Evaluation of radiation stability of electron beam irradiated Nafion® and sulfonated poly(ether ether ketone) membranes

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ Evaluation of radiation stability of electron beam irradiated Nafion® and sulfonated poly(ether ether ketone) membranes © 2022 The Authors. Published by Elsevier Ltd. Published version Pajuste, E.; Reinholds, I.; Vaivars, G.; Antuzevičs, A.; Avotiņa, L.; Sprūģis, E.; Rossi, Mikko; Kettunen, Heikki; Meri, R.M.; Kaparkalējs, R. Pajuste, E., Reinholds, I., Vaivars, G., Antuzevičs, A., Avotiņa, L., Sprūģis, E., Rossi, M., Kettunen, H., Meri, R.M., & Kaparkalējs, R. (2022). Evaluation of radiation stability of electron beam irradiated Nafion® and sulfonated poly(ether ether ketone) membranes. Polymer Degradation and Stability, 200, Article 109970. https://doi.org/10.1016/j.polymdegradstab.2022.109970 2022 Polymer Degradation and Stability 200 (2022) 109970 Contents lists available at ScienceDirect Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polymdegradstab Evaluation of radiation stability of electron beam irradiated Nafion® and sulfonated poly(ether ether ketone) membranes E. Pajuste a , b , ∗, I. Reinholds c , g , G. Vaivars a , d , A. Antuzevi ˇ cs d , L. Avoti ¸n a a , E. Spr ¯ u ģ is a , d , R. Mikko e , K. Heikki e , R.M. Meri f , R. Kaparkal ¯ ejs a , b a Institute of Chemical Physics, University of Latvia, Jelgavas iela 1, Riga, Latvia b Faculty of Chemistry, University of Latvia, Jelgavas iela 1, Riga, Latvia c Baltic Scientific Instruments, Ganibu dambis 26, Riga, Latvia d Institute of Solid State Physics, University of Latvia, Kengaraga iela 8, Riga, Latvia e Department of Physics, University of Jyväskylä, Survontie 9 C, Jyväskylä, Finland f Institute of Polymer Materials, Faculty of Materials Science and Applied Chemistry, Riga Technical University, Riga, Latvia g Institute of Food Safety, Animal Health and Environment “BIOR”, Lejupes iela 3, Riga, Latvia a r t i c l e i n f o Article history: Received 15 December 2021 Revised 7 April 2022 Accepted 5 May 2022 Available online 8 May 2022 Keywords: Proton exchange membranes Electron beam Ionising radiation Degradation Crosslinking Thermomechanical properties a b s t r a c t Proton exchange membranes (PEM), which have been commonly used in fuel cells have raised interest for the application in harsh environments involving ionizing radiation. Therefore, radiation stability and ability to sustain their functionality under the radiation environment are of great interest. Within this study, electron beam irradiation in dose range from 50 to 500kGy was used to evaluate the effects of radiation on the physico-chemical and mechanical properties of two types of PEM: commercial Nafion®117 and sulfonated poly(ether-ether-ketone) (SPEEK) with high degree of sulfonation (DS = 0.75 ±0.5). SPEEK membrane presented higher mechanical and thermal stability compared to that of Nafion® at doses up to 250 kGy, which was evidenced by infrared and electron paramagnetic resonance spectroscopy, thermal analysis, ion chromatography methods. Tensile tests at room temperature and dynamical mechanical analysis of irradiated membranes revealed improved strength, storage modulus at room and elevated temperatures (80 °C) for irradiated SPEEK as compared to pristine PEM. For comparison Nafion® exhibited notable deterioration of mechanical properties including elongation at break due to the predominant oxidation and chain scission already at doses exceeding 50 kGy. The study indicated that SPEEK could be perspective replacement of traditional PEM for application in fuel cells exposed to ionising radiation. ©2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ) 1. Introduction Nafion®, the sulfonated tetrafluoroethylene-based fluoropolymer–copolymer, first synthesised by the DuPoint TM is a standard membrane material used for polymer electrolyte membrane (PEM) applications in fuel cells (FC), electrolysis cells, batteries and other electrochemical devices [1] . The polymer consists of hydrophilic perfluoro sulfonic acid moieties attached to a hydrophobic fluorocarbon backbone ( Fig. 1a ) and possess high conductivity not only for hydrogen protons, but also for various kinds of cations due to the presence of negatively charged sulfonate groups [2] . However, broader application of Nafion® have ∗Corresponding author. E-mail address: elina.pa[email protected] (E. Pajuste) . several drawbacks such as high price, limited selectivity, permitted leakage between anode and cathode electrolyte compartments. All of this leads to the enhanced search of alternative PEM materials. In recent decades, SPEEK polymers, the sulfonated polyether-ether ketone membranes ( Fig. 1b ) have been evaluated as alternative replacement to conventional PEM materials due to relatively lower costs and advanced properties including adjustable thermal and mechanical performance depending on the degree of sulfonation (DS) of the crystalline PEEK polymer under the effect of sulfonic acid under controlled reaction conditions, whereas the proton conductivity may be slightly lower or comparable with that of commercial PEM Nafion [3–6] . PEM based technologies are being considered for the use in industries related to harsh environments such as nuclear facilities, where fuel and electrolysis cells have found an application https://doi.org/10.1016/j.polymdegradstab.2022.109970 0141-3910/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ) E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Fig. 1. The general chemical structures of Nafion® (a) and SPEEK (b) polymers. for separating and recovering heavy and radioactive hydrogen isotopes [ 7 , 8 ]. Electrolytic processing of water in order to decontaminate it from radioactive isotope of hydrogen, tritium, is an example of PEM being exposed to ionizing radiation [ 9 , 10 ]. Tritium is a beta ( β-) radiation emitter of a half-life of 12.3 years and is a byproduct in fission and a main fuel in nuclear fusion reactors. Therefore, radiation stability and ability to sustain functionality under the radiation environment of the membrane material is of great importance. Radiation stability studies of Nafion® membrane have demonstrated considerable reduction of mechanical and thermal stability due to the degradation of polymer backbone within free radical induced reactions. Radiation stability has been assessed after irradiation with either gamma of electron beam radiation in dose ranges up to 500kGy [11–15] . Moreover, a dedicated study of tritium beta radiation effects has been performed by Hongqiand et al. by immersing the membranes in the tritiated water ( ∼12TBq/L). Results of other author studies indicated to increased fluorine anion concentration released in the water containing Nafion® after exposure to 200kGy as a result of radical induced -CF 2 backbone degradation and release of fluorine anions and atoms as the primary products of radiolysis [16] . In addition, direct ionizing particle interaction with the polymer molecules radiation chemistry of the surrounding environment should be considered. In water environment water radiolysis products, such as H, OH radicals, H 2 O 2 , may have significant role in the radiolysis reactions as reported by Ghassemzadeh et al . [17] . Moreover, water radiolysis has an effect also due its high content in the polymer itself and in this case the products are formed within the polymer that leads to the destruction throughout all the volume of the material [11] . It must be emphasized that destruction products of the fluorine containing polymer are toxic, corrosive and possess danger both to the employees and environment [18] . For example Balko and Chaklos reported of several volatile products such as CF 4 , COF 2 , CF 2 = CF-CF 3 formed during degradation of Nafion® [11] . It can be expected that SPEEK possess higher radiation stability due to the aromatic polyether-ether ketone backbone. It is well known from the radiation chemistry of organic compounds that aromatic polymers are radiation resistant since π-electrons of the aromatic system are shared by the entire system and an excitation of π-electron does not represent a concentration of energy at a particular location in the molecule. Additionally, the aromatic system due to the resonance phenomena can also act as an energy sink for energy absorbed by other parts of the molecule, affecting increased thermal stability and stability under influence of exposure to non-ionising (UV) and high energy ionising radiation [19] . To assess mechanisms of radiation effects a low temperature (77K) irradiation has been performed by Li et al. and formation of radicals was observed during these studies by the electron spin resonance (ESR) measurements. However, after storage for less than 20 minutes or thermal annealing the ESR signal was diminished due to the rapid radical decay [20] . The mechanisms of the radiation effects of PEEK are considered to be simultaneous crosslinking and scission under the high dose irradiation of electron beam, whereas under gamma irradiation in air chain scission occurs mainly on the surface, whereas crosslinking within the bulk of the polymer [21] . As mentioned above, the irradiation environment is of great importance, therefore. There have been reported comparative studies of PEEK and other polymer radiation resistance under oxidative conditions, where PEEK demonstrated its superior properties regarding the exposure to harsh environments [21] . It must be emphasized that PEEK has been considered and already has been used as a base material for radiation shielding composite materials [22] . It has been approved for the application in spacecrafts and nuclear fusion reactors, where the exposure to radiation can reach 50MGy or greater doses [23] . Therefore, a high dose exposure studies have been implemented up to 100MGy [24] . The main interest of the high dose studies included assessment of the change of the thermal properties [25] . Aging studies of PEEK with simultaneous exposure to high temperature and gamma irradiation has shown that degradation is mainly induced by heat. Study performed by Yang et al. demonstrates that thermal degradation mechanism of PEEK is based on scission of the aromatic ether bonds generating radicals followed by the oxidation and crosslinking [26] . All of the studies described above regarding the radiation stability were performed for PEEK, however, there is a lack information on the sulfonated polymer available. Moreover, there is an interest of direct comparison of Nafion® and SPEEK performance in radiation environments. Currently, there is a data of comparably low dose Nafion® radiation resistance and high doses for PEEK. The aim of the study was to evaluate the effect of ionizing radiation on the mechanical, thermal properties and proton conductivity of Nafion® and sulfonated poly (ether ether ketone) (SPEEK) proton exchange membranes (PEM) under identical irradiation conditions. The irradiation dose range was chosen regarding the potential exposure conditions in future applications. In ITER, the International Thermonuclear Experimental Reactor, an electrolyser should maintain its performance during 2 years of operation in tritiated water of 9 TBq/kg which corresponds to the irradiation of about 530 kGy [27] . 2. Materials and methods 2.1. Materials Commercially available Nafion® 212 and Nafion® 117 membranes of 50.8 μm and 183 μm thickness both were purchased from Chemours Company (Wilmington, Delaware, USA). Before their use, the polyester cover sheets and backing films were removed. Poly(ether ether ketone) (PEEK) granules were obtained from Sigma Aldrich (MO, USA). Other materials and chemicals used for the sulfonation of PEEK, development, and testing of membrane materials (sulfuric acid, N,N-dimethylformamide, etc.) were of reagent grade and used as received. Sulphate anion (SO 4 −2 ) 2 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 standard (10 0 0 mg/L) for ion chromatography (IC) was purchased from Hach Lange GmbH (Dusseldorf, Germany). Fluoride (F −) standard (10 0 0 mg/L) for IC was purchased from Sigma Aldrich (USA). Ultrapure deionised water was generated by a Millipore Milli-Q TM system (Billerica, MA, USA). 2.2. Preparation of SPEEK membranes PEEK pellets were dried in a vacuum oven at 100 °C overnight. Then appropriate weight (10 g) of the PEEK was dissolved in 200 mL of concentrated sulfuric acid (98%) and vigorously stirred at room temperature for 24 h. Afterwards obtained polymer solution was placed in a thermostat and heated at 36 °C for 24 h. The polymer suspension was slowly poured into a large amount of cool deionized water. The precipitated SPEEK polymer yarns were filtered, washed with deionized water until neutral pH and dried at 80 °C for 24 hours. The procedure was repeated to obtain the necessary samples for membrane replicates. SPEEK membranes were obtained by solvent cast method as follows: 0.5 g of the SPEEK polymer were dissolved in 20 mL of dimethylformamide followed by stirring at 100 °C for 1 h. Then polymer solution was cast into Petri dish and dried for 24 h at 80 °C to remove the solvent. The thickness of the prepared SPEEK membranes was 120 ±20 μm. 2.3. Irradiation Irradiation was carried out by the Varian Clinac® medical accelerator at the Accelerator Laboratory of Jyväskylä University (Finland). Both Nafion® and SPEEK polymer films (3 replicates of each) were sealed in separate sterile polyethylene sample bags and irradiated in air at 0 (control), 50, 100, 250, and 500 kGy of 6 MeV accelerated electron beam (EB). In addition, membrane samples of 10 mg weight were immersed in deionised water and also irradiated at doses of 25 to500 kGy. The dose rate was equal to 1 kGy/min. 2.4. Methods of analysis 2.4.1. Degree of SPEEK sulfonation (DS) Degree of sulfonation (DS) was determined by acid-base titration as described in literature [28] . 0.5 g of the dried SPEEK precipitate was properly weighted in a 100 mL conical flask and soaked in 25 mL of 0.1 M NaCl solution to provide exchange of H + to Na + of all sulfonated groups of SPEEK. Then samples were titrated with standardized 0.1 M NaOH solutions. The DS was calculated using the dry weight of the sample and the quantity of exchanged protons. The triplicate measurements indicated DS to be 0.75 ±0.05. 2.4.2. Fourier-transform infrared (FTIR) spectroscopy and FTIR-TGA Bruker Vertex 70v vacuum spectrometer equipped with an attenuated total reflection (ATR) accessory was used in this study. Recording range 400 cm −1 - 40 0 0 cm −1 , spectral resolution ±2 cm −1 , in 2.95 hPa vacuum, at least 3 measurements par sample, 20 spectra per measurement, giving a sum of at least 60 spectra per sample were obtained. Average spectrum was calculated from the measured three replicate spectra. Data were collected using TRIOS Software v4.3.1 and FTIR program OPUS by Bruker, analysed within OriginPro v8.0 scientific graphing and data analysis software. 2.4.3. Ion chromatography Ion chromatography (IC) was used to determine concentrations of dissolved fluoride and sulphate anions as the degradation products of irradiated membranes in water samples. Analysis was performed on a Dionex TM Integrion TM high pressure IC system from Thermo Fisher (MS, USA) equipped with Dionex ionPacTM AS18- 4μm (4 ×150 mmm) analytical column, Dionex EGC III KOH potassium hydroxide Eluent Generator and Dionex AS-AP autosampler. A laboratory validated method for seven anion (fluoride, chloride, nitrate, nitrite, sulphate and phosphate) analysis was used. All chromatographic experiments were carried out in the isocratic mode, at 30 °C. The operating conditions were: 1 mL/min eluent flow rate, 250 μL loop volume, 25 μL sample injection volume, 124 mA current in suppressor. Water samples of 50 μL volume were diluted 1:10 in chromatographic vials with deionised water. Three replicates were tested for each of the samples including the control samples of non-irradiated deionised water. Seven-point calibration curves were obtained in the 0.05–0.5 mg/L concentration range for fluoride anions, and 0.05-5 mg/L for sulphate anions with the correlation coefficients R 2 > 0.999 to determine anion concentrations in the irradiated water samples. The method uncertainty (U, %) of 11% for F −and 10% for SO 4 −2 were used to evaluate the total RSDs (%). The levels of the quantification (LOQ) of the method were 0.05 mg/L for F −and 0.5 mg/L for SO 4 −2 , respectively. The sample weights were used to calculate the concentration of anions formed from the degradation of PEM membranes [13] . 2.4.4. Thermogravimetry analysis (TGA) TGA measurements were performed using an Exstar SII TG/DTA 6300 (Seiko Instruments Inc., Chiba, Japan) thermogravimetric instrument. Samples of Nafion® and SPEEK membranes ( ∼2.5 mg weight) were placed in alumina crucibles and tested under air flux (100 ml/min) from 25 to 950 °C with a heating rate of 10 °C/min. Three replicates were tested for each sample. The average TGA curves of the weight loss versus temperature and the derivative (DTG, %/ °C) were analysed. For comparative reasons, the TGA data of the sulfonate group decomposition were used to calculate DS for SPEEK membranes after their EB irradiation. The Eq. (1 ) described in the literature [29] was used: DS = n ( S O 3 H ) / n ( PEEK ) = M ( PEEK ) / [ ( m / m ) ×M ( S O 3 H ) ] (1) where M(PEEK) and M(SO 3 H) are the molecular masses of PEEK monomer (288.7 g/mol) and = sulfonic acid groups (81 g/mol), m is the mass of SPEEK at the starting of the desulfonation, and m is the mass loss due to the desulfonation. 2.4.5. Differential scanning calorimetry (DSC) Measurements of the calorimetric properties were provided on DSC 1/200 W equipment (Mettler-Toledo, Greifensee, Switzerland) equipped with intercooler. The preconditioned membranes samples at 55 °C under vacuum for 48 hours were used for DSC analysis. Then samples with approximate weight of ∼10 mg were placed in aluminium pans and tested under a constant nitrogen purge (60 mL/min). Samples were heated from −5 to 250 °C at a heating rate of 1 0 °C/min, conditioned at 250 °C for 5 minutes, followed by cooling from 250 to −5 °C at a cooling rate of 10 °C/min. Samples were stored at this temperature for 5 min and further heated to 250 °C at a heating rate of 10 °C/min. The data from the subsequent heating–cooling–heating cycles were analysed by triplicate. Calibration of the instrument was performed with indium sample (T m = 156.6 °C). Pristine and electron beam irradiated Nafion® and SPEEK membrane samples were dried under vacuum until reaching constant mass and kept in the desiccator until analysis to reduce the impact of the adsorbed water on the calorimetric characteristics. 2.4.6. Mechanical properties Mechanical properties were determined by constructing stressstrain curves at room temperature (23 ±2 °C) and relative humidity RH = 30 ±2% using a Tinius Olsen H1KS universal tensile testing machine (PA, USA) equipped with 1kN strain gage load cell. Samples were cut as rectangles with a length of 70 mm (the gauge length of 50 mm) and a width of 5 mm. Crosshead moving 3 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Fig. 2. FTIR spectra of pristine and electron irradiated Nafion® membranes. speed was set to 10 mm/min during the tests. From the stress– strain curves, the modulus of elasticity, the tensile strength as the maximum stress at rupture and the elongation at break (were determined as well as the stress at yield was estimated in the case of semi-crystalline SPEEK membranes as the point of the intersection of the tangent lines to the regions corresponding to the elastic and plastic deformation. Dynamic mechanical analysis (DMA) was used for characterisation of viscoelastic properties, thermal transition stages, including the storage modulus under the glass transition, including at room temperature and conventional operation temperatures of PEM in the case of FC applications (80 °C) that are lower to potential applications under hydrogen isotope exchange transition energies. For each type of membrane at least 5 parallel samples were tested, and the averaged values were compared. The experimental errors were estimated according to the Student’s t-distribution. 2.4.7. Dynamic mechanical analysis (DMA) Samples were preconditioned in vacuum oven at 55 °C for 48 h prior to analysis to remove the absorbed water during the storage. DMA testing was performed at 1 Hz from -10 °C to + 250 °C under nitrogen atmosphere on a DMA/SDTA861 instrument (Mettler Toledo, Columbus, OH, USA) in extension mode according to EN ISO 6721–2. Samples were cut in the pieces of 15 mm ×7 mm size and their dynamic mechanical thermal properties such as the dynamic storage modulus, loss modulus, and the damping factor (tan δ) were determined at the heating rate of 5.0 °C/min. 2.4.8. Electron paramagnetic resonance (EPR) spectroscopy EPR spectra were measured with Bruker ELEXSYS-II E500 CWEPR spectrometer at room temperature. Magnetic field modulation parameters were 100 kHz and 0.1 mT; microwaves were generated at 9.834 GHz frequency and 20 mW power. Membranes were cut into smaller pieces (3 ×4 mm) and placed into EPR sample tubes. To estimate uncertainties, 5 different sam ple pieces were measured for each irradiation dose. EPR signal intensities have been calibrated to sample mass. EPR spectra simulations were performed in EasySpin software [30] . 2.4.9. Proton conductivity Conductivity measurements of non-irradiated and electron beam irradiated SPEEK and Nafion® membranes were performed through-plane using impedance analysis with various cell configurations as described in the previous study [31] . The conductivity was determined by means of impedance measurements using a BioLogic VMP3 multichannel potentiostat / galvanostat frequency response analyser (FRA) at AC amplitude of 10 mV with EC-Lab® software. The frequency interval employed for the measurements ranged from 10kHz to 10mHz. For the ionic conductivity, the membranes were immersed in a de-ionized water for 24 hours and kept in a chamber with a water bath. Samples were sandwiched between two copper discs with a diameter 10 mm. The pressure is kept constant and defined by the maximum limits of screws used to close the probe arrangement. Dry film thickness values were measured with a micrometre. Proton conductivity of the samples was calculated using the following Eq. (2) : σ= L / R ∗A , (2) where σis the proton conductivity (S cm −1 ), L the thickness(cm) of the polymer film, A the contact area between the electrodes and the polymer film, and R the bulk membrane resistance calculated from the Nyquist plot. 3. Results and discussion 3.1. FTIR spectra analysis Fig. 2 shows typical spectra of non-irradiated and irradiated Nafion® membranes. All the spectra exhibited strong absorption bands at 1144 cm −1 and 1199 cm −1 assigned to the asymmetric and symmetric stretching vibrations of the branched -CF 2 groups, a narrow band at 1014 cm −1 associated with symmetric stretching S –O vibration of the –SO 3 2 −group, and two strong absorption bands allocated at 980 cm −1 and 960 cm −1 due to C –F and C – O –C stretching vibrations [32] . Some weak peak bands presented at 805 cm −1 , 630 cm −1 , 511 cm −1 and 550 cm −1 were associated with C –S and C –F stretching and bending vibrations [33] . Spectra of electron beam irradiated samples showed a formation of new weak broad signal around 1720 cm −1 , which could be associated with the presence of carbonyl groups or possible presence of low levels of hydrated protons or protonated water molecules in the form of hydronium ions [34] . Irradiation introduced broad weak band with a maximum around 1775 cm −1 that can be attributed to defluorination and formation of CF = CF bonds 4 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Table 1 Anion contents in water extracts of irradiated Nafion® and SPEEK samples. Dose (kGy) Nafion® SPEEK SO 4 2- (mg/g Polym. ) ±RSD F −(mg/g Polym. ) ±RSD SO 4 2- (mg/g Polym. ) ±RSD 0 < 0.005 - < 0.001 - < 0.005 - 50 0.07 0.01 0.15 0.02 < 0.005 - 100 0.10 0.01 0.21 0.02 < 0.005 - 250 0.39 0.04 0.76 0.08 0.14 0.01 500 1.15 0.12 1.49 0.16 0.26 0.03 and also the radiation induced scission at the vinyl ether groups and formation of C = O bonds as a result of chain destruction and radiation induced oxidation during irradiation in air conditions [ 13 , 35 ]. Signal intensity at 1775 cm-1 correlates with the dose indicating the decomposition of polymer and formation of oxidation products [34] . Slight increase of the bond intensities at 511 cm −1 and 550 cm −1 has also been observed indicating the radiation induced decomposition of polymer (CF 2 )n backbone. FTIR spectra of pristine SPEEK contained several absorption bands, which are characterised in Fig. 3 . The most intensive signal bands observed in SPEEK spectra were attributed to O –H and S = O bond vibrations of the sulfonic acid at 1155 - 1217 cm −1 , stretching vibrations of C = O at 1645 cm −1 and 1610 cm −1 attributed to acid and ketone groups, signals at 1480 cm −1 , 1217 cm −1 for benzene -Ph- groups and Ph –CO –Ph groups. and Number of signals (1020, 1080, and 1250 cm −1 some of most intense) can be attributed to for O = S = O band stretching and bending vibrations [ 36 , 37 ]. Broad peak at around 2820 cm −1 could be assigned to C - H stretching [38] , and at 2950 cm −1 to C –H asymmetric stretch [39] , whereas broad signal centred at around 3075 cm −1 could be assigned as –CH stretching [40] . Signals at 709cm −1 [41] , 1308cm −1 [42] , 1410cm −1 [43] corresponds to benzoyl ring, C = O band and COO, respectively. Electron beam irradiation at doses below 500 kGy did not cause notable changes to most of these absorption bands of SPEEK membranes, whereas at 500 kGy increase of band intensities at 1280 cm −1 , 679-685 cm −1 and 790-840 cm −1 , as well as crystalline phase signals at 586 cm −1 was observed ( Fig. 4 ). Changes of the crystalline phase signal at 586 cm −1 could be attributed to the effect of high dose irradiation on reorganization of the chemical bonds and crosslinking, whereas the increase of O = S = O group signals at 1280 cm −1 associated with the formation of crosslinks between the molecules. In most cases, when crosslinking of SPEEK was reported at doses up to 400 kGy, the ionising radiation treatment induced activation of crosslinking promoters containing active double bond (di- or tri-vinyl or acryl) additives, which acted as the secondary crosslinking agents affecting formation of crosslinking bounds between SPEEK molecules [ 37 , 44 ]. In this case, the absence of crosslinking agent and rather low irradiation time at doses below 250 kGy resulted in no changes in the structure comparing to irradiation up to 500 kGy. 3.2. Ion chromatography Nafion® and SPEEK membranes were also irradiated in water and post irradiation measurement of the concentration of sulphate SO 2 − 4 and fluorine F −ions were done by the means or ion chromatography. Results are summarized in Table 1 and demonstrated in Figs. 5 and 6 . Water radiolysis products might play a significant role regarding the radiation resistance of the membrane materials. Radical reactions during irradiation of Nafion® have been described by Ghassemzadeh et al. [17] and simultaneous effect of water and temperature by Yamaguchi et al . [45] . Ion emission indicates the degradation of the polymer. In case of Nafion® the fluoride ions are being released because of scission of the polymer chains. Results demonstrated strong correlation between the irradiation dose and fluorine ion concentration from 0.15 mg/g Polym after irradiation with 50kGy up to 1.49 mg/g Polym after - 500kGy. To compare both polymers the concentration of sulphate ions was also measured. An increase of sulphate ions was observed already at irradiation dose 50kGy for Nafion®, whereas for SPEEK slight release of sulphate ions was observed only at doses reaching 250 and 500kGy. Stability of macromolecular structures of Nafion® and SPEEK can be differently affected as further discussed in TGA results, which indicated lower stability of polytetrafluoroethylene backbone as compared to that of aromatic poly(etherether-ketone) backbone of SPEEK. The IC data were in good agreement with the results of FTIR analysis indicating increase of the degradation products in Nafion® correlating with the radiation dose above 100 kGy. 3.3. TGA analysis Thermal stability is an important factor considering the PEMs, which are typically operated at temperatures around 80 °C. When operating in harsh conditions, temperatures may also expand 100 °C leading to potential thermal oxidation combined with ionising radiation effects resulting in reduction of the material performance of proton conductivity and decrease of operation lifetime. Figs. 7 and 8 show thermal decomposition behaviour of pristine and electron beam irradiated Nafion® and SPEEK membranes heated up to 950 °C in air atmosphere. TGA analysis of both SPEEK and Nafion® indicated that mass decrease was accompanied with exothermic signals due to exothermic oxidation of polymers. Nafion® 117 membranes underwent three main thermal transitions. The initial stage at temperatures above 100 °C was due to the desorption of water bounded to the sulfonic groups of Nafion® with the average weight loss reaching 4% at doses below 250 kGy ( Fig. 7a ). The calculated weigh loss was 7% at 500 kGy due to the radiation induced oxidation and formation of adsorbed hydrolysis products [46] . The second weight loss undergo at temperatures between 280 and 380 °C and was associated with the degradation of sulfonic groups, decomposition of side chains –OCF 2 CF 2 –SO 3 H [47] . The third, multistep stage with several peaks determined by DTG ranged between 380 and 550 °C ( Fig. 7b ) attributed to multistep radical reactions during the degradation of the polymer backbone chains CF 2 –CF 2 [48] . From the DTG curves it can be well seen that ionising radiation facilitates the degradation at doses above 100 kGy. Mainly it is attributed to oxidation reaction induced chain scission and formation of backbone scission products that was confirmed from the FTIR spectra analysis. The DTG curves of Nafion® irradiated up to 250 and 500 kGy shifted to lower temperatures demonstrating reduced thermal stability ( Fig. 7b ). In the DTG of Nafion® two main radiation induced changes were observed at around 357 °C. Starting temperature for rapid decomposition differed by more than 10 °C comparing non-irradiated polymer 5 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Fig. 3. Typical FTIR spectrum of non-irradiated SPEEK tested in this study. 6 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Fig. 4. FTIR spectra of pristine and electron beam irradiated SPEEK membranes. Fig. 5. Concentrations of sulphate ions in the irradiated water samples of Nafion® and SPEEK membranes. Fig. 6. Concentration of fluoride ions in the irradiated water samples of Nafion® membranes. and polymer irradiated with 500 kGy. For the Nafion® the mass loss up to 400 °C is due degradation of sulfonic groups, whereas the sharp weight loss above 400 °C is due polymer backbone, decomposition of main chains. However, at doses below 250 kGy Nafion® 117 membrane expressed thermal stability up to 240 °C, which make the suitable application of Nafion® membranes after low dose radiation treatment. SPEEK demonstrated notably higher thermal stability if compared to Nafion®, which was completely degraded at 500 °C. Almost 50% of the SPEEK fraction remained at this temperature. Electron beam irradiation induced transformations in the SPEEK cluster structure resulting in higher thermal stability compared to pristine SPEEK as can be seen in Fig. 8a [49] . High thermal stability is also associated with the thermal-resistance of the aromatic polyketone backbone leading to the reduction of the thermal energy. Irradiation dose above 100 kGy introduced changes in the ionic clusters of the sulfonated groups leading to moderate reduction of the thermal properties. As noted in the literature [50] , degree of sulfonation has notable impact of the thermal behaviour of SPEEK membrane material [51] . TGA curves of SPEEK demonstrated three decomposition regions typical to highly sulfonated SPEEK membranes ( Fig. 8a ). First mass loss stage was related to the evaporation of water adsorbed to the surface of polymer ( Fig. 8a ). Second weight loss of about 20% at a temperature range of 300–390 °C can be associated with the thermal degradation of sulfonic acid groups in SPEEK. This stage of the structure degradation was slightly affected by the electron beam irradiation. The third weight loss region at temperatures ranged between 500 and 650 °C can be attributed to the exothermic degradation as a result of thermal oxidation and decomposition of the main chain of the polymer. It can be observed that increase of the irradiation dose up to 250 kGy has increased thermal resistance of the SPEEK backbone ( Fig. 8b ). DTG curves of SPEEK indicated several parallel processes occurring during thermal degradation above 500 °C, with similar shift of the peaks to higher temperatures in case of samples irradiated up to 250 kGy. Irradiation up to 500 kGy caused reduced thermal degradation due to both radiation induced rearrangement and loss of the order in the ionic clusters of the SPEEK as a result of radiation induced crosslinking. Recent reports of electron beam effect on the graft crosslinked SPEEK in the presence of crosslinking agents have indicated benefits on increased thermal stability in case of electron beam modified SPEEK [ 44 , 52 ]. Mass change analysis indicated that both nonirradiated and irradiated SPEEK is thermally stable up to around 300 °C and that is by at least 60 °C higher temperature compared to Nafion®. 3.4. DSC thermal analysis Typical DSC curves for nonirradiated and irradiated Nafion® and SPEEK membranes are presented in Figs. 9 and 10 . Both polymers demonstrated rather different character of thermal transitions that determined different heating cycles, which were used for the characterisation of membrane materials. As it can be seen in Fig. 9a , during the first heating cycle of Nafion® there was a sharp endothermic peak at temperature ranging between 100-180 °C assigned to molecular rearrangements inside the polar clusters of the polymer [ 53 , 54 ]. This is commonly associated as the first glass transition followed by the second glass transition at higher temperatures. Authors of recent studies reported that the 1 st glass transition is caused by the mobility of the main chain in the polymer matrix, whereas the 2 nd glass transition region is attributed to the vibrations of the sulfonyl groups attached to the polymer backbone [55] . In this report, the first glass transition for Nafion® 117 was tested under similar conditions to that provided in the present study. The glass transition temperature Tg of completely dried membrane was 132 °C, whereas the second glass transition associated with cluster transformations was determined at temperatures exceeding 190 °C [48] . Due to the reduction of the thermal stability, second cooling and heating cycles did not provide information of 1 st glass transition of Nafion®. Mainly the changes at higher temperatures were attributed to oxidation and decomposition of Nafion® backbone - CF 2 groups. That is why only the DSC curves of the 1 st melting cycle were analysed. For nonirradiated Nafion® membrane a peak centred at 170 °C. At doses up to 100 kGy there was little changes in the temperature, whereas a gradual decrease from 173 to 144 and 140 °C was determined with increase of the radiation dose from 100 to 250, and 500 kGy, respectively, that is associated with radiation induced oxidation and chain scission ( Fig. 9b ). 7 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 Fig. 7. TGA (a) and DTA (b) curves of pristine and electron beam irradiated Nafion® membranes. Fig. 8. TGA (a) and DTA (b) curves of pristine and electron beam irradiated SPEEK membranes. 8 E. Pajuste, I. Reinholds, G. Vaivars et al. Polymer Degradation and Stability 200 (2022) 109970 [31] D. Fedorenko, G. Vaivars, Different approaches in sulfonated poly (ether ether ketone) conductivity measurements, IOP Conference Series: Materials Science and Engineering, 2019 . [32] Z. Liang, W. Chen, J. Liu, S. Wang, Z. Zhou, W. Li, G. Sun, Q. Xin, FT-IR study of the microstructure of Nafion® membrane, J. Membr. Sci. 233 (1) (2004) 39–44 . [33] A . Gruger, A . Régis, T. Schmatko, P. Colomban, Nanostructure of Nafion® membranes at different states of hydration: an IR and Raman study, Vib. Spectrosc. 26 (2) (2001) 215–225 . [34] A.S. Rao, K.R. Rashmi, D.V. Manjunatha, A. Jayarama, R. Pinto, Role of UV irradiation of Nafion membranes on ionic groups responsible for proton conduction and mechanical strength: a FTIR spectroscopic analysis, Mater. Today Commun. 25 (2020) 101471 . [35] L. Ghassemzadeh, K.-D. Kreuer, J. Maier, K. Müller, Chemical degradation of nafion membranes under mimic fuel cell conditions as investigated by solid-s- tate NMR spectroscopy, J. Phys. Chem. C 114 (34) (2010) 14635–14645 . [36] M. Kumari, H.S. Sodaye, R.C. Bindal, Cross-linked sulfonated poly(ether ether ketone)-poly ethylene glycol/silica organic–inorganic nanocomposite membrane for fuel cell application, J. Power Sources 398 (2018) 137–148 . [37] J.-M. Song, J. Shin, J.-Y. Sohn, Y.C. Nho, Preparation and characterization of SPEEK membranes crosslinked by electron beam irradiation, Macromol. Res. 19 (10) (2011) 1082 . [38] M. Jafarpour, A. Rezaeifard, V. Yasinzadeh, H. Kargar, Starch-coated maghemite nanoparticles functionalized by a novel cobalt Schiff base complex catalyzes selective aerobic benzylic C–H oxidation, RSC Adv. 5 (48) (2015) 38460– 38469 . [39] E. Bogomolny, M. Huleihel, Y. Suproun, R.K. Sahu, S. Mordechai, Early spectral changes of cellular malignant transformation using Fourier transform infrared microspectroscopy, J. Biomed. Opt. 12 (2) (2007) 024003 . [40] P. Maity, S.V. Kasisomayajula, V. Parameswaran, S. Basu, N. Gupta, Improvement in surface degradation properties of polymer composites due to pre-pro- cessed nanometric alumina fillers, IEEE Trans. Dielectr. Electr. Insul. 15 (1) (2008) 63–72 . [41] M. Zhou, Z. Xiong, D. Yang, Y. Pang, D. Wang, X. Qiu, Preparation of slow release nanopesticide microspheres from benzoyl lignin, Holzforschung 72 (7) (2018) 599–607 . [42] A. Alexandropoulos, Y.S. Al Jabbari, S. Zinelis, T. Eliades, Chemical and mechanical characteristics of contemporary thermoplastic orthodontic materials, Aust. Orthod. J. 31 (2) (2015) 165–170 . [43] Antônio Sávio G. Magalhães, Manuel P. Almeida Neto, Maslândia N. Bezerra, Nágila M. P. S. Ricardo, Judith P. A. Feitosa, Application of ftir in the determination of acrylate content in poly(sodium acrylate-co-acrylamide) superabsorbent hydrogels, Química Nova 35 (7) (2012) 1464–1467 . [44] P. Khomein, W. Ketelaars, T. Lap, G. Liu, Sulfonated aromatic polymer as a future proton exchange membrane: a review of sulfonation and crosslinking methods, Renew. Sustain. Energy Rev. 137 (2021) 110471 . [45] M.. Yamaguchi, K. Ohira, Gamma radiolysis of perfluorosulfonic acid ionomers and their side chain model compounds in water, Radiat. Phys. Chem. 159 (2019) 89–94 . [46] C. Gong, L. Pinatti, G. Lavigne, M.T. Shaw, D.A. Scola, Thermal stability of end– capped and linear sulfonated polyimides, sulfonated polystyrene, and Nafion 117, J. Appl. Polym. Sci. 135 (3) (2018) 45694 . [47] R. Sigwadi, M.S. Dhlamini, T. Mokrani, F. Ṋ emavhola, P.F. Nonjola, P.F. Msomi, The proton conductivity and mechanical properties of Nafion®/ZrP nanocomposite membrane, Heliyon 5 (8) (2019) e02240 1–9 . [48] L. Sun, J.S. Thrasher, Studies of the thermal behavior of Nafion® membranes treated with aluminum(III), Polym. Degrad. Stab. 89 (1) (2005) 43–49 . [49] A. Ramgobin, G. Fontaine, S. Bourbigot, A case study of polyether ether Ketone (I): investigating the thermal and fire behavior of a high-performance material, Polymers 12 (8) (2020) 1789 . [50] P. Knauth, H. Hou, E. Bloch, E. Sgreccia, M.L. Di Vona, Thermogravimetric analysis of SPEEK membranes: thermal stability, degree of sulfonation and cross-linking reaction, J. Anal. Appl. Pyrol. 92 (2) (2011) 361–365 . [51] G. Wu, S.-J. Lin, I.-C. Hsu, J.-Y. Su, D.W. Chen, Study of high performance sulfonated polyether ether ketone composite electrolyte membranes, Polymers 11 (7) (2019) 1177 . [52] S. Han, M.-S. Zhang, J. Shin, Y.-S. Lee, 1,6-Bis(4-vinylphenyl)hexane as a crosslinking agent for the preparation of crosslinked sulfonated poly(ether ether ketone) membranes by EB irradiation, Radiat. Phys. Chem. 97 (2014) 313–318 . [53] S. Mollá, V. Compañ, Polyvinyl alcohol nanofiber reinforced Nafion membranes for fuel cell applications, J. Membr. Sci. 372 (1) (2011) 191–200 . [54] M. Iijima, Y. Sasaki, T. Osada, K. Miyamoto, M. Nagai, Nanostructure of clusters in nafion studied by DSC, Int. J. Thermophys. 27 (6) (2006) 1792–1802 . [55] H.-Y. Jung, J.W. Kim, Role of the glass transition temperature of Nafion 117 membrane in the preparation of the membrane electrode assembly in a direct methanol fuel cell (DMFC), Int. J. Hydrogen Energy 37 (17) (2012) 12580–12585 . [56] L. Unnikrishnan, S. Mohanty, S.K. Nayak, Proton exchange membranes from sulfonated poly(ether ether ketone) reinforced with silica nanoparticles, High Perform. Polym. 25 (7) (2013) 854–867 . [57] M.J. Parnian, S. Rowshanzamir, F. Gashoul, Comprehensive investigation of physicochemical and electrochemical properties of sulfonated poly (ether ether ketone) membranes with different degrees of sulfonation for proton exchange membrane fuel cell applications, Energy 125 (2017) 614–628 . [58] E. Sgreccia, J.F. Chailan, M. Khadhraoui, M.L. Di Vona, P. Knauth, Mechanical properties of proton-conducting sulfonated aromatic polymer membranes: stress–strain tests and dynamical analysis, J. Power Sources 195 (23) (2010) 7770–7775 . [59] J.M. Song, D.W. Shin, J.Y. Sohn, Y.C. Nho, Y.M. Lee, J. Shin, The effects of EB-ir- radiation doses on the properties of crosslinked SPEEK membranes, J. Membr. Sci. 430 (2013) 87–95 . [60] J.M. Song, S.Y. Lee, H.S. Woo, D.W. Shin, J.Y. Sohn, Y.M. Lee, J. Shin, EB-crosslinked SPEEK electrolyte membrane with 1,4-butanediol divinyl ether/triallyl isocyanurate for fuel cell application, J. Membr. Sci. 469 (2014) 209–215 . [61] S.H. Almeida, Y. Kawano, Effects of X-ray radiation on Nafion membrane, Polym. Degrad. Stab. 62 (2) (1998) 291–297 . [62] B. Jaleh, N. Gavary, P. Fakhri, N. Muensit, S.M. Taheri, Characteristics of PVDF membranes irradiated by electron beam, Membranes 5 (1) (2015) 1–10 . [63] E.M. Kornacka, G. Przybytniak, A. Nowicki, Radical processes initiated by ionizing radiation in PEEK and their macroscopic consequences, Polym. Adv. Technol. 30 (1) (2019) 79–85 . [64] A. Bosnjakovic, S. Schlick, Nafion perfluorinated membranes treated in fenton media: radical species detected by ESR spectroscopy, J. Phys. Chem. B 108 (14) (2004) 4332–4337 . [65] A. Bosnjakovic, M.K. Kadirov, S. Schlick, Using ESR spectroscopy to study radical intermediates in proton-exchange membranes exposed to oxygen radicals, Res. Chem. Intermed. 33 (8) (2007) 677 . [66] M. Danilczuk, A. Bosnjakovic, M.K. Kadirov, S. Schlick, Direct ESR and spin trapping methods for the detection and identification of radical fragments in Nafion membranes and model compounds exposed to oxygen radicals, J. Power Sources 172 (1) (2007) 78–82 . 15