2.2.1. Materials

Poly(L-lactide-co-ε-caprolactone) (PLCL, 70/30) and poly(L-lactide-co-D,L-Lactide) (PLDLA 70/30) were commercially purchased from RESOMER® (Evonik Health Care Evonik Industries AG, Essen, Germany). Chloroform (CHCl3) was selected as a solvent and was supplied by Showa, Tokyo, Japan. All other reagents and solvents were of analytical grade and were used without further purification.

#### 2.2.2. Sample Irradiation

PLCL and PLDLA films were prepared by a solution casting method using CHCl3 as a solvent. The PLCL and PLDLA powders were dissolved in CHCl3 to obtain 10 wt% and 4 wt% polymer solutions, respectively. The PLCL and PLDLA films were fabricated by pouring a polymer solution into a well-cleaned glass plate and evaporating the solvent in air at room temperature. Dried films were peeled off manually from the glass plate and dried in vacuum oven for 24 h at room temperature. Prior to gamma irradiation, the prepared films were packed with nitrogen gas in glass vials. Gamma irradiation was thereafter performed using a gamma 60Co source on the samples with different radiation

doses (25, 50, 100, 200, and 500 kGy) at a dose rate of 10 kGy/h. The 60Co source (MDS Nordion, Ottawa, Canada, IR 221 n wet storage type C-188) was located at the Korea Atomic Energy Research Institute (KAERI), Jeongeup, Republic of Korea.

#### 2.2.3. Attenuated Total Reflection Fourier Transform Infrared Spectroscopy (ATR-FTIR)

ATR spectra of gamma-ray-irradiated biodegradable polymer films were obtained using an ATR-FTIR spectrophotometer (Bruker TENSOR 37, Bruker Corporation, Billerica, MA, USA). The spectra were measured in the wavenumber range from 500 to 4000 cm<sup>−</sup><sup>1</sup> in ATR mode. Spectra were recorded using Bruker OPUS software (version 8.5, Bruker Corporation, Billerica, MA, USA) at a resolution of 4 cm<sup>−</sup>1.

#### 2.2.4. Thermal Analysis

The thermal properties of the polymer films were tested by difference scanning calorimetry (DSC, Q100, TA Instruments, New Castle, DE, USA). The DSC thermograms of the polymer films were measured from −50 to 250 ◦C under a nitrogen atmosphere at a heating rate of 5 ◦C/min, and then were cooled at room temperature. A second heating cycle was then observed by heating the samples from −50 to 250 ◦C.

#### 2.2.5. Average Molecular Weight

Changes in the molecular weights of the irradiated film were determined by gel permeation chromatography (GPC, PL-GPC 110, Polymer Laboratories, Church Stretton, UK) with the eluent of CHCl3 at a flow rate of 1.0 mL/min at 40 ◦C. The GPC system was equipped with columns of PLgel Guard column 5 μm, PLgel 10 μm Mixed B and PLgel 5 μm 10,000 A (Polymer Laboratories, Church Stretton, UK) calibrated with polystyrene standards.

The changes in molecular weight are related to the radiation chemical yields of crosslinking ( *G*x) and chain scission ( *G*s), which determines the extent of chain scission or crosslinking during gamma-ray irradiation, and can be calculated from the following equations [27]:

$$1/M\_{\rm W} = 1/M\_{\rm W,0} + (G\_{\rm s}/2 - 2G\_{\rm x})D \times 1.038 \times 10^{-6} \tag{8}$$

$$1/M\_{\rm n} = 1/M\_{\rm n,0} + (G\_{\rm 5} - G\_{\rm x})D \times 1.038 \times 10^{-6} \tag{9}$$

where *M*w,0 and *M*n,0 are the weight and number average molecular weight of unirradiated films. *M* w and *M*n are the corresponding values following exposure to irradiation dose, *D*. A ratio of *G*s/*G*x greater than 4 would indicate that chain scission is more prominent for nitrogen atmospheres [29].

#### **3. Results and Discussion**

#### *3.1. Modeling and Simulation*

Figure 2 shows the number average molecular weight change according to the gamma irradiation dose of the radiation damage model PLCL and PLDLA films through Geant4 simulation. The blue triangle represents the simulation result considering both chain scission and crosslinking in the radiation damage model, whereas the red square represents the simulation result considering only the chain scission in the radiation damage model. The *M*n,0 and *G*-values ( *G*s and *G*x) of PLCL and PLDLA obtained from the experimental results were used in Equation (7) for the radiation damage model of the polymer considering the radiation cleavage and crosslinking derived in this study, respectively.

**Figure 2.** Number average molecular weight of (**a**) PLCL and (**b**) PLDLA radiation damage models depending on the gamma-ray irradiation in simulation (blue triangles: simulation of chain scission and crosslinking; red squares: simulation of chain scission).

In the simulation results, the average molecular weight of the radiation damage model PLCL decreased with the increasing irradiation dose (Figure 2a). In the two damage models, the difference in the number average molecular weight in the range 25–200 kGy of the irradiation dose appeared to gradually widen, but the gap seemed to narrow at the irradiation dose of 500 kGy; it appeared that both the chain cleavage and crosslinking reactions occurred up to the range before the irradiation dose of 200 kGy, but the chain cleavage reaction was expected to be relatively dominant at a high irradiation dose of 500 kGy or more.

Figure 2b shows the number average molecular weight changes, according to the gamma radiation dose of the radiation damage model PLDLA film, through the Geant4 simulation. In the simulation results, the number average molecular weight of the radiation damage model PLDLA decreased with the radiation dose. Interestingly, when only chain cleavage was considered in the radiation damage model, there was no significant difference from when both chain cleavage and crosslinking were considered, that is, in the PLDLA damage model, it can be inferred that the chain cleavage reaction prevails over crosslinking.

#### *3.2. Radiation Assessment for Simulation Verification*

#### 3.2.1. ATR-FTIR Spectroscopy

Figure 3 shows the ATR-FTIR spectra of PLCL and PLDLA before and after gammaray irradiation. The ATR-FTIR spectrum of the PLCL film was observed from 1300 to 1000 cm<sup>−</sup>1, which is related to the stretching of C−O bonds in the ester, found in the long alkyl chain of the polymer structure [30,31]. The C−O and C−O−C groups exhibited stretching peaks at 1188, 1037, and 1079 cm<sup>−</sup><sup>1</sup> for the PLCL film both before and after exposure. The stretching of −C=O (carbonyl) appeared as an intense peak at 1750 cm<sup>−</sup>1. In addition, the asymmetric stretching vibration of the −CH3 and −CH2 groups were observed at 2992 and 2943 cm<sup>−</sup><sup>1</sup> for the PLCL film before and after exposure [29,31,32]. Another bond related to the symmetric vibration of −CH2 was observed at 2872 cm<sup>−</sup><sup>1</sup> [29]. Bonds associated with the asymmetrical and symmetrical stretching of the −CH3 group exhibited peaks at 1450 and 1361 cm<sup>−</sup>1, respectively [33]. Small peaks were also observed for the C−H bending vibrations at 757 and 864 cm<sup>−</sup><sup>1</sup> [34].

**Figure 3.** ATR-FTIR spectra before and after gamma-ray irradiation: (**a**) PLCL and (**b**) PLDLA polymer films.

Figure 3b shows the ATR-FTIR spectra of PLDLA before and after gamma-ray irradiation. The C−O and C−O−C groups exhibited stretching peaks at 1182 and 1081 cm<sup>−</sup><sup>1</sup> for the PLDLA film both before and after exposure. The stretching of −C=O (carbonyl) appeared as an intense peak at 1746 cm<sup>−</sup>1. In addition, the asymmetric stretching of −CH was observed at 2994 and 2945 cm<sup>−</sup><sup>1</sup> for the PLDLA film before and after exposure. Another bond related to the bending vibration of −CH3 was observed at 1452 cm<sup>−</sup><sup>1</sup> [33]. Small peaks were also observed for the C−H bending vibrations at 759 and 872 cm<sup>−</sup><sup>1</sup> [34].

The PLCL and PLDLA films did not display significant differences before and after exposure due to minor radiation-induced chemical changes occurring in the polymer chain [35,36]. This implies that there was no change in the functional group inside the polymer after irradiation, and this result indicated that the possibility of the creation of a new bond, which was one of the parts to be considered in the decomposition mechanism of the polymer, could be excluded.

#### 3.2.2. Thermal Analysis

Figure 4 shows the DSC curves of PLCL and PLDLA before and after gamma-ray irradiation. The DSC curve of the PLCL film before exposure had a melting point (*T*m) of 159.78 ◦C. No significant differences were observed up to a dose of 100 kGy. However, *T*m decreased rapidly when the dose was increased to 200 and 500 kGy, indicating that gamma-ray irradiation resulted in the degradation and scissioning of the main chain [20,37].

**Figure 4.** DSC thermograms of the before and after gamma-ray irradiation (**a**) PLCL and (**b**) PLDLA polymer films.

In the DSC curve of the PLDLA film before irradiation, a glass transition (*T*g) was observed at 41.75 ◦C. After gamma irradiation, there was no significant change at 100 kGy or less, and a decrease in *T*g was observed at 200 kGy. The *T*g of a polymer was related

to its molecular weight, and *T*g decreases as the average molecular weight decreases [38]. Consequently, it was confirmed that the temperature at which thermal transition occurs changed when the dose was greater than 100 kGy.

#### 3.2.3. Average Molecular Weight

The changes in the average molecular weights (*M*w and *M*n) of the PLCL and PLDLA films before and after gamma-ray irradiation are shown in Figure 5. The oxidation of polymers during exposure reduces crosslinking, increases degradation, or causes chain scissions [24]. Thus, the formation of radicals after exposure results in chain scissions, which in turn lowers the molecular weight. Chain scissions usually occur when polymers are in the amorphous phase [25].

**Figure 5.** Average molecular weight (black squares: *M*w, white circles: *M*n) of (**a**) PLCL and (**b**) PLDLA depending on the gamma-ray irradiation.

At higher doses of at least 200 kGy, chain scission occurs because of the alkyl free radicals that react with oxygen to form peroxyl free radicals through hydrogen abstraction [26,27]. This type of chain scission has no significant effect on the decrease in the average molecular weight, given the relatively higher increase in chain scission, compared to crosslinking events under higher doses [39]. The number of alkyl free radicals was greater than that of peroxyl free radicals under higher radiation doses because oxygen was not present under our experimental conditions. Alkyl free radicals had less influence on chain scissions than peroxyl free radicals, and were more likely to undergo rebonding or crosslinking in crystalline and amorphous segments [40].

Hydrogen abstraction due to chain-breaking radicals in the weakest C–H bonds contributes to an increase in alkyl radicals or –C(CH3)– radicals. The average molecular weight decreased with the degradation of the polymer during exposure [41]. From Table 2, the high *G*-value (*G*s/*G*x = 8.7) exhibits a degree of chain scission in the irradiated PLCL film after exposure. The decrease in molecular weight at low radiation doses was due to chain scission by alkyl free radicals [26]. High-energy irradiation forms radicals played a role in the degradation of polymers [41]. In other words, the chain appears to be affected at a dose of 30 kGy or higher, but exhibits a more distinct difference in molecular weight at higher doses of at least 200 kGy. Even in the absence of oxygen in the air, radicals that cause chain cleavage may be formed by radiation energy, which may affect the average molecular weight of the polymer.

**Table 2.** Chain scission (*G*s) and crosslinking (*G*x) radiation yields of PLCL and PLDLA irradiated.


Polymers with oxygen atoms are known to exhibit a very high sensitivity to radiation [2], and similar studies were conducted on structures containing oxygen in the polymer backbone [29]. Methine groups in the polymer backbone appear to be important in the radiolysis of biodegradable polymers containing oxygen atoms. Previously, Nugroho et al. reported a study on the gamma-ray degradation of PLA [42]. This polymer contains an ester linkage and a cleavage site in the methine group. The cleavage of the ester bond at this cleavage site causes crosslinking with a relatively low yield, whereas cleavage at the methine group causes the chain scission of the polymer [17,42].

In general, the radicals generated at the ends of polymer chains generate new radicals, move to adjacent polymer chains, or cause hydrogen abstraction [42]. In addition, double bonds are formed at the chain ends after hydrogen abstraction. In PLDLA, a copolymer comprising of isomers of PLA, and both crosslinking reactions, can cause decomposition reactions due to cleavage of ester bonds in the polymer and hydrogen abstraction of the methine groups [20]. In this study, when interpreted only as a result of the decrease in the average molecular weight of PLDLA according to gamma radiation dose, the radicals generated inside the polymer react well with each other and the decomposition reaction is relatively dominant; therefore, it is believed that primarily chain cleavage occurs. Consequently, it is observed that the polymer chain is affected by radiation even at a low gamma irradiation of 25 kGy or higher, but it can be confirmed that a distinct change is exhibited with a molecular weight at 50 kGy or higher.

#### 3.2.4. Modeling and Simulation Verification

Figure 6 shows the results of the simulation of the radiation damage model and the change in the number average molecular weight of the PLCL and PLDLA films according to the gamma radiation dose. The blue triangle represents the simulation result considering both chain scission and crosslinking in the radiation damage model, the red square represents the simulation result considering only the chain scission in the radiation damage model, and the white circle represents the experimental result obtained from the radiation measurement evaluation.

**Figure 6.** Comparison of number average molecular weight of (**a**) PLCL and (**b**) PLDLA in simulation and measurement evaluation (blue triangles: simulation of chain scission and crosslinking, red squares: simulation of chain scission, white circles: experimental value).

The simulation results predicted that, in the radiation damage model, both chain cleavage and crosslinking reactions occurred at low doses before 200 kGy, but chain cleavage reactions were relatively dominant at high doses of 500 kGy or higher (Figure 6a). The number average molecular weight of the PLCL film decreased according to the gamma radiation dose in the radiation measurement results, which exhibited a similar trend to the Geant4 simulation results. The actual evaluation results were similar to those of the radiation damage model when both chain scission and crosslinking were considered at a radiation dose in the range 0–200 kGy. Meanwhile, the actual evaluation results at a dose of 500 kGy were similar to the simulation results of the radiation damage model considering

only chain scission. Consequently, it can be determined that PLCL has a more dominant chain cleavage reaction than crosslinking at a high radiation dose of at least 500 kGy, and the simulation of the radiation damage model can be verified through a comparison with the experimental results.

The simulation results of PLDLA exhibited no significant difference when only chain cleavage was considered in the radiation damage model, compared to when both chain cleavage and crosslinking were considered (Figure 6b). Therefore, the PLDLA damage model predicted that the chain cleavage reaction would prevail at the total irradiation dose (25–500 kGy). The results of the radiation measurement demonstrated that the polymer chain was affected by radiation even at a low irradiation dose of 25 kGy or higher, and that the chain cleavage reaction was dominant, with a distinct molecular weight change at 50 kGy or higher. Consequently, unlike PLCL in which cleavage was predominant at high doses, in the case of PLDLA, it could be determined that the chain cleavage reaction was dominant at the total irradiation dose (25–500 kGy), and the simulation could be verified through a comparison with experimental results.

In addition, in Equation (7) derived from this study, the slopes *α* and *β* are correlated with the G value, and thus the radiation damage model is an important factor in simulating the actual evaluation experiment. To reflect the change in the initial number of molecules (*N*0) due to molecular formation through chain cleavage and crosslinking, 0.5 and 2 were applied to the slopes *α* and *β*, respectively. The average molecular weight change curve was also determined.

Regarding the degradation of biodegradable polymers, as well as the results of our study, the results of modeling and simulation studies by other researchers were also reported [43–45]. This work focused on the average molecular weight required to conduct a basic simulation of radiation damage analysis for biodegradable polymers. However, the analysis reported in this paper corresponds to a simplified model of the interaction of ionizing radiation through only single ionization. It was not fully considered that, as a result of radiolysis, apart from molecular hydrogen, molecular oxygen was also released [46]. Thus, it is necessary to carry out additional research considering post-radiation, oxidative degradation processes.
