Influence of zeolite nanoparticles on photostability of ethylene vinyl alcohol copolymer (EVOH)

Polymer Degradation and Stability 121 (2015) 137e148

Contents lists available at ScienceDirect

Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab

Influence of zeolite nanoparticles on photostability of ethylene vinyl alcohol copolymer (EVOH) Ievgeniia Topolniak a, b, Jean-Luc Gardette a, b, Sandrine Therias a, b, * a b

CNRS, UMR 6296, Institut de Chimie de Clermont-Ferrand, BP 80026, F-63171 Aubi
ere, France Clermont Auvergne Universit e, Universit e Blaise Pascal, Institut de Chimie de Clermont-Ferrand, BP 10448, F-63000 Clermont-Ferrand, France

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 June 2015 Received in revised form 24 August 2015 Accepted 25 August 2015 Available online 31 August 2015

New ethylene vinyl alcohol copolymer EVOH nanocomposites based on zeolite particles were processed by solution or melt-mixing to synthesise potential encapsulants for flexible organic solar cells (OSC). The photochemical behaviour of pristine EVOH copolymer and the influence of zeolite particles with different size and increasing nanoparticle loading were studied to evaluate the photostability of such packaging materials. Photooxidation of pristine EVOH films (with two different ethylene unit contents (27% or 44%) was performed under accelerated artificial conditions that allow for the proposal of a comprehensive mechanism of EVOH oxidation. The photochemical behaviour of EVOH/zeolite nanocomposites with different morphology and size of particles was compared, revealing a higher oxidation of the polymer in the presence of zeolite particles, which can be mainly attributed to iron impurities. © 2015 Elsevier Ltd. All rights reserved.

Keywords: EVOH Photooxidation Zeolite Nanocomposites

1. Introduction Over the previous 20 years, polymer nanocomposites (PNCs), which are particle-filled polymers with at least one dimension of the dispersed particles in the nanometre range, have attracted much attention. It has been demonstrated that introducing inorganic nanoparticles (usually at low content up to 5 wt%) into a polymer matrix improves several properties of the material in comparison with the pristine polymer. Among these properties are heat resistance [1e4], thermal [5,6], mechanical [7], electrical [8] properties, and barrier properties, such as a decrease of oxygen [6] and water vapour [9] transition rates (abbreviated OTR and WVTR, respectively) for nanocomposites with low nanofiller content (approximately 3e5%). These enhanced performances are attributed to the unique morphology with large interfacial area of nanoparticles as the supramolecular organisation of the particles is important to reach the desired macroscopic properties. It is known that organic solar cells (OSCs), especially the active layer and electrodes, are sensitive to trace quantities of oxygen and humidity, which jointly with light, can strongly decrease the performance and the life-time of the device [10]. One solution to

, Universite  Blaise Pascal, * Corresponding author. Clermont Auvergne Universite Institut de Chimie de Clermont-Ferrand, UMR 6296, ICCF, BP 80026, F-63171 Aubiere, France. E-mail address: [email protected] (S. Therias). http://dx.doi.org/10.1016/j.polymdegradstab.2015.08.014 0141-3910/© 2015 Elsevier Ltd. All rights reserved.

overcome this problem is the use of encapsulation to prevent oxygen and moisture from reaching the fragile components of the cell. The development of encapsulation materials for the photovoltaic module protection with high barrier properties is then one of the challenges for OSC competitiveness in the market [11]. The barrier requirement range for OSCs encapsulation is of ~10
3 cm3 m
2 day
1 and ~10
6 g m
2 day
1 for OTR and WVTR, respectively. Moreover, these materials have to be stable within the operating life of the cells. Among the polymers commonly used as barrier materials for applications such as food packaging, ethyleneevinyl alcohol copolymers (EVOH) have good barrier properties to oxygen (OTR ¼ 0.05 cc day
1 atm
1) for EVOH with 32 wt% of ethylene part and thickness 10 microns [12], considerable chemical resistance and high transparency, which make them good candidates for encapsulation coatings. Poly (ethylene vinyl alcohol) copolymers (EVOH) are randomly formed semi-crystalline polymers with different ratio of ethylene and vinyl alcohol groups from 24 up to 55 wt% of ethylene part. With increasing ethylene content in the copolymer, the water barrier properties increase, whereas the oxygen gas barrier properties decrease [13]. This opposite effect provides an opportunity to tune the chemical composition of the polymer for the desired property. The oxygen barrier properties of EVOH are extremely sensitive to moisture, therefore EVOH should be used as a core layer in multilayer packaging structures [14,15]. In terms of permeability coefficients, even the best polymeric

138

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

barriers remain several orders of magnitude higher than the OPV requirements. Improving the barrier properties of the polymer is thus required, which could be achieved through the incorporation of inorganic particles into the polymer. The resulting increase of the barrier effect caused by the introduction of such passive particles is well known. This increase is majorly attributed to a lengthening of the permeation path due to the tortuous path [16] provoked by the particles; passive particles usually do not interact with the permeating gases either through sorption or chemical processes. It has been shown that in the case of EVOH the barrier properties can be increased by adding various types of nanofillers, such as organoclays (bentonite [17], kaolin [13], kaolinite [18], montmorillonite [9], layered double hydroxides [19]) or carbon nanotubes [20]. An alternative approach to the passive nanocomposite is the use of active fillers, which strongly interact with a permeating gas. These active fillers are commonly known as getters and their activity is to either consume a permeating gas through chemical reaction or to trap permeating gases via sorption processes. The common types of getters that can trap water through adsorption include molecular sieves such as zeolites. Zeolites are aluminosilicates, which have an ordered, highly porous structure that allows for a high water sorption capacity. As these getters need a high surface area to adsorb large quantities of gas, they are usually highly porous, often with pore diameters on the order of only a few nanometres. There exists a wide variety of zeolite structures, but zeolites are generally based on cage-like structures created by silicate and aluminosilicate tetrahedral (Scheme 1). These cages form a series of ordered pores with the smallest pore diameters in the range of 6e14 nm. Of particular interest are zeolites as they have the ability to be manufactured as nanoparticles. Nanoparticles of zeolites have been well researched; however, their use in barrier composites has received only very limited attention [21,22]. For example, zeolites were proposed for use in electronics in 1990 [23] but were recently introduced into a device as a relatively thick non-transparent layer of particles in the bottom side of organic light-emitting device (OLED) [24] to ensure a dry environment. It is worth considering that nanofillers can decrease the oxygen permeability of the polymer, which in turn could limit the oxidation rate of the polymer rate in the case of an oxygen diffusion controlled reaction. However, a drawback to the use of nanofillers is their negative impact on the photostability of the polymer. The addition of particles to polymers has been shown many times to affect the photostability of nanocomposites. Although inorganic particles rarely experience significant photodegradation under

sunlight exposure, they can change the mechanisms and/or the rate of photooxidation of the polymer matrix and must therefore always be accounted for when determining the photostability of the nanocomposite. For example, some metal oxides such as zinc oxide and titanium oxide, can have a photocatalytic effect [25]. Conversely, lamellar nanofillers such as clays are reported to have a prodegradant effect on the polymer photooxidation [26e28] due to the presence of transition metal ions such as iron [29e32] that can produce a catalytic decomposition of hydroperoxides. In summary, one should pay attention to the fact that the observed rate of oxidation results in a balance between the two competing processes, O2 barrier and photocatalytic effects, with the photocatalytic effect being highly dependent on the chemistry and purity of both the nanofillers and the polymer. This study is part of a work devoted to the development of EVOH/nanocomposites based on inorganic fillers such as zeolites for potential encapsulation coating for OSC. The present paper focuses on two fronts: on one side, the photooxidation of EVOH copolymers was studied through identification of the photoproducts, both in the solid and the gas phases; and on the other side, the photochemical behaviour of the EVOH/zeolite nanocomposites was characterised depending on the particle size and the zeolite loading of the materials. This research aims to propose nanocomposite coatings for flexible solar cells taking into account the photostability of the barriers used for encapsulation. 2. Experimental part 2.1. Materials Ethylene vinyl alcohol copolymers (EVOH) containing 27 wt% (EVOH27) and 44 wt% (EVOH44) ethylene, were supplied by Scientific Polymer Products, INC. Most of the results were obtained with EVOH44 as polymer, and the polymer was named EVOH. Dimethylsulfoxide (DMSO) solvent of 99.9% purity was purchased from Sigma Aldrich Ltd (St. Louis, MO) and used without further purification. Synthetic zeolites were kindly supplied by Clariant Produkte (Deutschland) GmbH with the trade names Lucidot NZL 40 LP3533 (NZL) and Lucidot DISC (DISC). Particle specifications provided by the supplier are displayed in Table 1. The chemical composition of zeolite particles (NZL and DISC) was checked by elemental analysis (Service central d'Analyse of the CNRS at Vernaison) (Table 2.). According to the analysis, the zeolites have the following formula: [K9(H2O)16] [Si27Al9O72] with Si/Al ratio close to 3, which is typical for L-type zeolites. The results indicated the presence of iron impurities in both samples, and other elements were detected in trace amount (20 ppm). 2.2. Preparation of EVOH/zeolite nanocomposites and film processing Nanocomposites were prepared by two different methods: either solution casting (method A) or melt-mixing using a twin screw extruder (method B).

Scheme 1. Zeolite structure.

2.2.1. Method A The polymer was dissolved in DMSO at 190 g/l. At the same time, a solution of zeolite particles with concentration of 95 g/l was placed in an ultrasonic bath for 30 min and then treated by an ultrasonic processor (a Vibra-Cell VCX 130 PB) for 5 min at a frequency of 20 kHz and amplitude of 75%, immediately preceding the mixing with polymer solution. After sonication, aliquots of zeolite suspension were added to achieve nanocomposites with 10, 20 and 30 wt% of particles to the fixed volume of EVOH solution. The

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

139

Table 1 Specifications of zeolite particles. Specification

NZL

DISC

Size of primary particles Water content, % Specific surface area (BET) of the hydrated material, m2/g

15e60 nm 10e13 150e170

0.5e2 mm 7e8 50e55

Table 2 Elemental analysis of zeolites. Chemical element

NZL DISC

H (%)

Si (%)

Al (%)

K (%)

Fe (ppm)

Zn (ppm)

Cu (ppm)

Co (ppm)

Mn (ppm)

Na (ppm)

1.20 1.33

24.75 28.87

9.07 8.80

13.31 12.00

63 610

<20 20

<20 20

<20 20

<20 20

20 39

Bold indicates significant amount of impurity.

nanocomposites were sonicated for 5 min using the ultrasonic processor with the same conditions as for the zeolite suspension. Films of pristine EVOH and EVOH/zeolite nanocomposite were obtained with an Erichsen Coatmaster 809 MC film applicator onto glass sheets and left to dry at 40  C overnight. After solvent evaporation, films with thickness between 15 and 40 mm were obtained and then placed in an oven at 60  C under vacuum for one week to remove trace solvents. Any residual presence of the solvent was verified by IR spectroscopy with the IR characteristic bands of DMSO at 950 and 1020 cm
1. 2.2.2. Method B A melt-mixing intercalation process was also used for nanocomposite preparation. Polymer and zeolite powders were meltmixed in a twin-screw extruder (Thermo MiniLab Haake Rheomex CTW5, Germany) for 15 min at 210  C and a screw speed of 40 rpm. Films were then obtained from solid polymer/zeolite nanocomposite pellets using a hydraulic press with 2 heated plates at a working temperature of 200  C and the pressure of 200 bar for 1.5 min. Films with thickness between 30 and 60 mm were obtained. 2.3. Irradiation UVevisible light irradiation (l ¼ 300 nm) was carried out in a SEPAP 12/24 unit from Atlas [33] at 60  C in presence of air. This device is equipped with four medium pressure mercury lamps (Novalamp RVC 400 W) positioned vertically at each corner of the square chamber. The glass envelope filters wavelengths below 300 nm. SEPAP 12/24 has been designed for studying polymer photodegradation in artificial medium-ageing conditions. Irradiations were performed on pristine EVOH, EVOH/zeolite nanocomposite films and on zeolite particles included in KBr pellets.

t ¼ 0 h were normalised for the same absorbance at 1460 cm
1, and normalisation coefficients were applied to corresponding spectra during photooxidation. It is worth noting that this allows for comparison of the degradation rates of the polymer with and without nanoparticles considered as an individual component and does not correspond to the degradation of the nanocomposite material. The resulting data can clearly show the effect of particles on the polymer photostability. UVevisible absorption spectra were performed between 200 and 800 nm using a Shimadzu UV-2600 spectrophotometer provided with an integration sphere for analysing the total light transmission, both direct and diffuse. Transmission electron microscopy (TEM) analysis was carried out in a Hitachi H7650 microscope with 80 kV accelerating voltage and equipped by a Hamamatsu TN camera. TEM observation of zeolite particles was performed on the particle suspensions (95 g/l) into DMSO solution deposited onto a membrane placed on a grid. Nanocomposite films for TEM were prepared by blocking films in cured epoxy resin and cutting them by ultramicrotomy. 2.4.1. Chemical derivatisation treatments Photooxidation products were identified by performing chemical derivatisation treatments, which are based on a selective reaction of the specific chemical groups with a reagent that can consequently be observed in the modifications in the IR spectra [34,35]. The reagents were ammonia (NH3) and 2,4dinitrophenylhydrazine (DNPH). NH3 reacts with carboxylic acids to form ammonium salts and with esters and anhydrides to form amides. NH3 treatment was performed at room temperature in simple flow reactors in a sealed system. 2,4-DNPH reacts with aldehydes and ketones to form a derivative of 2,4dinitrophenylhydrazone. Irradiated films were placed into MeOH solution for 18 h to eliminate low molecular weight oxidation products and were then treated by 2,4-(DNPH) methanol solution.

2.4. Characterisation methods Infrared spectroscopy (IR) analysis was performed using a Nicolet 760 Magna spectrophotometer, working with OMNIC software. Spectra were recorded in transmission mode with 32 acquisitions and 4 cm
1 resolution. The photooxidation kinetics of the polymer and the nanocomposites were followed by IR spectroscopy in the carbonyl domain (1730 cm
1) and the region of CeH vibration (2850 cm
1) by plotting the absorbance variation from initial spectrum (AteAt0) at 1730 and 2850 cm
1 as a function of irradiation time. The absorption band at 1460 cm
1 is a characteristic band of the polymer, and was used to compare the same polymer amount. All spectra at

2.4.2. Gas phase analysis: solid-phase microextraction (SPME) SPME was run to identify volatile photodegradation products. To collect these products, the samples were irradiated in sealed vials containing air. The volatile compounds were absorbed during 2 min at 60  C by CarboxenePDMS fibre (75 mm) purchased by Supelco (Bellefonte, PA, USA). Upon concentration in the fibre, the chemical compounds were injected and thermally desorbed in a 6890 Network System Gas Chromatography (GC) from Agilent Technologies coupled to a 5973 Agilent mass sensitive detector. The GC was equipped with a Supelcowax TM 10 column (30 m  0.25 mm) from Supelco. Desorption into injector at 280  C for 2 min in the splitless mode was performed. The temperature

140

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

programme started from isotherm at 60  C during the first 5 min and then increased up to 200  C at a rate of 10  C/minute, and the final temperature was maintained for 15 min. The temperature of the transfer line was kept at 280  C, and the ions' source temperature was fixed at 230  C. The ionisation was produced by electronic impact with 70 eV energy under the electron ionisation mode. The mass spectra as well as the reconstructed chromatograms were obtained by a mass scanning in the interval m/z ¼ 20e400. 3. Results and discussion 3.1. Characterisation of zeolite particles and EVOH/zeolite nanocomposites 3.1.1. EVOH-zeolite nanocomposite 3.1.1.1. Microscopy analysis. TEM images of both initial zeolite powders (Fig. 1) show the specific morphology of the zeolite nanoparticles NZL (Fig. 1a) and the micrometric size of DISC particles (Fig. 1b). TEM images of cross sections of an EVOH/NZL nanocomposite film (Fig. 2) show that the specific nanoparticles morphology was maintained within the EVOH matrix with a nanometric size of the primary particles connected to each other to form aggregates. Moreover, a good distribution of the particles into the polymer matrix is observed for the films processed by method A (Fig. 2aec) and for particles at different loading 5e20 wt% (not shown.). NZL particles form aggregates in the size range of 90e170 nm for films processed by methods A and B. This might be due to specific interactions in the solvent for the system polymerparticles due to the hydrophilic nature of zeolite in the case of method A. In contrast, for the films prepared by method B, areas with a high concentration of particles with a diameter range of 5e20 mm can be observed. 3.1.1.2. Spectroscopy analysis. Films of pristine EVOH and of EVOH/ zeolite nanocomposites prepared by solution casting (named as method A) with different particle contents (0e30 wt%) were characterised by UVeVisible and IR spectroscopies (Figs. 3e4.). 3.1.1.3. UVevisible analysis. UVevisible spectra of pristine EVOH and EVOH/NZL nanocomposite films with different particle loadings are shown in Fig. 3. Due to the nanometric size and the good dispersion of the filler particles, the EVOH/NZL nanocomposite

films maintain good transparency even with 30 wt% particle loading. This may bring the opportunity to include relatively high percentages of particles into the polymer matrix to increase the water scavenger properties of the material while maintaining the optical properties required for OSCs encapsulation. 3.1.1.4. IR analysis. The characteristic IR band assigned to EVOH [36e40] and to zeolite L-type particles [41e43] have been previously described and are summarised in Table 3. The IR spectra of EVOH/zeolite nanocomposite films (Fig. 4) show the expected contribution of the IR absorption bands that are related to the polymer (2850e3350 cm
1) and to the zeolite structure (430e650 cm
1, 1000e1200 cm
1). The IR bands at 475, 1033 and 1098 cm
1 can be used to quantify the amount of zeolite. As expected, the absorbances are proportional to particle content in the nanocomposite (Fig. 5b), as well as for the films obtained by method A as for those obtained by method B. 3.2. Photooxidation of EVOH at l > 300 nm) 3.2.1. UVevisible analysis Under exposure in the presence of air, the UVevisible spectra of EVOH film show a shift in absorbance to longer wavelengths without any defined maximum (Fig. 5a). The increase of absorbance is limited below 400 nm, and even after 750 h of exposure, the EVOH film does not show any yellowing. 3.2.2. IR analysis Photooxidation of EVOH provokes dramatic modifications of the IR spectra, as shown in Fig. 5b. The main modifications are in the hydroxylated and the carbonylated domains (Fig. 6), with a decrease of ReOH (3340 cm
1) and CeH (2930, 2910, 2855 cm
1) vibration bands. A broad band formed in the carbonyl region with a maximum at 1715 cm
1 and of two weak shoulders with maxima at approximately 1730 cm
1 and 1780 cm
1. These modifications of the infrared spectra observed during exposure of EVOH fit quite well those reported in the case of polyethylene photooxidation [44,45]. Formation of absorption bands at 1640 cm
1 and 910 cm
1 can also be noticed and can be attributed to double bonds, as observed in the case of PE photooxidation [46] (Scheme 2). The analysis by IR spectroscopy seems to indicate that the photooxidation of EVOH is similar to that of PE.

Fig. 1. TEM images of zeolite particles dispersed in DMSO (95 g/l): (a) NZL and (b) DISC.

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

141

Fig. 2. TEM images of EVOH/NZL nanocomposites processed by (aec) melt-mixing in the twin-screw extruder and (def) solution casting.

Fig. 3. UVeVisible spectra of (a) EVOH and (bed) EVOH/NZL nanocomposite films (e ¼ 20 mm) with different zeolite content: (b) 7 wt%; (c) 18 wt%; (d) 30 wt%.

3.2.3. Chemical treatments Chemical derivatisation treatments by NH3 and 2,4-DNPH were performed (Figs. 7e8) on photooxidised EVOH films to help in the identification of the photoproducts in the carbonyl region and to confirm the similarities with the photooxidation of PE. The IR spectra recorded after treatment by NH3 of a photooxidised film shows a decrease of absorbance at 1710 cm
1 correlated with an increase of absorption at approximately 1565 cm
1

(Fig. 7). On the basis of the numerous publications on NH3 derivatisation treatments [47e49], these modifications reflect the reaction of carboxylic acids in the dimer form observed at 1710 cm
1 with NH3 to form carboxylates. Reactions with 2,4-DNPH of either aldehydes or ketones give a derivative of 2,4-dinitrophenylhydrazone [50]. The modifications observed in Fig. 8a indicate the formation of 2,4dinitrophenylhydrazone with the appearance of two new bands at 1618 cm
1 (C]N in hydrazone) and 1592 cm
1 (n(C]C) of aromatic ring) correlating to a small decrease in absorbance at 1717 cm
1 (these modifications are not observed in the case of nonirradiated samples). The formation of 2,4-dinitrophenylhydrazone was also monitored by UVeVisible spectroscopy (Fig. 8b), which shows the formation of a band at 365 nm that is characteristic of the products obtained by reaction of 2,4-DNPH. Photolysis, which is light irradiation in the absence of oxygen, of photooxidised EVOH films was carried out. It is well-known that ketones are unstable under UV-light irradiation and react according to the reactions of Norrish Type [51]. Fig. 9 shows the IR spectrum resulting from the difference between pre- and post-photolysis of EVOH in the carbonyl domain. A decrease of absorbance at 1710 cm
1 can be observed after 20 h of photolysis for photooxidised EVOH film, whereas no modification was observed in the case of non-irradiated samples, which indicates that the photoproducts that have reacted in the treatment with 2,4-DNPH are ketones. Table 4 summarises the IR bands assigned to photoproducts. 3.2.4. Gas phase analysis (SPME) As noted above, EVOH consists of ethylene and vinyl alcohol

142

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

Fig. 4. a) FTIR spectra of pristine EVOH(A) and EVOH/NZL(A) nanocomposites with different particles content (7e30 wt%); b) Absorbance at 1033 cm
1, 1093 cm
1, and 476 cm
1 as a function of particle content in EVOH/NZL(A) nanocomposites.

Table 3 Assignment of infrared bands of EVOH/L-zeolite composites. Bands, cm
1

Attribution

Origin

3350 2929 2852 1460 1098 1035 845 827 763 723 643 609 575 478 435

n(OH) stretching nas(CH2) stretching nsym(CH2) stretching n(OH) skeleton vibrations nas(SieO) stretching

Polymer

Zeolite

n(CH2) rocking

Polymer

Symmetric stretch of internal tetrahedra as well as for external linkages Vibration of tetrahedral structure

Zeolite Polymer þ Zeolite Zeolite

Related to TeO bending mode Pore opening of external linkages

Zeolite Zeolite

Fig. 5. a) UVeVisible and b) FTIR spectra of EVOH(A) film during photooxidation at l > 300 nm.

units. The IR analysis of EVOH film after 750 h photooxidation reveals the oxidation of the PE units. However, previous studies on poly(vinyl alcohol) PVA photooxidation [52] have shown that most of the PVA photoproducts are volatile and consequently might not be detected by IR analysis of the photooxidised film for irradiation time less than 2000 h [52], due to their migration out of the polymer matrix. Gas phase analysis of EVOH during photooxidation

was performed to detect any degradation products coming from the PVA units of EVOH. The composition of the gas phase of EVOH photooxidation was then analysed using SPME. Several compounds were identified using this technique (Table 5), including ketones, lactones and carboxylic acids. To determine the influence of the relative content of ethylene and vinyl alcohol groups on the quantities of detected

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

143

Fig. 6. FTIR subtracted spectra of EVOH film during photooxidation: (a) 3000e2750 cm
1, (b) 1850e1600 cm
1.

Scheme 2. Mechanism of photooxidation of PE [46] and PVA [52] units.

products, SPME experiments were performed on EVOH materials with 44 wt% and 27 wt% of ethylene groups. The results reported in Table 5 show that products such as g-valerolatone, acetic and propionic acids were found in higher amounts in the case of EVOH27, whereas the quantities of acetone, 2-butanone, 2propanol, and 2-pentanone were higher for EVOH44. From previous work carried out by our group [52], the same products as those observed in Table 5 were identified by SPME analysis of the gas phase of photooxidised PVA, except hexan-2,5dione, propanol and pentanone. These three compounds are present in both EVOH copolymers, the last two being detected at higher concentrations in the case of EVOH44, the one with the highest ethylene group content. 3.2.5. Ionic chromatography To propose a more comprehensive mechanism of photooxidation of EVOH, ionic chromatography (IC) was performed to analyse the low-molecular-weight photoproducts which were trapped in

the film. A photooxidised film was soaked in water and the aqueous solution was analysed to identify the extracted molecular photoproducts. By comparing these products with standards, the primary released products were identified as acetic acid, formic acid, malic acid, valeric acid and glutaric acid (Table 6). Where acetic acid and formic acid can be detected in the case of PVA, pentanoic acid, glutaric acid and a-hydroxylated malic acid were observed only in the case of the copolymers EVOH. Their formation can be explained from oxidation of sequences of “PE-PE-PE-PVA” or “PVA-PE-PVA” as described below in Scheme 3. 3.2.6. Photooxidation mechanism A comprehensive mechanism of EVOH photooxidation can be proposed on the basis of all of the detected photoproducts. This mechanism involves photoproducts formed in the case of both the PE units and the PVA units (Scheme 2) but also specific photoproducts that are only formed in the case of EVOH due to alternating sequences of PE units and PVA units (Scheme 2).

144

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

Fig. 7. Derivatisation treatment with NH3 of EVOH films after 155 h photooxidation, FTIR spectra before (black) and after 1 h of NH3 treatment (red), subtracted spectra afterebefore (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

The results obtained by analysis of the solid film by infrared, analysis by SPME of the gas phase molecular photoproducts and analysis of extracts in solution by ionic chromatography show that various carboxylic acids are detected. Other classes of compounds, such as ketones and esters, are also detected but at clearly lower concentrations. The identification of the various products formed gives evidence that the photochemically induced oxidation of EVOH involves the oxidation either on the tertiary (C1) and/or on the secondary carbon (C2 and C3) in a- and b-positions to hydroxyl group (Scheme 3) as well as in the PE units and the PVA units and that the sequences PVA-PE-PVA and PVA-PE-PE also have to be considered. The mechanisms are based on the classical sequences hydroperoxidation/decomposition of hydroperoxides/b-scission of alkoxy radicals, which have been reported many times in the literature to explain the oxidation of organic polymers [53]. The molecular carboxylic acids identified by ionic chromatography are formed by two oxidation steps. Glutaric acid is formed by oxidation on two tertiary carbon atoms, C1 on a PVA-PE-PVA segment unit. The secondary carbon noted (C2) reacts in PVA photooxidation (Scheme 2) and leads to an a-hydroxylated acid, which is then

Fig. 9. FTIR spectrum of difference (afterebefore photolysis) of EVOH film previously photooxidised (380 h).

Table 4 Assignment of infrared bands in the carbonyl domain. Wavenumber (cm
1)

Photooxidation products

Treatment

1715 1717 1730 1785

Carboxylic acids Ketones Esters Lactones and anhydrides

NH3 2,4-DNPH NH3 NH3

followed by an oxidation on a secondary carbon in b-position atom (C3), leading to malic acid after chain scission. Oxidation on a tertiary carbon atom, C1 in a PVA-PE-PE segment unit, leads to a carboxylic function. Oxidation on a C3 carbon atom can give after chain scission to pentanoic acid or propanoic acid depending on which PE unit is oxidised. 3.3. Photooxidation of EVOH/NZL nanocomposites at long wavelengths (l > 300 nm) A similar study of the photochemical behaviour of EVOH/NZL nanocomposites was studied in the presence of oxygen to evaluate the influence of zeolite on the photochemical reactivity of the polymer. To start with, we studied EVOH/zeolite nanocomposites with nanoparticles NZL and at 20 wt% loading and then studied

Fig. 8. a) FTIR and b) UVeVisible spectra of 2,4-DNPH treatment of EVOH: before irradiation (blue); after 180 h of photooxidation (black); after 18 h of soaking in MeOH (red); after 2,4-DNPH treatment (green). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148 Table 5 Photoproducts of photooxidised films of EVOH detected in gas phase by SPME-GCMS. N

1 2 3 4 5 6 7 8

tR

2.3 3.3 3.7 4.9 21.2e22.7 22.3 22.8 23.8e24.0

Compound

Acetone 2-Butanone 2-Propanol 2-Pentanone Acetic acid 2,5-hexanedione Propionic acid Gamma valerolactone

Area, Pct EVOH 44

EVOH 27

17.1 5.8 8.1 3.7 31.5 2.2 2.7 2.9

8.8 3.5 3.9 1.7 51.1 2.5 8.1 9.1

145

both the influence of the zeolite type (NZL or DISC) and the influence of the amount of zeolites (from 5 wt% to 20 wt%). 3.3.1. IR analysis The modifications of the IR spectra during photooxidation of EVOH/NZL nanocomposite film (Fig. 10a) are similar to those of pristine EVOH (Figs. 5e6). This indicates on one hand that zeolite particles are stable under irradiation and on the other hand that the zeolite particles do not modify the photooxidation mechanism of the polymer. 3.3.2. UVevisible analysis UVevisible spectra of EVOH/NZL nanocomposite film (Fig. 10b) show a shift of the absorption front towards the longer

Table 6 Carboxylic acids extracted from irradiated EVOH and detected by ionic chromatography. Compound

Formula

C, mg/l (of extract)

C, mg/g (of solid film)

EVOH 44

EVOH 27

EVOH 44

EVOH 27

Formic acid Acetic acid Pentanoic acid Glutaric acid Malic acid

HeCOOH H3CeCOOH CH3e(CH2)3eCOOH HOOCe(CH2)3eCOOH HOOCeCH2eCHOHeCOOH

14 48 8.4 6.4 10.8

14 38 3.9 6 10.4

1.3 4.5 0.8 0.6 1.0

1.1 3.0 0.3 0.5 0.8

Scheme 3. Mechanism of EVOH copolymer photooxidation.

146

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

Fig. 10. a) FTIR and b) UVeVisible spectra of EVOH/NZL (20%)(A) nanocomposite film during photooxidation.

wavelengths, but it is worth noting that as previously observed for pristine polymer, the absorption does not overlap the domain of the visible light, and as a consequence, the nanocomposite film remains transparent and uncoloured. This is an important property for applications for encapsulation of electronic devices such as OSCs. 3.4. Influence of zeolite on photooxidation rate of EVOH 3.4.1. Influence of zeolite type To investigate the effect of the morphology and the size of the zeolite particles on the polymer photostability, two zeolite particles (DISC and NZL, which are microsized and nanosized, respectively) were included into the polymer matrix (20 wt%) by method A and irradiated in the conditions described above. As performed in the case of pristine EVOH, the oxidation rates of the polymer were evaluated by plotting the increase of absorbance at 1730 cm
1 and the decrease at 2850 cm
1 as a function of irradiation time. Fig. 11 compares the oxidation kinetics of pristine material and EVOH loaded with 20 wt% NZL prepared following procedure A. It is interesting to note that in presence of both zeolite particles, the induction period for oxidation was shorter (100 h vs 150 h) and that the oxidation rate was higher than in the case of pristine EVOH. This result suggests that both zeolite particles have a prodegrading effect on EVOH. In regards to the composite stability in terms of particles type, DISC has a slightly higher prodegrading effect than that of NZL as can be observed in the carbonyl domain (Fig. 11a) or

with the decrease of CeH bands (Fig. 11b). One can also notice a slightly higher oxidation rate in the case of the micro-zeolite (DISC). Such behaviour can be explained by the photoinducing effect of iron impurities that is present in both zeolite particles, with an even higher amount in the case of DISC (10 times higher concentrations). However, the prodegradant effect of iron impurities on the photooxidation of a polymer is not proportional to the amount of iron, since it is expected that iron acts through a catalytic cycle. Iron is known to induce the photodecomposition of hydroperoxides [54e56], producing radicals that can initiate oxidation of the polymer. This effect has been reported in the case of natural clays such as montmorillonite [57e59,26,60,61,27]. Another parameter that has to be taken into account is the specific surface of the particles, since nanoparticles (NZL) have a higher surface interaction than microparticles (DISC). Even if the amount of iron is lower in the case of the NZL nanoparticles, the observed effect on the polymer oxidation reflects very certainly the high specific surface of the zeolite nanoparticles. 3.4.2. Influence of NZL loading The influence of the zeolite loading on the photooxidation rate of EVOH was studied by plotting the increase of absorbance at 1730 cm
1 and the decrease at 2850 cm
1 as a function of irradiation time (as described in Section 3.4.1), for EVOH/NZL nanocomposites with particles at 5 to 20 wt% (Fig. 12). The kinetic curves clearly show that the prodegradant effect of NZL increases with an

Fig. 11. a) formation of carbonylated products and b) a decrease of CeH bands intensity of EVOH and EVOH/composite during photooxidation: pristine EVOH (black), EVOH/ NZL(20%)(A) (red), EVOH/DISC(20%)(A) (green). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

147

Fig. 12. a) formation of carbonylated products and b) decrease of CeH bands absorbance for EVOH(A) and EVOH/NZL(A) nanocomposite films with different particle contents (5; 10; 15; 20 wt%).

increase of NZL particles content. The higher particle amount leads to higher oxidation of the polymer. 4. Conclusion In this paper, we have shown that it was possible to obtain EVOH/zeolite nanocomposites with well-dispersed and homogeneously distributed nanoparticles up to 20 wt% loading by two different methods, solution mixing and melt-mixing. The identification of the oxidation photoproducts of EVOH films (with two different ethylene unit contents, 27% or 44%) under accelerated artificial conditions allowed us to propose a comprehensive mechanism accounting for the photochemical behaviour of EVOH. This mechanism indicates that oxidation of EVOH involves the oxidation either on the tertiary and/or on the secondary carbon in a- and b-positions to hydroxyl group as well in the PE units as in the PVA units and that the sequences PVA-PE-PVA and PVA-PE-PE have also to be considered. We have shown that zeolites are prodegradant and accelerate the photochemical oxidation of EVOH. This effect was attributed to the iron impurities in the zeolite particles. This is a reasonable assignment as iron ions are indeed known to induce the photodecomposition of hydroperoxides, which in turn increases the rate of oxidation of the polymer. This effect is observed in the case of both the morphologies that were investigated (nano and micro). It is more important in the case of the micromorphology because this morphology has a higher iron impurity content. Moreover, as shown in the case of nanoparticles, this effect is proportional to the amount of zeolite in the composite (from 5 wt% to 20 wt%). Further studies for photostabilisation of these polymer/zeolite encapsulants are currently being undertaken to achieve barrier materials for long-lived flexible OSCs. Acknowledgements The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2011) under grant agreement ESTABLIS no. 290022. deutour is warmly thanked for administrative Dr. M. Pe assistance. References [1] F. Samyn, S. Bourbigot, C. Jama, S. Bellayer, S. Nazare, R. Hull, et al., Crossed characterisation of polymer-layered silicate (PLS) nanocomposite

[2]

[3] [4]

[5] [6] [7]

[8]

[9]

[10]

[11]

[12]

[13] [14]

[15]

[16] [17]

[18] [19]

[20]

morphology: TEM, X-ray diffraction, rheology and solid-state nuclear magnetic resonance measurements, Eur. Polym. J. 44 (6) (2008) 1642e1653. F. Samyn, S. Bourbigot, C. Jama, S. Nazare, B. Biswas, R. Hull, et al., Characterisation of the dispersion in polymer flame retarded nanocomposites, Eur. Polym. J. 44 (6) (2008) 1631e1641. P. Kiliaris, C.D. Papaspyrides, Polymer/layered silicate (clay) nanocomposites: an overview of flame retardancy, Prog. Polym. Sci. 35 (7) (2010) 902e958. D.Y. Wang, X.Q. Liu, J.S. Wang, Y.Z. Wang, A.A. Stec, T.R. Hull, Preparation and characterisation of a novel fire retardant PET/a-zirconium phosphate nanocomposite, Polym. Degrad. Stab. 94 (4) (2009) 544e549. T. Ebina, F. Mizukami, Flexible transparent clay films with heat-resistant and high gas-barrier properties, Adv. Mater. 19 (18) (2007) 2450e2453. D.J. Chaiko, A.A. Leyva, Thermal transitions and barrier properties of olefinic nanocomposites, Chem. Mater. 17 (1) (2005) 13e19. A. Zhu, A. Cai, J. Zhang, H. Jia, J. Wang, PMMA-grafted-silica/PVC nanocomposites: mechanical performance and barrier properties, J. Appl. Polym. Sci. 108 (4) (2008) 2189e2196. H.N. Chandrakala, B. Ramaraj, Shivakumaraiah, Joong Hee Lee, Siddaramaiah, Polyvinyl alcohol/carbon coated zinc oxide nanocomposites: electrical, optical, structural and morphological characteristics, J. Alloys Compd. 580 (1) (2013) 392e400. S. Seethamraju, P.C. Ramamurthy, G. Madras, Flexible poly (vinyl alcohol-coethylene)/modified MMT moisture barrier composite for encapsulating organic devices, RSC Adv. 3 (31) (2013) 12831e12838. A. Rivaton, S. Chambon, M. Manceau, J.L. Gardette, N. Lemaître, S. Guillerez, Light-induced degradation of the active layer of polymer-based solar cells, Polym. Degrad. Stab. 95 (3) (2010) 278e284. J. Ahmad, K. Bazaka, L.J. Anderson, R.D. White, M.V. Jacob, Materials and methods for encapsulation of OPV: a review, Renew. Sustain. Energy Rev. 27 (2013) 104e117. R.B. Armstrong, EVOH map (modified atmosphere packaging) for fresh meat and the effects of moisture on oxygen barrier of EVOH, in: TAPPI-polymers Laminations & Coatings Conference Proceedings, 2001. R.H. Foster, Ethylene vinyl alcohol copolymer (EVOH) resins, in: A.A. Tracton (Ed.), Coatings Technology Handbook, third ed., CRC Press, 2005, p. 57-1-5. J.A. Ratto, D. Froio, S. Schirmer, M. Bernacsoni, D. Vanderwalker, C. Thellen, J. Lucciarini, Nanocomposite multilayer films of ethylene co-vinyl alcohol with kaolin for high barrier food packaging, in: 69th Annual Technical Conference of the Society of Plastics Engineers 2011ANTEC-conference Proceedings, 2011, pp. 2e6. S. Schirmer, D. Froio, C. Thellen, M. Mannarino, J. Lucciarini, J.A. Ratto, Multilayer films of ethylene co-vinyl alcohol (EVOH) nanocomposites and polypropylene (PP) for food packaging applications, ANTEC Conf. Proc. 4 (2007) 2324e2328. K. Yano, A. Usuki, A. Okada, Synthesis and properties of polyimide-clay hybrid films, J. Polym. Sci. A Polym. Chem. 35 (11) (1997) 2289e2294. ~ oz, ModificaJ.P. Cerisuelo, J. Alonso, S. Aucejo, R. Gavara, P. Hern andez-Mun tions induced by the addition of a nanoclay in the functional and active properties of an EVOH film containing carvacrol for food packaging, J. Membr. Sci. 423e424 (2012) 247e256. nez, J.M. Lagaron, R. Gavara, J.J. Saura, Development of L. Cabedo, E. Gime EVOH-kaolinite nanocomposites, Polymer 45 (15) (2004) 5233e5238. L. Moyo, W.W. Focke, D. Heidenreich, F.J.W.J. Labuschagne, H.J. Radusch, Properties of layered double hydroxide micro- and nanocomposites, Mater. Res. Bull. 48 (3) (2013) 1218e1227. E.J. Lee, J.S. Yoon, E.S. Park, Preparation and properties of the highly porous poly (ethylene-co-vinyl alcohol)/multiwalled carbon nanotube nanocomposites prepared by a simple saponification method, J. Appl. Polym. Sci. 125 (S1) (2012) E691eE704.

148

I. Topolniak et al. / Polymer Degradation and Stability 121 (2015) 137e148

[21] C.S. Wu, J.Y. Liao, S.Y. Fang, A.S.T. Chiang, Flexible and transparent moisture getter film containing zeolite, Adsorption 16 (2010) 69e74. [22] V. Valtchev, L. Tosheva, Porous nanosized particles: preparation, properties and applications, Chem. Rev. (2013), http://dx.doi.org/10.1021/cr300439k. [23] A.A. Shores, Method of Providing Moisture-free Enclosure for Electronic Device, US Patent 5,401,536, 28 March 1993. [24] M. Erdmann, N. Schall, A. Kirtikar, P-63: getter paste for OLED application, SID Symp. Dig. Tech. Pap. 37 (1) (2006) 436e439. , P.O. Bussie
re, J.L. Gardette, M. Murariu, P. Dubois, [25] S. Therias, J.F. Larche Photochemical behavior of polylactide/ZnO nanocomposite films, Biomacromolecules 13 (10) (2012) 3283e3291. [26] S. Morlat, B. Mailhot, D. Gonzalez, J.L. Gardette, Photooxidation of polypropylene/montmorillonite nanocomposites. Part I: influence of nanoclay and compatibilising agent, Chem. Mater. 16 (3) (2004) 377e383. [27] S. Morlat-Therias, B. Mailhot, D. Gonzalez, Gardette JL photooxidation of polypropylene/montmorillonite nanocomposites. Part II: interactions with antioxidants, Chem. Mater. 17 (5) (2005) 1072e1078. [28] J. Gaume, A. Rivaton, S. Therias, J.L. Gardette, Influence of nanoclays on the photochemical behaviour of poly(vinyl alcohol), Polym. Degrad. Stab. 97 (4) (2012) 488e495. [29] F.P. La Mantia, N. Tzankova Dintcheva, V. Malatesta, F. Pagani, Improvement of photo-stability of LLDPE-based nanocomposites, Polym. Degrad. Stab. 91 (12) (2006) 3208e3213. [30] S. Morlat-Therias, E. Fanton, J.L. Gardette, N. Tzankova Dintcheva, F.P. La Mantia, V. Malatesta, Photochemical stabilization of linear low-density polyethylene/clay nanocomposites: towards durable nanocomposites, Polym. Degrad. Stab. 93 (2008) 1776e1780. [31] S. Bocchini, K. Fukushima, A. Di Blasio, A. Fina, A. Frache, F. Geobaldo, Polylactic acid and polylactic acid-based nanocomposites photooxidation, Biomacromolecules 11 (2010) 2919e2926. [32] M. Morreale, N.T. Dintcheva, F.P. La Mantia, The role of filler type in the photooxidation behavior of micro- and nano-filled polypropylene, Polym. Int. 60 (7) (2011) 1107e1116. [33] J.L. Philippart, C. Sinturel, J.L. Gardette, Influence of light intensity on the photooxidation of polypropylene, Polym. Degrad. Stab. 58 (3) (1997) 261e268. [34] C. Wilhelm, J.L. Gardette, Infrared identification of carboxylic acids formed in polymer photooxidation, J. Appl. Polym. Sci. 51 (8) (1994) 1411e1420. [35] D.J. Carlsson, R. Brousseau, Z. Can, D.M. Wiles, Identification of Products from Polyolefin Oxidation by Derivatization Reactions. Chemical Reactions on Polymers, American Chemical Society, 1988, pp. 376e389. [36] K. Holland-Moritz, D.O. Hummel, Structural studies on linear aliphatic polyesters by electron diffraction, infrared, and Raman spectroscopy, J. Mol. Struct. 19 (1973) 289e300. [37] H. Jiang, P. Wu, Y. Yang, Variable temperature FTIR study of poly(ethylene-covinylalcohol)-graft-poly(ε-caprolactone), Biomacromolecules 4 (5) (2003) 1343e1347. [38] K. Holland-Moritz, H.W. Siesler, Infrared spectroscopy of polymers, Appl. Spectrosc. Rev. 11 (1) (1976) 1e55. [39] G.S. Hu, Z.Y. Ding, Y.C. Li, B.B. Wang, Crystalline morphology and melting behavior of nylon11/ethylene-vinyl alcohol/dicumyl peroxide blends, J. Polym. Res. 16 (3) (2009) 263e269. nez, J.J. Saura, Degradation of high barrier ethyleneevinyl [40] J.M. Lagaron, E. Gime alcohol copolymer under mild thermal-oxidative conditions studied by thermal analysis and infrared spectroscopy, Polym. Int. 50 (6) (2001) 635e642. [41] B.I. Shikunov, L.I. Lafer, V.I. Yakerson, I.V. Mishin, A.M. Rubinshtein, Infrared spectra of synthetic zeolites, Bull. Acad. Sci. USSR Div. Chem. Sci. 21 (1) (1972) 201e203. [42] A.J.M. de Man, R.A. van Santen, The relation between zeolite framework

structure and vibrational spectra, Zeolites 12 (3) (1992) 269e279. [43] M.F. Edith, K. Hassan, A.S. Herman, Infrared structural studies of zeolite frameworks, in: Molecular Sieve Zeolites-I, American Chemical Society, 1974, pp. 201e229. [44] S. Morlat-Therias, E. Fanton, J.L. Gardette, N.T. Dintcheva, F.P. La Mantia, V. Malatesta, Photochemical stabilization of linear low-density polyethylene/ clay nanocomposites: towards durable nanocomposites, Polym. Degrad. Stab. 93 (10) (2008) 1776e1780. [45] M. Sugimoto, A. Shimada, H. Kudoh, K. Tamura, T. Seguchi, Product analysis for polyethylene degradation by radiation and thermal ageing, Radiat. Phys. Chem. 82 (2013) 69e73. € ldes, B. Puk [46] M. Gardette, A. Pertue, J.L. Gardette, T. Janecska, E. Fo anszky, S. Therias, Photo- and thermal-oxidation of polyethylene: comparison of mechanisms and influence of unsaturation content, Polym. Degrad. Stab. 98 (11) (2013) 2383e2390. [47] J.F. Pilichowski, J. Lacoste, J. Mallegol, E. Michel, F. Delor, Modification de
res par traitement gazeux e I. Polyme
res alcools, polyme
res acides, polyme Eur. Polym. J. 33 (10e12) (1997) 1591e1599. [48] A. Rivaton, Photochemistry of poly(butyleneterephthalate): 2 e identification of the IR-absorbing photooxidation products, Polym. Degrad. Stab. 41 (3) (1993) 297e310. [49] M. Manceau, A. Rivaton, J.L. Gardette, S. Guillerez, N. Lemaître, The mechanism of photo- and thermooxidation of poly(3-hexylthiophene) (P3HT) reconsidered, Polym. Degrad. Stab. 94 (6) (2009) 898e907. [50] D. Lin-Vien, N.B. Colthup, W.G. Fateley, J.G. Grasselli, Double bonds containing nitrogen atoms, in: The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules, Academic Press, San Diego, 1991, pp. 191e211. [51] S.K.L. Li, J.E. Guillet, Photochemistry of ketone polymers. Photodegradation of an amorphous ethylene-propylene copolymer, Macromolecules 17 (1) (1984) 41e50. [52] J. Gaume, P. Wong-Wah-Chung, A. Rivaton, S. Therias, J.L. Gardette, Photochemical behavior of PVA as an oxygen-barrier polymer for solar cell encapsulation, RSC Adv. 1 (8) (2011) 1471e1481. [53] J.L. Gardette, A. Rivaton, S. Therias, Photodegradation Processes in Polymeric Materials. Photochemistry and Photophysics of Polymer Materials, John Wiley & Sons, Inc., 2010, pp. 569e601. [54] C. David, M. Trojan, A. Daro, W. Demarteau, Photodegradation of polyethylene: comparison of various photoinitiators in natural weathering conditions, Polym. Degrad. Stab. 37 (1992) 233e245. [55] Z. Osawa, Role of metals and metal-deactivators in polymer degradation, Polym. Degrad. Stab. 20 (3e4) (1988) 203e236. [56] B. Ranby, J.F. Rabek, Photodegradation, Photooxidation and Photostabilization of Polymers, Interscience Publishers, New York, 1975. [57] G. Scott, H.H. Sheena, A. Harriman, Mechanisms of antioxidant action: a photochemical study of transition metal dibutyl dithiocarbamates (and their photo-stabilizing role in a polyolefin matrix), Eur. Polym. J. 14 (12) (1978) 1071e1077. [58] A. Tidjani, C.A. Wilkie, Photo-oxidation of polymeric-inorganic nanocomposites: chemical, thermal stability and fire retardancy investigations, Polym. Degrad. Stab. 74 (1) (2001) 33e37. rard, Photo[59] B. Mailhot, S. Morlat, J.L. Gardette, S. Boucard, J. Duchet, J.F. Ge degradation of polypropylene nanocomposites, Polym. Degrad. Stab. 82 (2) (2003) 163e167. [60] H. Qin, C. Zhao, S. Zhang, G. Chen, M. Yang, Photo-oxidative degradation of polyethylene/montmorillonite nanocomposites, Polym. Degrad. Stab. 81 (3) (2003) 497e500. [61] T.O. Kumanayaka, R. Parthasarathy, M. Jollands, Accelerating effect of montmorillonite on oxidative degradation of polyethylene nanocomposites, Polym. Degrad. Stab. 95 (4) (2009) 672e676.