitrogen, regardless of the photoinitiator level. The degree of crosslinking of the ink with 4 percent photoinitiator compound cured under nitrogen was greater than the ink containing 13 percent photoinitiator cured under oxygen.
When the inks were cured under oxygen, degree of cure increased as more photoinitiator was added. The presence of additional photoinitiator didn’t significantly affect the degree of cure when the inks were cured under nitrogen.
One of the concerns with using UV inks on some applications like food packaging is the potential for migration of raw materials or raw material fragments into the packaged food. (Herlihy, 2002)
Several potential routes for contamination of the food exist. In this test, the level of extractables/migratables resulting from offset migration was tested. Offset migration occurs when the printed side of the material comes in contact with the underside of the next layer of printed material, such as in a roll or stack. Mobile materials can transfer to the underside of the sheet and can then taint the packaged food since that side of the sheet becomes the inside of the package. (Herlihy, 2002)
When food is tainted by UV inks, it is due to either the presence of extractable raw materials or to poor cure or a combination of those two factors. (Herlihy, 2002) Since we included materials known to be extractable, the purpose of the test was to demonstrate whether there was a difference in the level of extractable material due to better or poorer cure under the different curing atmospheres.
Procedure: Prints were conditioned face-to-back under pressure, heat and humidity for one week to simulate extreme storage conditions. The back side of each print was then washed with a strong solvent to remove extractable materials. The extract was concentrated and then analyzed by GC-MS. The materials detected were quantified in micrograms of material per square centimeter.
Note: Some extractable materials were purposely used in this project to illustrate how formula changes and curing conditions affect the level of extractable materials.
Results: Lower levels of photoinitiator (PI) in the ink resulted in lower levels of extractable PI. Since the ink can be cured under nitrogen with less photoinitiator, the level of extractable PI was lowest in the ink with 4 percent PI.
The level of extractable monomer increased as the PI level decreased, since the ink was not cured as thoroughly. Since the crosslink density of the inks cured under nitrogen was greater, the level of extractable monomer for the ink with 4 percent PI cured under nitrogen was less than the ink with 13 percent PI cured under air.
Procedure: 240 cm2 of print were sealed in a glass jar and placed in a 38°C oven for one day, then cooled. An odor panel of two men and two women then rated the odor of the samples on a scale of 1-6, with higher numbers corresponding to less odor.
Results: At the lowest level of photoinitiator cured under both conditions, the prints cured under nitrogen had less odor than prints of the same ink cured under oxygen. At higher levels of photoinitiator, there was not a significant difference between the odor of the prints cured under oxygen and cured under nitrogen. The prints with 4 percent photoinitiator that were cured under nitrogen had an odor level comparable to the prints with higher levels (10 percent and 13 percent) of photoinitiator that were cured under oxygen.
Procedure: Two photoinitiators with similar molecular weights and absorption spectra were chosen for comparison.
Inks containing the two photoinitiators at different levels were cured under nitrogen and under air and the ink films were tested using
the Ink Cure Analyzer.
Results: The tests showed that the crosslink density with these two photoinitiators was similar under oxygen and under nitrogen. Although there were no significant differences between the efficiency of these two, nitrogen blanketing might provide a greater benefit with some photoinitiators relative to others.
• Curing under nitrogen increased the crosslink density of UV flexo inks.
• With nitrogen, less photoinitiator was needed and the cure still improved.
• Since less photoinitiator was used, less potentially extractable material was present.
• Since the crosslink density was higher, less residual material could migrate out of the cured prints.
• Since less photoinitiator was needed, one might be able to use non-extractable photoinitiators exclusively.
For instance, Bis (2,4,6-trimethylbenzoyl)-phenylphosphineoxide was not extracted, but it is only soluble up to 9 percent in acrylate monomer and up to 3 percent in oligomer. (Ciba, 1999) However, if only 4 percent photoinitiator is required to cure the ink under nitrogen instead of the 10 percent or more required under ambient air, this photoinitiator might be soluble enough to be used by itself and eliminate the need for the extractable (and more soluble) photoinitiators.
• Since the ink cures better when a nitrogen blanket is used, the level of extractable monomer is reduced, even when less photoinitiator is used.
• At lower levels of photoinitiator, the prints cured under nitrogen had less odor. Since the inks with low photoinitiator don’t cure well under oxygen, there could be uncured material present that creates an odor. At high levels of photoinitiator, unused photoinitiator could cause an odor.
• There weren’t significant differences between the two photoinitiators compared under air and nitrogen in this project, but this method could be used to evaluate which photoinitiators are most ideal for curing under nitrogen. Adjustments would be needed to optimize the photoinitiator package for curing under nitrogen.
• Ciba Specialty Chemicals Inc. 1999. Irgacure 651 Product Data Sheet. [Brochure]
• Ciba Specialty Chemicals Inc. 1999. Irgacure 184 Product Data Sheet. [Brochure]
• Ciba Specialty Chemicals Inc. 1999. Irgacure 819 Product Data Sheet. [Brochure]
• CON-TROL-CURE Ink Cure Analyzer Technical Manual. (2003). Chicago, IL: UV Process Supply, Inc. Retrieved April 22, 2003 from the World Wide Web: http://www.uvprocess.com/products/Curecon/Degrpol/ICA/iica.htm
• Herlihy, Dr. Shaun. The Use of Multifunctional Photoinitiators to Achieve Low Migration in UV Cured Printing Applications. The RadTech 2002 International North America Exhibition and Conference. 28 April – 1 May. Technical Conference Proceedings. 413-427.
• Sartomer Company. 1998. Esacure KIP 100F Product Data Sheet. [Brochure]
• Walia, Jagjit. (2001). The Process and Benefits of Nitrogen Inerting for UV Curable Coatings with and without the presence of Photoinitiator. The RadTech Europe 2001 Exhibition and Conference. 8-10 October 2001. Retrieved April 2002 from the World Wide Web: http://www.coatings.de/ articles/ecs01papers/walia/walia.htm
The author acknowledges the assistance of BASF; IST, RadTech, NPIRI, Colleen DeKay, Paul Gupta and Deju Koziol.
Oxygen inhibits the free-radical curing of UV flexo inks. When UV flexo inks are cured in an environment with abundant oxygen, such as ambient air, the oxygen reacts with free radicals from the photoinitiator (PI) and growing polymer chains to form peroxy radicals. (Walia, 2001) Peroxy radicals are more stable than the radicals formed by the photoinitiator or propagating monomer, so the curing reaction does not continue. (Walia, 2001) This results in the ink being undercured for a given energy dose and photoinitiator level.
Flooding the curing environment with nitrogen dramatically reduces the oxygen concentration and minimizes the inhibition of the reaction. Curing printing ink under a nitrogen blanket provides several advantages. Test results demonstrate that UV flexo inks cured under nitrogen have higher crosslink density, even with a reduced level of photoinitiator. This provides more flexibility in optimizing the formulation since less of the formula needs to be reserved for photoinitiator. Inks with higher crosslink density also have increased resistance properties.
Photoinitiators and their fragments can cause many undesirable side effects, such as odor and migration, so reducing the level of photoinitiator required to cure the inks is very beneficial. Levels of extractable materials and residual odor are also lower for prints cured under nitrogen. Low odor and extractables levels can make UV flexo a more attractive option for a variety of markets, such as food and pharmaceutical packaging.
Experimental
In this study, we looked at how the crosslink density and residual odor and extractable materials were affected by the atmosphere in which the ink was cured. We formulated a model ink containing some extractable materials. This ink was prepared with a photoinitiator compound at 4 percent, 7 percent, 10 percent and 13 percent. We also made the ink with two different single photoinitiators. Color strength of the inks was held constant so it wouldn’t affect curing.
We cured these inks under air and under a nitrogen blanket. A comparison of extractables, odor and cure response for the various combinations was then done. For these tests, prints were made on foil with a 600 line, 2.52 BCM anilox and cured with a 400 W/in mercury-vapor bulb at 120 m/min. For prints cured under nitrogen inerting, an oxygen level of 100-400 ppm was maintained.
Cure Analysis
Procedure: Degree of cure was tested using an Ink Cure Analyzer. This apparatus uses a test solution made of two solvents that penetrates the ink film. The test method is based on the premise that a well-crosslinked coating will not allow a solvent to penetrate it as much as a coating with a lower crosslink density. The first solvent is a volatile solvent that is deposited on the ink surface and penetrates it to varying degrees depending on the degree of crosslinking. The volatile solvent evaporates quickly, leaving only the second solvent, which has a low vapor pressure and contains a small amount of C14 marker. A Geiger-Mueller counter detects how much of the C14 marker is on the surface after various increments of time. An attached computer performs calculations to determine how deeply the film was penetrated. (ICA Technical Manual, 2003)
The instrument generates indices to correspond to the degree of crosslinking at different depths in the film. Indices I, II, III, and IV correspond to the crosslink density at different depths in the film. Index I reflects the degree of surface cure, while Index IV represents the degree of through-cure. Higher numbers correspond to a higher degree of crosslinking.
Results: The degree of crosslinking was much greater when the inks were cured under nitrogen, regardless of the photoinitiator level. The degree of crosslinking of the ink with 4 percent photoinitiator compound cured under nitrogen was greater than the ink containing 13 percent photoinitiator cured under oxygen.
When the inks were cured under oxygen, degree of cure increased as more photoinitiator was added. The presence of additional photoinitiator didn’t significantly affect the degree of cure when the inks were cured under nitrogen.
Extractables
One of the concerns with using UV inks on some applications like food packaging is the potential for migration of raw materials or raw material fragments into the packaged food. (Herlihy, 2002)
Several potential routes for contamination of the food exist. In this test, the level of extractables/migratables resulting from offset migration was tested. Offset migration occurs when the printed side of the material comes in contact with the underside of the next layer of printed material, such as in a roll or stack. Mobile materials can transfer to the underside of the sheet and can then taint the packaged food since that side of the sheet becomes the inside of the package. (Herlihy, 2002)
When food is tainted by UV inks, it is due to either the presence of extractable raw materials or to poor cure or a combination of those two factors. (Herlihy, 2002) Since we included materials known to be extractable, the purpose of the test was to demonstrate whether there was a difference in the level of extractable material due to better or poorer cure under the different curing atmospheres.
Procedure: Prints were conditioned face-to-back under pressure, heat and humidity for one week to simulate extreme storage conditions. The back side of each print was then washed with a strong solvent to remove extractable materials. The extract was concentrated and then analyzed by GC-MS. The materials detected were quantified in micrograms of material per square centimeter.
Note: Some extractable materials were purposely used in this project to illustrate how formula changes and curing conditions affect the level of extractable materials.
Results: Lower levels of photoinitiator (PI) in the ink resulted in lower levels of extractable PI. Since the ink can be cured under nitrogen with less photoinitiator, the level of extractable PI was lowest in the ink with 4 percent PI.
The level of extractable monomer increased as the PI level decreased, since the ink was not cured as thoroughly. Since the crosslink density of the inks cured under nitrogen was greater, the level of extractable monomer for the ink with 4 percent PI cured under nitrogen was less than the ink with 13 percent PI cured under air.
Odor Comparison
Procedure: 240 cm2 of print were sealed in a glass jar and placed in a 38°C oven for one day, then cooled. An odor panel of two men and two women then rated the odor of the samples on a scale of 1-6, with higher numbers corresponding to less odor.
Results: At the lowest level of photoinitiator cured under both conditions, the prints cured under nitrogen had less odor than prints of the same ink cured under oxygen. At higher levels of photoinitiator, there was not a significant difference between the odor of the prints cured under oxygen and cured under nitrogen. The prints with 4 percent photoinitiator that were cured under nitrogen had an odor level comparable to the prints with higher levels (10 percent and 13 percent) of photoinitiator that were cured under oxygen.
Photoinitiator Comparison
Procedure: Two photoinitiators with similar molecular weights and absorption spectra were chosen for comparison.
Inks containing the two photoinitiators at different levels were cured under nitrogen and under air and the ink films were tested using
the Ink Cure Analyzer.
Results: The tests showed that the crosslink density with these two photoinitiators was similar under oxygen and under nitrogen. Although there were no significant differences between the efficiency of these two, nitrogen blanketing might provide a greater benefit with some photoinitiators relative to others.
Conclusions
• Curing under nitrogen increased the crosslink density of UV flexo inks.
• With nitrogen, less photoinitiator was needed and the cure still improved.
• Since less photoinitiator was used, less potentially extractable material was present.
• Since the crosslink density was higher, less residual material could migrate out of the cured prints.
• Since less photoinitiator was needed, one might be able to use non-extractable photoinitiators exclusively.
For instance, Bis (2,4,6-trimethylbenzoyl)-phenylphosphineoxide was not extracted, but it is only soluble up to 9 percent in acrylate monomer and up to 3 percent in oligomer. (Ciba, 1999) However, if only 4 percent photoinitiator is required to cure the ink under nitrogen instead of the 10 percent or more required under ambient air, this photoinitiator might be soluble enough to be used by itself and eliminate the need for the extractable (and more soluble) photoinitiators.
• Since the ink cures better when a nitrogen blanket is used, the level of extractable monomer is reduced, even when less photoinitiator is used.
• At lower levels of photoinitiator, the prints cured under nitrogen had less odor. Since the inks with low photoinitiator don’t cure well under oxygen, there could be uncured material present that creates an odor. At high levels of photoinitiator, unused photoinitiator could cause an odor.
• There weren’t significant differences between the two photoinitiators compared under air and nitrogen in this project, but this method could be used to evaluate which photoinitiators are most ideal for curing under nitrogen. Adjustments would be needed to optimize the photoinitiator package for curing under nitrogen.
References
• Ciba Specialty Chemicals Inc. 1999. Irgacure 651 Product Data Sheet. [Brochure]
• Ciba Specialty Chemicals Inc. 1999. Irgacure 184 Product Data Sheet. [Brochure]
• Ciba Specialty Chemicals Inc. 1999. Irgacure 819 Product Data Sheet. [Brochure]
• CON-TROL-CURE Ink Cure Analyzer Technical Manual. (2003). Chicago, IL: UV Process Supply, Inc. Retrieved April 22, 2003 from the World Wide Web: http://www.uvprocess.com/products/Curecon/Degrpol/ICA/iica.htm
• Herlihy, Dr. Shaun. The Use of Multifunctional Photoinitiators to Achieve Low Migration in UV Cured Printing Applications. The RadTech 2002 International North America Exhibition and Conference. 28 April – 1 May. Technical Conference Proceedings. 413-427.
• Sartomer Company. 1998. Esacure KIP 100F Product Data Sheet. [Brochure]
• Walia, Jagjit. (2001). The Process and Benefits of Nitrogen Inerting for UV Curable Coatings with and without the presence of Photoinitiator. The RadTech Europe 2001 Exhibition and Conference. 8-10 October 2001. Retrieved April 2002 from the World Wide Web: http://www.coatings.de/ articles/ecs01papers/walia/walia.htm
Acknowledgements
The author acknowledges the assistance of BASF; IST, RadTech, NPIRI, Colleen DeKay, Paul Gupta and Deju Koziol.