Electron Beam Curing in PVC Wide-Web Applications

One of the biggest drawbacks of EB curing on films is the effect of ionizing radiation on several plastics including polyvinylamide (PA) and polyvinylchloride (PVC), where discoloration of the film or an odor could occur as a result of degradation of the film (Table 1). The film that is most sensitive to ionizing radiation is polyvinyl chloride, due to degradation the PVC backbone. However, based on mechanical and physical properties, it also is one of the films most suitable for flooring applications. The most commonly treated substrates are polyethylene and copolymers to induce crosslinking, which results in improvements in thermal and mechanical properties.²

The flooring industry utilizes UV/EB coating technology in three primary areas: 1) floor tiles, including luxury vinyl flooring (LVT); 2) wide-width residential and commercial sheet goods; and 3) wood coatings on solid, engineered and rigid core technology. A residential floor tile is composed of a decorative (printed) vinyl layer laminated/embossed to a tile base typically comprising calcium carbonate and a binder material. To enhance the performance properties of the vinyl wear layer on a floor tile, a UV/EB-curable topcoat is applied as either high gloss or low gloss. Improved scratch resistance, gloss retention and stain resistance are just a few benefits obtained by utilizing UV/EB coating technology. For a UV/EB system, the gloss level desired for the flooring product dictates the process used to cure the formulation.

In the case of flooring applications, the choice of UV vs. EB technology is dictated by processing requirements for the substrate and throughput of material. Electron beam offers several advantages over UV cure, as documented by several authors.^1-4 Some of these advantages include fast line speeds for topcoat web applications, where the ionizing radiation and dose rates do not affect the substrate properties by degradation that affects the visual characteristics of the final product. Other advantages include the fact that photoinitiators are not required, thereby lowering the cost of the final product. However, several disadvantages exist as well: high capital cost, cost of nitrogen inerting and – from a process chemistry side – inability to change gloss of the product by using a single UV/EB formulation without the use of pre-UV lamps. In addition, any unexpected failure in the thin foil – due to age or substrate hitting the window – will result in a substantial amount of down time due to changeover.

In the mid-’90s, low-energy (100 to 150 kV) electron beam (EB) technology curing was integrated into Armstrong’s radiation program, with development in the areas of high performance wear layers for flooring applications. This led to the first Electro Curtain ESI machine being installed for coated film products for tile in 1995. Three years later a second ESI machine was added for curing adhesive on the back of tile (Photo 1).

One of the problems associated with wide-width processing of rigid vinyl film is discoloration of the film after EB curing the topcoat – due to penetration of the electrons into the sensitive substrate, as illustrated in by the depth dose diagram for the ESI Electro Curtain machine being used for this process.¹¹ At low kV energy, electrons barely penetrate into the rigid vinyl film, as depicted in Figure 1 by the percent frontal dosage at the interface of the coating and film at a specific density of the material being irradiated.⁵ As the kV increases, the extent of electron penetration increases, as does the percent frontal dosage – as shown for two different electron beam units operating at 150 kV or 200 kV. Yellowing occurs within the wound roll due to the formation of conjugated polyene sequences due to electron degradation of the PVC (equations 1 and 2) by radical induced reaction. A “zipper elimination” occurs, in which a chromophore is formed due to polyene sequences of 5 to 25 bonds.¹² The hydrogen chloride (HCl) released from the dehydrochlorination reaction of the PVC diffuses out rapidly from the ends of the wound roll but becomes trapped in its center. This residual HCl further catalyzes PVC degradation to make the initial polyene sequences longer. The longer the polyene sequence, the more apparent the yellow color becomes to the naked eye. Thickness variation in the ink layer of the printed RVF will cause tension variation within the wound BC topcoat film roll. This gives rise to yellowing in different sections across machine direction (AMD) of the PVC topcoat film, where the maximum amount of yellowing is typically observed toward the core. A second phenomenon that causes variation in tension across machine direction is baggy film as a result of the calendaring process to produce the film that can be tight in the center but looser on the ends.

Wide-width radiation curing processes present end users with an array of material and process challenges not normally faced by narrow-web processors. This paper reviews methods to control the dehydrochlorination reaction to prevent yellowing within BC topcoat production roll:

• Reducing the depth of electron penetration into the RVF during electron beam processing, thereby reducing the degree of yellowing.
• Refrigeration of the white pattern three-wide topcoat rolls immediately after EB processing to slow down the rate of the dehydrochlorination reaction.
• Rewind EB-cured topcoated film immediately after production, using a rotograuvure press oven to remove residual HCl with high air impingement.

Experimental

The degree of yellowing can be measured by use of a calorimeter that measures tristimulus color values of “a,” “b” and “L” – where the color coordinates are designated as +a (red), -a (green), +b (yellow), -b (blue), +L (white) and -L (black). It is more appropriate to express the degree of yellowing as “Delta b,” or difference in b values between the initial and final values. A Delta b difference greater than 1 can generally be detected by the naked eye. Nine positions across the 40-inch web were measured at specific footages into the roll.

To gain a quicker understanding of this problem, an accelerated test was developed to simulate center yellowing of three-wide white topcoat film. Electron beam-cured topcoat film rolls were immediately wrapped in plastic and placed in a hot box 90°F to 125°F for a period of one to four weeks. Then, rolls would be unwound and color measurements taken at 250-ft. intervals within the roll, where the outside of the roll is identified as 0 ft. The number of AMD positions measured with a colorimeter ranged from 6 to 12.

Terminology used within this report is defined as follows:

• *Delta b width = (maximum b value across film pattern) – (minimum b value across film pattern) at specific footage into roll
• *Delta b roll = (maximum b value) – (minimum b value) for entire topcoat roll

Double-bond conversion

The samples were scanned using a Cary 620 Fourier-transform infrared (FTIR) spectrometer with a diamond attenuated total reflection (ATR) accessory with a zinc selenide (ZnSe) engine. The samples were placed directly on the sample compartment base plate of the spectrometer without a salt crystal. Analysis was performed in absorbance mode using 64 scans at 8 cm^-1 resolution. Degree of double-bond cure for the LED-cured coatings was estimated by comparing the intensity of the IR stretch for the C=C group around 1400 cm^-1 of uncured liquid (0% cure) to cured coating (100% degree of cure). This is based on the relative ratio of the measured intensity of each peak relative to carbonyl intensity at ~1700cm-1 (constant). The 1400cm^-1 stretch is absent in a fully cured coating, indicating the polymerization of the C=CH₂ in the acrylic resin. This would be 100% degree of cure. The results are presented as the IR degree of cure.

FTIR scope

Films were mounted in a microchuck, microtoning cross-section ally and mounting the cross-sections in a KBR micro-compression cell. A total of 512 scans were taken on each sample in triplicates. A total of 5 slices were prepared for each sample, representing the total thickness of the topcoat of 2 mils.

Mechanical testing

All mechanical testing was conducted using an Instru-Met instron at a rate 0.5″/min. Samples were prepared by using machined 0.5.00″ X 6.00″ template to prepare samples of coating/film composites.

Differential scanning calorimetry

Differential scanning calorimetry (DSC) experiments were conducted using a TA Instruments Model Q-2000 Differential Scanning Calorimeter. About 5.0 mg of the samples were weighed into an aluminum pan and analyzed. Initial and reheat data were obtained while heating the samples from -50°C to 190°C at a rate of 20°C/min. in a nitrogen atmosphere. The samples were quench-cooled using the RCS-90 chiller between the initial and reheat scans.

Results and discussion

To illustrate the effect of EB ionizing radiation on center yellowing, a 1000-ft. roll was prepared using white RVF film topcoated with a polyurethane acrylate coating and EB-processed at 110 kV, 4 Mrads at 100 fpm. This roll was subsequently stored at 120°F and plastic-wrapped for six days prior to color measurements. Six color measurements were recorded AMD at 750 ft. into the 40-inch-wide roll (Figure 2). In comparing the non-EB-treated film to the outside layer of film exposed to ionizing radiation (0 ft), the degree of yellowing was found to be about 0.5 Delta b units, indicating a slight effect of ionizing radiation. Color measurements taken AMD circa 750 ft. into the roll shows a near symmetrical yellowing, with a max b value around 4.6 units. The observation of a symmetrical b value profile AMD indicates that the tension within the roll is fairly uniform, thereby allowing HCl gases to readily diffuse out each end of the wound roll, leaving the highest concentration trapped in the center.

At 110 kV and 4 Mrads at 100 feet per minute (fpm), an increase in b values for AMD positions 3 through 10 is observed. The Delta b difference between the edges of position 1 to center of the row position 6 is 4.5; therefore, the degree yellowing would be easily detected on laminated tile. A Delta b value difference between the maximum b value, obtained at the edge of a tile pattern, minus the minimum b value, obtained at the edge of a tile pattern, should be no greater than 1, as this degree of yellowing can be detected by a well-trained naked eye.

For topcoat test film processed at 105 kV at 8 Mrads, plots of b values vs. position across the three-tile width of the web (AMD) for samples within the roll clearly indicates a trend of increasing center yellowing from the outside of the roll into its core (Figure 3). Two observations are apparent from this example: 1) “center yellowing,” where yellowing occurs primarily in row B (positions 4 through 8), is not apparent, but rather yellowing is observed in rows A and B, trailing off in row C. The observed nonuniform yellowing across machine direction (AMD) is presumed, due to either ink thickness variation that gives rise to differences in tension AMD and/or flatness of the film prior to wind-up. Hence, the right side of the graph depicted in Figure 3, where b values are low, reflects a loose wrap that allows for HCl to diffuse out of the roll more readily. The second observation is that yellowing of BC topcoat film does not always increase from the outside of the roll to its core. This may be due to the temperature gradient of the coated film from outside into the core of the roll or variation in tension upon rewind.

Test tile produced from wide-width topcoat film laminated to a base further shows the problem of yellowing across machine direction (Figure 4). At 120 kV and 3.3 Mrads at 100 fpm, an increase in b values for AMD positions 2 through 5 is observed. The Delta b difference between the edge of position 1 and position 4 (max value) is 4 units, and center of the row B tile is 4.5; therefore, the degree yellowing would be detected by the human eye. A Delta b value difference, between the maximum b value obtained at the edge of a tile pattern minus the minimum b value obtained at the edge of a tile pattern, should be no greater than 1 unit, as this degree of yellowing can be detected.

Reduction in depth of electron penetration into the RVF

An approach to eliminate/reduce center yellowing in wound PVC topcoat film is to minimize the effective dosage reaching the coating/PVC interface. This can be achieved by optimizing electron beam cure conditions to achieve the necessary cure at the coating both on the surface and interface, and at the same time minimize electron penetration into the PVC. Figure 1 illustrates penetration profiles at accelerating energies from 100 kV to 150 kV for the Electro Curtain EB production unit. These profiles are unique to each EB unit and are determined by the use of 9µm & 20µm dosimetry chips. The Y axis represents the percent of front dosage at a given thickness (X axis) into the desired organic substrate. A distinct feature of each kV profile is the instantaneous slope at the coating thickness of 2 mils or 58g/m2. The instantaneous slope decreases from 150 kV, where near flat at 92% front surface, to less than 5% front surface dosage for the 100 kV profile. The ramification of a smaller instantaneous slope is that the electron does not penetrate into the substrate as deeply as would an electron at higher kV, where the instantaneous slope is greater. A summary of interface dosage at 2 mil and 2.2 mil coating thickness at various accelerating energies has been calculated from these profile curves (Table 2). These tabulations are represented graphically in Figure 5 for application weights of 58g/m₂ and 65g/m₂, where interfacial dosages are higher for 100 kV vs. 105 kV until 120 kV, where the same interfacial energy is calculated. Optimum conditions to minimize electron penetration into the RVF, thus preventing dehydrochlorination, would be achieved at low interfacial dosage at low accelerating energy.

To determine the effect of kV and Mrad processing conditions on yellowing of PVC, three 1000-ft. urethane topcoated PVC rolls were processed at the same dosage but at different accelerating energies: 110 kV, 4 Mrads; 105 kV, 4 Mrads; and 100 kV, 4 Mrads. Based on penetration profiles from ESI, the film interface dosage was determined to be 0.6 Mrads (110-4), 0.24 Mrads (105-4) and 0.12 Mrads (100-4). Immediately after topcoating, these three rolls were plastic-wrapped and placed in a hot box at 120°F for 6 days for accelerated testing of center yellowing. Maximum b values obtained for each roll condition after storage are illustrated in Figure 6. Reducing the accelerating energy from 110 kV to 105 kV reduced the amount of center yellowing. The maximum Delta b value (maximum b value in roll – minimum b value in nonstorage sample) computed decreases from 3.3 at 110 kV to 1.2 at 105 kV or almost by a factor of 2.75. The decrease in interface dosage was calculated to be from 0.6 Mrads at 110 kV to 0.25 at 105 kV, a factor of 2.5. At 100 kV, only a sign of center yellowing is observed for the corresponding interface dosage of 0.12 Mrad. The maximum Delta b value is reduced to 0.31 from 3.3 at 110 kV, almost a factor of 10.

Although the data clearly indicate significantly lower yellowing of the topcoated PVC film when processed at 100 kV 4 Mrads vs. 110 kV 4 Mrads, adhesion of the coating to the film needs to be considered. At low kV – using conditions of 100 kV, 4 Mrads – adhesion of coating to the film is at 5B rating, as determined by the ASTM method D3359-17. Increasing the thickness of the coating from 58g/m2 (2mils) to 65g/m2 (2.25mils) results in poor adhesion, with adhesion test rating of 2B. This can be explained by the interfacial dosage decrease from 0.12 Mrad to 0 Mrad at the higher thickness. Although the change in thickness seems reasonable as a high limit value, the effect on adhesion is huge relative to performance of the product. Upper control limits and lower control limits must be taken into consideration when considering process conditions for manufacturing.

To investigate the difference in adhesion, the effect of IR conversion at the surface and coating film interface was determined by two techniques, IR surface only and FTIR-scope, to determine cure vs. depth. IR %C=C conversions for each kV conditions were found to be relatively the same, with no trend observed for the various kV settings (Table 4).

In the case of the FTIR-scope, a total of 5 slices were prepared for each sample, representing the total thickness of the topcoat of 2mils. In comparing IR double-bond conversions for the slice 5 (top surface) to slice 1 (bottom surface) for the 100 kV 4 Mrad condition, only a slight decrease in double-bond conversion is noted at the interface going from ca. 85% top through slices 5, 4 and 3, and decreasing to ca. 82% at interface. In contrast, for the 110 kV 4 Mrad condition, the top surface double-bond conversion is 85% and rises to 97% at the interface (Figure 7). The error bars indicate that slices 2, 3 and 4 are very similar in C=C double-bond conversion.

Effect of dose rate on roll yellowing

The dose applied to the surface of the coating is mathematically related to beam current, line speed and amperage to the filament, as defined by Equation 3. In EB processing, the speed at which dose is delivered to the surface is termed dose rate. The dose can be held constant at elevated line speeds (the energy in Mrads imparted to the sample) by adjusting the beam current. The constant K is specific for the EB machine being used and derived from dosimetry.²

Eq. 3. Dose = Beam Current * K / Line Speed

Scaling up lab processes can result in restrictions in line speeds based on capability of EB lab equipment. This can lead to unanticipated results when line speeds are increased for manufacturing, resulting in physical property changes to the substrate or cured coating that can compromise critical to quality (CTQ) attributes – such as stain resistance, cleanability and wear resistance – in flooring wear layers.

To determine the effect of dose rate on yellowing of topcoated RVF, four 2000-ft. white printed rolls were prepared at 100 kV 6 Mrads at line speeds of 100 fpm, 125 fpm, 150 fpm and 200 fpm.

These rolls were plastic-wrapped and stored at 90°F for two weeks and the degree of yellowing measured by using a Minolta Colorimeter. Six color measurements were taken across machine direction of the 40-inch web, and Delta b width values were computed at various distances into the roll (Figure 8).

Results show that line speed had very little effect on yellowing within the four 2000-ft. rolls. The Delta b width ranged from 0.1 to 0.46 units and is well within the acceptable range of minimal detection by the naked eye.

At constant dosage of 6 Mrads, the same amount of energy is delivered over a shorter time period, resulting in an increase of initiating radicals that will increase for the propagation and termination steps.11 The effect of dose rate at 100 kV 6 Mrads processing conditions for the topcoated film on double-bond conversions on the top 15µms and bottom 15µms showed a slight trend of reduced double-bond conversion from 100 fpm to 200 fpm (Figure 9). The highest conversion for the top was for the slowest line speed of 100 fpm, and lowest conversion was for the highest line speed at 175 fpm. The double-bond conversion for the bottom 15µms did not show any significant differences among different line speeds. Double-bond conversions ranged from 84% to 86%.

The effect of dose rate on elongation properties of the coating film composites also was investigated for EB processing conditions of 100 kV 6 Mrads at line speeds of 100, 125, 150 and 200 fpm. In general, elongation slightly increased with increasing line speed at constant dose from 100 fpm to 200 fpm, where maximum elongation was observed at 175 fpm (Figure 10). This observed difference was presumed to be due to a decrease in the Tg of the urethane acrylate coating, as well as slight degradation of the PVC film.^10-11

The effect of dose rate on Tg of the coating film composites was briefly investigated for EB processing conditions of 107 kV 4.5 Mrads at line speeds of 100, 150, 175 and 200 fpm (Figure 11). Increasing the line speed from 100 fpm to 200 fpm, doubling dose rate, resulted in a slight reduction in Tg of the coating (Table 5), indicating lower molecular weight (MW) polymer backbone is being formed. The Tg decreased from 100 fpm at 15.3°C to 13.5°C at 150 fpm and 13.2°C at 175 fpm. At 200 fpm the Tg increased slightly to 14°C. More pronounced effects of dose rate have been observed for model compounds of phenyl acrylate, benzyl acrylate and 2-phenyl ethylacrylate by Schissel.¹¹

Effect of Refrigeration

The effect of post-refrigeration of EB-cured topcoat PVC film was investigated to reduce the rate of the dehydrochlorination reaction occurring in the roll to prevent the formation of polyene sequences. The rate constants for dehydrochlorination, HCl evolution, has been determined to be equal to the rate constants for polyene formation. Initial studies indicate that, for one month of storage, the refrigeration temperature of BC topcoat film would need to be below 50°F. The effect of refrigeration temperature on yellowing after one month of storage is readily observed by comparing Delta b width values at 250-ft. intervals within each 1000 ft. of BC topcoat, 110 kV 4 Mrads processed roll (Table 5). On average, Delta b width values for the 60°F refrigerated roll are about twice that of the 50°F processed roll. In contrast, the control roll stored at 75°F for one month shows significant yellowing past 250 ft. into the roll. The Delta b width value increases from 0.26 at 250 ft. to 1.53 at 500 ft. into the 1000-ft. roll. The dehydrochlorination reaction is dependent on temperature and increases with increasing temperature.¹⁰

To further evaluate the effect of storage temperature on in-roll yellowing, a refrigeration study was conducted on two 9000-ft. rolls to determine what cold storage temperature range would be required to minimize yellowing. These topcoat rolls were initially processed at 110 kV 4 Mrads and individually stored for a period of one month at 40°F and 50°F, respectively. At the end of the storage period, samples were retrieved from the rolls every 500 ft. to determine the extent of yellowing (Table 3). A plot of Delta b width vs. footage within each refrigerated roll is illustrated in Figure 12. In general, the Delta b width values for the 40°F roll are significantly less than those values obtained for the 50°F roll up to 7000 ft. Yellowing develops much sooner in the 50°F roll – beginning to escalate from 3000 ft. to 4500 ft., to a Delta b width value of 0.52 – and then decreasing slightly and leveling off. Whereas in the 40°F refrigerated roll, yellowing is acceptable up to 7000 ft., and then takes off and escalates up to 8500 ft. to a Delta b width of 0.63. In comparing high b values obtained for each footage in the roll for each roll, there is a trend toward higher values for the 50°F vs. 40°F roll. The Delta b film roll values from outside wrap to roll core for the 50°F and 40°F rolls were found to be 0.77 and 0.92, respectively, which would be considered acceptable.

First order rates constants, activation enthalpies and entropies for the dehydrochlorination reaction at different temperatures have been measured under nitrogen atmosphere by Fisch and Bacalouglu, where the “rate constants of initiation K1 are found to be equal to the rate of polyene sequence formation.” If assumed that the rate of polyene sequence formation due to dehydrochlorination is directly proportional to the Max b color value within the roll at various footages, a plot of High b value vs. footage is illustrated in Figure 13 for the portion of Figure 13 that shows the most constant change over time until a maximum is achieved. It appears that, for low concentrations of polyene formation, polyene concentrations increase linearly vs. footage within the roll, as published for PVC for polyene sequences vs. time.10

To evaluate the effect of refrigeration on yellowing for laminated film over a longer period of time, a second 9000-ft roll was processed at 110 kV 4 Mrads and stored for one month at 40°F, one month at 50°F, two weeks at 40°F and one week at 60°F. This roll was then laminated to tile base, where the pattern produced three tile images across machine direction. The laminated tile was found to have an acceptable degree of yellowing, based on Delta b tile edge width values taken at the beginning of the film and at 500-ft. intervals to the end of the film, at 9000 ft. (Figure 14). The Delta b width values recorded are below 1 Delta b unit, where the average was determined to be 0.3162 Delta b units (std dev 01346), thereby indicating refrigeration temperatures in the range of 40°F to 50°F would be suitable for storage of topcoat PVC film rolls.

Effect of rewinding the film immediately after EB processing

To determine the effect of warm air impingement on yellowing within topcoated PVC film immediately after processing, films were rewound through a nine-dryer ink press. Initial studies demonstrated that a reduction in yellowing is observed by removal of residual HCl (Figure 16). There are two aspects that explain our observations. First, the warm air impingement on the ink side of the PVC film is presumed to allow the film to reach a “pre-glassy” state, thereby allowing HCl to diffuse out of the PVC more rapidly. Second, the high velocity air (2000 fpm) impingement on the PVC film is presumed to remove HCl at the immediate surface, thereby lowering the concentration gradient of HCl within the PVC, hence increasing the permeation rate of HCl out of the film.

Post heat-treatment using dryer stations

Using a dryer at 180°F at 100 fpm, two different cure conditions were used for this study: 110 kV 4 Mrads and 107 kV 4.5 Mrads. Again, treated and nontreated rolls were plastic-wrapped and placed in a hot box at 125°F for one week for simulated yellowing. A significant decrease in center yellowing is observed for post heat-treated rolls (Figure 15). In comparing nonpost heat-treated to heat-treated for the 110 kV 4 Mrad condition, the Delta b width maximum value decreased 3.52 to 0.58 for this processed film. Reducing the accelerating energy to 107 kV resulted in less yellowing. The maximum Delta b width values for treated and nontreated rolls are 0.33 and 2.39, respectively.

Conclusions

Wide-web processing of EB-cured topcoated PVC film poses several challenges to the end user that can be overcome by the following:

• Choose proper EB process parameters to reduce the interfacial dosage to a minimum without causing other critical-to-quality issues, such as adhesion of coating to film. FTIR-scope can be used as a guiding tool to manage process condition to optimize the effect of kV on double-bond C=C conversion. For studies presented, minimal effect of dose rate on percent elongation and Young’s modulus were observed in the process range at 100 kV at 4.5 Mrads from 100 fpm to 200 fpm. Similarly, the Tg of the topcoat urethane acrylate was found to decrease slightly with increasing line speeds from 100 fpm to 175 fpm.
• Using refrigeration to cool the roll after EB processing proved to be a good method of choice at the expense of costs to use refrigeration post-production and transportation of coated film. The Delta b max within a roll was reduced from 4.5 for topcoated PVC nonrefrigerated to 1.79 after 40°F refrigeration for one month. At low concentrations of polyene formation, polyene concentrations increase linearly vs. footage.
• The effect of dryer treatment also proved to be useful in reducing in roll yellowing but can be costly from a productivity standpoint of running the film through a dryer post EB cure.

Acknowledgements

This work was performed in collaboration with co-workers Cliff Rosenau, Craig DeSantis and Eric Chambers for factory testing and Dick Papez for testing and analysis.

References

1. Biro, A. and Bishop, J. Advances in Electron Beam Curing in Wide Web Flexible Package Printing. PCI, 2016.
2. Lapin, S. EB Curing Q&A. UV&EB Technology, Vol. 4, No. 1, 2018.
3. Lapin, S. Modification of Polymer Substrates using Electron Beam Induced Graph Copolymerization. RadTech 2014 Proceedings.
4. Carignano, A. Low Energy Electron Beam Curing for Piezo Inkjet. PCI, June 30, 2016.
5. Rangwalla, I. Development in Low Voltage EB Curing for High Product Throughput Applications. RadTech 2014 Proceedings.
6. Drobny, J.G. Ionizing Radiation and Polymers: Principles, Technology, and Application. Plastics Design Library/Elsevier Publishing, 2013.
7. Atchison, G.J. Color and Radical Formation in Irradiated Polyvinyl Chloride. J.Polym. Sci., 49, p385 1961.
8. Mark, H.F. Radiation Chemistry of Polymers, Encyclopedia of Polymer Science and Technology, Concise, p1011. John Wiley and Sons, New York, NY, 2013.
9. Sigel. G., Eshbach, J. and Bagley, G. US Patent 6616792 B2.
10. Dawes, K. and Glover, L.C. Effects of Electron Beam and Gamma Irradiation on Polymeric Materials, Physical Properties of Polymers Handbook, Chapter 41, pp134-138K, Mark, J.E., ed. AIP Press.
11. Schissel, S., Lapin, S. and Jesop, J. Accelerating EB’s Potential: Understanding the Effects of Dose Rate in Electron-beam Polymerization, RadTech 2016 Proceedings.
12. Fisch, M. and Bacaloglu, R. Kinetics and Mechanism of the Thermal Degradation of Poly(Vinyl Chloride), J. of Vinyl & Additive Technology, Vol. 1, No. 4, December 1995.

Gary A. Sigel, Ph.D., is senior principal scientist at Armstrong Flooring. He can be reached at [email protected] or 717.672.7908.