**1. Introduction**

Having its beginnings in the 1950s, electron beam (e-beam) industrial processing has been developing rapidly and used extensively in many areas of industry, where there is a need for enhancement of the physical and chemical properties of materials or a reduction of undesirable contaminants [1]. Low-energy electron beams, which are typically of energies ranging between 80 and 300 keV [2], have found several applications in the radiation processing industry, starting from polymer curing [3], cross-linking of coatings and inks [4] and polymer functionalization by grafting [5], through decontamination of packaging materials [1], up to sterilization of surfaces [6,7] and pharmaceutical components [8]. Low-energy e-beam accelerators offer a number of advantages, such as (1) electrical energy conversion of the order of 70% [9], (2) high dose rates (~100 kGy/s), allowing rapid processing [1], (3) radiation protection in a form of self-shielded equipment [1], (4) environmental benefits, as they reduce the emission of volatile organic compounds and air pollutants [10], and (5) low costs [1].

However, low-energy electrons are able to penetrate matter only up to a few hundred micrometers [2], which in fact becomes challenging in terms of absorbed dose determination. The dosimetry requirements are especially rigorous when performing the surface

**Citation:** Skowyra, M.M.; Ankjærgaard, C.; Yu, L.; Lindvold, L.R.; Skov, A.L.; Miller, A. Characterization of a Radiofluorogenic Polymer for Low-Energy Electron Beam Penetration Depth Visualization. *Polymers* **2022**, *14*, 1015. https:// doi.org/10.3390/polym14051015

Academic Editors: Bozena Jarz ˛ ˙ abek and Mohammad Afsar Uddin

Received: 28 November 2021 Accepted: 28 February 2022 Published: 3 March 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

sterilization process of, for example, pharmaceutical or medical products [11]. There, only the external surfaces of, for example, pre-sterilized tubs (of vials or syringes) should be irradiated before they can enter an aseptic filling area. In this way, the production of radiolysis products within the volume of a tub is avoided [12].

Several dosimetry systems are used in the radiation processing industry, such as alanine-EPR [13], calorimeters [14] or radiochromic dye films [13,15]. They are all excellent tools for high-dose measurements (~kGy). However, at low electron energies, dose gradients are introduced over the thickness of a dosimeter, and the measured average doses will depend on the chosen system [16]. In addition, they are not able to measure the dose in the first microns of irradiated material, the so-called surface dose. Thus far, the surface dose has been extrapolated from obtained depth–dose curves and corrected to the average dose in the first micrometer—Dμ [16]. Schuster et al. [17] proposed a thin dosimetric spray-coating material for surface dose measurements of complex surface geometries (exemplified by PET bottles). The concept relies on a readout of the dose-dependent luminescence decay time of an inorganic phosphor under near-infrared excitation. However, fading of the phosphor was a limiting factor that this dosimetry system so far has not been able to overcome, and hence, the challenge of directly measuring the dose in the first micrometer of a material is still to be addressed.

Here, we investigate a new type of material that could be applied as a low-energy electron beam dosimeter, focusing on surface dose verification. As a starting point, a radiochromic and radiofluorogenic polymer dosimeter developed by Bernal-Zamorano et al. [18–20], which was initially intended to work as a dosimeter for the medical dosimetry dose range (~Gy), was selected. However, industrial radiation processing applies doses of the order of tens or even hundreds of kGy, so the material had to be appropriately modified. Furthermore, the material's composition, its mechanical, optical and thermal properties, hydrophilicity as well as the photocuring process had not yet been fully optimized or characterized. The system previously developed by Skowyra et al. [21] was a proof of concept where the solid state polymer material responded to kilogray radiation doses of low-energy electron beam irradiations by its change in fluorescence intensity, measured using an epi-fluorescence widefield microscope. However, issues related to the background fluorescence signal, shelf life and environmental stability of the material prompted further investigation into establishing and demonstrating suitable material requirements before optimizing the readout technique into confocal laser scanning microscopy.

In this paper, we investigate the influence of the polymer material's composition on the properties of the material. This includes the polymer matrix and the optimal solvent for the radiation-sensitive dye. We present the material's properties in terms of its thermomechanical behavior, particularly the decomposition process and glass transition temperature, as well as hardness and rheological studies. Additionally, the optical quality of the polymer, with a focus on refractive index measurement, was determined, and for fluorescence imaging purposes, the water affinity of the material was evaluated. Finally, we show the absorbance and fluorescence response spectra of pararosaniline leuco-dye in both a solution and solid material, and demonstrate the spectral response of the material irradiated with a low-energy electron beam.

#### **2. Materials and Methods**

#### *2.1. Composition*

The dosimeter material consisted of (1) host polymer–PPGDA (poly(propylene glycol) diacrylate, 1.12 M, Sigma-Aldrich, Darmstadt, Germany, CAS: 52496-08-9, inhibitors: 100 ppm BHT and 100 ppm MEHQ); (2) photoinitiator–TPO (diphenyl (2,4,6-trimethylbenzoyl) phosphine oxide, 28 mM, Sigma-Aldrich, Germany, CAS: 75980-60-8); (3) leuco-dye (4 ,4 ,4 triamino-triphenylaceto-nitrile, 6.89 mM, NCK A/S, Farum, Denmark); and (4) acetonitrile (0.99 M, Sigma-Aldrich, Germany, CAS: 75-05-8, max. 0.005% H2O).

The solvents tested for solubility were acetone (CAS: 67-64-1, ≥99.5%), acetonitrile (CAS: 75-05-8) and ethanol (CAS number: 64-17-5, ≥99.9%) supplied by Sigma-Aldrich (Merck, Germany) and 2-ethoxy ethanol (CAS: 110-80-5, ≥98.5%) provided by Fluka Chemie (Buchs, Switzerland). The concentration of the dye solution was 20 mg/mL.

#### *2.2. Fabrication*

The leuco-dye, the selected solvent (acetonitrile) and the host polymer were mixed at room temperature using a magnetic stirrer in a closed, UV-blocking glass vial for 30 min. After that, the photoinitiator was added, and the solution was mixed for another half an hour. The mixed composition was then placed in a refrigerator at 4 ◦C until the photocuring was carried out to lower the oxygen diffusion and avoid a viscosity increase [22]. The samples were prepared by means of a photopolymerization reaction using a LED lamp of a 385 nm wavelength as described in [21]. Three different sample types were manufactured: (1) cuboids of dimensions 3 × 3 × 45 mm, (2) cuboids of dimensions 1 × 3 × 45 mm and (3) cylinders 10 mm in diameter and 3 mm in thickness. The samples were cured in silicone molds [21] of corresponding shapes with a plastic foil placed on the top surface to block oxygen inhibition during the reaction [23]. The illumination time was 5 min, with the intensity of the UV light measured to be 15.0 ± 0.3 mW/cm2. Unless otherwise stated, the uncertainties throughout the performed measurements were given at one standard deviation (*k* = 1).

Two resins for 3D-printed molds were tested as alternatives for the silicone mold: (1) traditional resin (black, ELEGOO standard [24]) and (2) acrylonitrile butadiene styrene (ABS)-like resin (transparent, ELEGOO ABS-like [25]). The compositions of the resins were: epoxy resin (50%), hexamethylene diacrylate (40%) monomer, N-(dimethylcarbamoyl) glycine (5%) and hydroxycyclohexyl phenyl ketone (5%) photoinitiator. The molds were 3D printed using stereolithography (SLA) technology.
