Radiochemical ageing of butyl rubbers for space applications

Polymer Degradation and Stability 98 (2013) 682e690

Contents lists available at SciVerse ScienceDirect

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

Radiochemical ageing of butyl rubbers for space applications M. Smith a, b, *, S. Berlioz a,1, J.F. Chailan a, 2 a b

Laboratoire Matériaux Polymères Interfaces Environnement Marin (MAPIEM EA 4323), Université du Sud Toulon Var, ISITV, BP 56, 83162 La Valette du Var Cedex, France CNES (Centre National d’Etudes Spatiales), 18 Av. Edouard Belin, 31401 Toulouse Cedex 9, France

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 July 2012 Received in revised form 12 October 2012 Accepted 14 October 2012 Available online 5 November 2012

The effect of 60Co gamma irradiation in inert atmosphere at 25
C and 70
C on a butyl elastomer filled with carbon blacks has been investigated by SEM, 13C NMR, swelling measurements and mechanical tests. An increase of the tensile strength of the material was observed during ageing. This increase of the mechanical properties is due to a modification of the rubber network structure. This hypothesis was confirmed by swelling measurements done before and after ageing at various irradiation doses associated with an estimation of the crosslink density by the FloryeRehner equation. Crosslinking and chains scission reactions occurred under irradiation and the contribution of both processes was estimated thanks to the CharlesbyePinner equation. The high level of carbon blacks in the butyl rubber formulation plays also an important role in the degradation process. Correlations between mechanical properties and crosslink density are also presented. The modification of the rubber mechanical properties underlines that the degradation mechanism is strongly influenced by the temperature especially at high irradiation doses. At 25
C, the chain crosslinking process predominates over the chain scission reactions whereas the two phenomena are in competition at the ageing temperature of 70
C. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Butyl rubber Gamma irradiation Temperature Inert atmosphere Mechanical properties 13 C NMR

1. Introduction Over the years, synthetic rubbers have being used in a variety of technical applications and so in different conditions which widely vary and are often complex due to the number of potential stresses. Butyl rubber is a copolymer of isobutene and isoprene which is well known for its strong capability of energy dissipating combined with a good chemical stability and a good moisture ozone resistance [1,2]. Butyl rubbers and particularly their halogenated derivatives have recently received increasing interests thanks to their remarkable damping properties. Halogenated butyl rubbers could also be used for anti-vibrational systems in satellite applications. Therefore, these rubbers should fulfill the space technology requirements during all the satellite lifetime, which varies between 5 and 10 years. The space environment is especially harsh towards elastomers [3e6]. In most cases, radiations can be considered as the most important factor leading to the degradation of the rubber

* Corresponding author. Laboratoire Matériaux Polymères Interfaces Environnement Marin (MAPIEM EA 4323), Université du Sud Toulon Var, ISITV, BP 56, 83162 La Valette du Var Cedex, France. Tel.: þ33 494142628. E-mail addresses: [email protected] (M. Smith), [email protected] (S. Berlioz), [email protected] (J.F. Chailan). 1 Tel.: þ33 494142976. 2 Tel.: þ33 494142462. 0141-3910/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.polymdegradstab.2012.10.013

properties. Damaging effects can obviously be aggravated by the other environmental factors (temperature, vacuum.). The use of rubbers in so severe conditions requires to have a reliable experimental method to obtain detailed data concerning the evolution of the material properties arising from this environment and so insure the durability of the satellites components. This necessitates accelerated degradation tests at laboratory scale in order to predict the material properties changes in service conditions. In oxidative atmosphere, radiation effects on butyl rubbers have been studied by various authors [7e11], but only a limited number of works has been reported on the irradiation of butyl rubbers in inert atmosphere [9]. When polymers are subjected to gamma radiations, many chemical reactions may occur affecting the physical and mechanical properties of the material [10,12,13]. The primary event which occurs when a molecule interacts with ionizing radiation involves the ejection of an electron to form a radical which could lead to macromolecular chain scissions, crosslinkings, changes in stereochemistry or formation of grafts through complex chemical reactions processes. The role of the temperature on the degradation process is crucial for exposed materials [12] and should be considered carefully. The simultaneous action of high temperature and radiations on rubber materials could lead to dramatic modifications of the material properties since the heat treatment alone leads to thermal degradation in many cases [14]. Concerning the

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

extent of degradation, it is difficult to make predictions since all the reactions which occur at room temperature, benefic as well as detrimental, may occur at elevated temperatures with higher rates. Hence, there is no general rule to predict the behavior of a specific rubber submits simultaneously to radiation and temperature and each case must be studied independently. Many techniques are used in order to investigate the resulting changes of the material and particularly the molecular structure (FTIR) [15,16], the thermal stability and the glass transition behavior (DSC, TGA) [17,18] or the mechanical properties (Dynamic Mechanical Analyses (DMA), tensile tests.) [19e21]. The chain scission process usually leads to a reduction of the tensile strength whereas the crosslinking process results in an increased tensile strength and a reduced elongation at break [22]. The predominance of one phenomenon above the other depends on several factors among them macromolecular structure, material composition, irradiation conditions (absorbed dose, dose rate and exposure environment), temperature, atmosphere . Butyl rubber is a copolymer composed of isobutylene units with a small percentage (1%) of isoprene units. It is known that polyisobutylene undergoes predominantly scission under irradiation [14,23,24]. In butyl rubber, crosslinking would be expected trough the isolated isoprene units but Hill al. pointed out that scissions of the isobutylene sequences predominate [23]. Chandra et al. [25] observed the same effect of ionizing radiations and reported the evolution of the molar mass, tensile strengths and density of butyl rubbers submit to gamma irradiation at 25
C. However, the mechanism of radiation-induced scission of this polymer is not completely understood. FTIR spectroscopic studies carried out by Turner and Higgins indicated the formation of vinylidene double bonds, ethyl groups and tri-substituted double bonds, accompanied by a drop in the substituent methyl groups content, indicating a methyl abstraction. They proposed a mechanism for the chain scission which implies that one vinyl group is formed for each scission [26]. This mechanism was confirmed by Bremer [24] who carried out Quantitative NMR analyses to determine the radiation yield of scission and assumed that every main chain scission results ultimately in both an unsaturated and a saturated chain end. However, it restrains this conclusion to relatively low radiation dose. While butyl rubber is known to undergo predominantly chain scission during exposure to high energy radiation, a drastically different response towards high energy radiation has been found for bromobutyl rubbers [23]. In bromobutyl rubber, carbonehalogen bonds are weaker than carbonecarbon and carbonehydrogen bonds, and the main effect of radiation is the carbonehalogen bond break, giving an organic free radical. Carswell-Pomerantz et al. [27] who studied the degradation of bromobutyl rubbers irradiated with high energy radiation through Electron Spin resonance characterizations pointed out that most of the radicals are generated on the halogenated isoprene units, due to the labile CeBr bound and they observed that crosslinking predominates over chain scission up to a limiting dose of 50e100 kGy. The studies concerning the degradation of bromobutyl rubber under radiation found in the literature reported complex phenomena with radicals formation which could lead to chains scissions, crosslinking or other reactions [27]. These reactions occur simultaneously and make difficult to visualize a distinct trend as many parameters are implied in the predominance of one reaction above the others: environment, dose, additives, temperature, curing agents and curing conditions (number of sulfur per bridge). Moreover, the materials properties as tensile strength for example are a complex function of crosslink chemical density and crosslink nature, both chemical and physical interactions involved. The aim of this paper is to evaluate the behavior of a halogenated butyl rubber submit to gamma radiations and temperature

683

ageing. The morphological, chemical and mechanical properties modifications are studied using classic tensile tests, Dynamic Mechanical Analysis (DMA), Scanning Electron Microscopy (SEM) and Nuclear Magnetic Resonance (NMR) investigations. A better comprehension of the physico-chemical mechanisms involved during ageing under simultaneous action of ionizing radiation and temperature is targeted.

2. Experimental 2.1. Material A butyl rubber (LanxessÒ) was mixed with 66 phr of reinforcing carbon blacks (CabotÒ). Two carbon blacks were used: 60 phr of Medium Thermal (MT) and 6 phr of Semi Reinforcing Furnace (SRF) with an average particle radius of 201e500 nm and 61e100 nm respectively. Chemical processing additives were added to the compound and the mixture was then cured with sulfur as vulcanizing agent. The glass transition temperature of vulcanizated rubber was 64
C  0.5
C, measured by Differential Scanning Calorimetry (DSC). The sheets were compression molded in an electrically heated press at a temperature of 170
C for 14 min.

2.2. Irradiation of the sample A series of gamma irradiation experiments were performed using a 60Co source at the SCK. CEN. Center (Belgium). Five different doses (5, 10, 50 and 100 kGy) were applied under argon atmosphere at room temperature and at 70
C. The dose rate was fixed at 500 Gy/h for all selected doses. These conditions were chosen in order to extrapolate the effects of irradiations to which the material will be subjected during its service life.

2.3. Scanning Electron Microscopy (SEM) The dispersion of the carbon black fillers and other additives before and after irradiation ageing was studied by Scanning Electron Microscopy. Samples were freeze-fractured before analysis and coated with gold to increase the conductivity. They were examined using a Zeiss Supra 40 VP Field Emission Scanning Electron Microscope with a low acceleration voltage (1e5 keV) in order to minimize the degradation caused by the electron beam on the rubber.

2.4. Mechanical and dynamic mechanical properties The tensile properties of unaged and irradiated rubbers were carried out on an MTS DY35 testing machine at 25  2
C according to the NF T46-002 specifications. The crosshead speed was fixed to 500 mm min-1 and five separate measurements were recorded and averaged to assure the reliability of the results and to obtain mean values of the tensile strengths at 50 and 100% of elongation and of the stresses and strains at break. Dynamic mechanical analysis was performed on a TA Instruments DMA 2980 analyzer. A single cantilever tool was used at a frequency of 10 Hz with oscillation amplitude of 6 mm. The temperature range was from 100 to 40
C with a 3
C min1 heating rate. The sample dimensions were approximately 4e5 mm thick, 17.5 mm long and 15 mm wide. The evolutions of tan d and storage modulus G0 versus temperature were recorded.

684

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

2.5. Crosslink density measurement

The CharlesbyePinner theory relates the sol fraction with the radiation dose by the following Eq. (8):

Small fractions of rubber samples exposed at different ageing levels were immersed in cyclohexane at 25  2
C in order to evaluate the crosslink density. Unaged samples were also tested in order to have a reference value. Each presented value was obtained by the measurement of five samples with nearly equal dimensions and weights. At various intervals, specimens were removed, gently wiped with a dry paper to remove solvent excess at the sample surface and reweighted after 30 s. Swelling ratios were calculated after an immersion of one week, which corresponds to the equilibrium swelling time. For the solegel fractions estimation, samples weights were also measured after drying at 60
C in vacuum. The swelling ratios were calculated using Eq. (1):

Dmð%Þ ¼

ðswollensample weightinitial sample weightÞ 100 initial sample weight (1)

The crosslink density is defined as the number of elastically active network chains totally included in a perfect network per unit volume and evaluated according to the following FloryeRehner Eq. (2) [28]:

n ¼

i h  ln ð1  Vr Þ þ Vr þ cVr2   1 Vr V  Vr3  2

Vr ¼   Xr

rr

Xr

rr þ

Xs



(3)

rs

(4) (5)

In order to quantitatively evaluate the crosslinking and chain scission yields of irradiated samples, [s þ s1/2] vs. [1/absorbed dose] from the CharlesbyePinner equation (Eq. (8)) were plotted for the different aged samples. The sol (s) and gel (g) fractions of the irradiated rubbers were calculated by Eqs. (6) and (7):

s ¼ 1g g ¼

dry sample weight initial sample weight

2.6. Nuclear Magnetic Resonance (NMR) Solid State NMR spectra were carried out on a Brucker Avance 400 MHz spectrometrer with a standard CP/MAS probe for 4 mm external diameter rotors using magic angle spinning rotation frequency of 10 kHz. The crosspolarization technique was applied with short contact times (1 ms). Small squares (0.5e1 mm2) of rubber were cut and set in NMR rotor tube. All tests were performed at room temperature. The contribution of each different carbon was determined as the integration of the corresponding NMR peaks normalized by the whole spectra area.

SEM was used in order to evaluate the surface morphological and aggregates dispersion modifications during ageing. An SEM image of unaged butyl rubber cross-section showing the good fillers distribution in the matrix before ageing is presented in Fig. 1A. The samples irradiated at 25
C (Fig. 1B) showed minor cracks of submicron size in one or two places. The samples exposed to radiation at 70
C (1 C) present a surface texture quietly different from the samples irradiated at room temperature with major cracks, wrinkles and fractures. In other studies, the same damages were observed on bromobutyl rubber submit to thermal ageing [32]. Concerning the fillers, no measurable changes of their size or their distribution in the matrix were detected by SEM. 3.2. Mechanical properties

ðswollen sample weightðgÞ  original sample weightðgÞÞ swollen sample weightðgÞ

Xr ¼ 1  Xs

where p is the scission density per unit dose in kGy, q is the density of crosslinks per unit dose in kGy, D is the radiation dose in kGy and u1 is the number of average degree of polymerization. p/q is the intercept of the straight line in the plot and gives an idea of the ratio of chain scission to crosslinking [31].

3.1. SEM

where Xr is the mass fraction of rubber, rr is the density of raw rubber (1062 g cm3), Xs is the mass fraction of cyclohexane and rs is the density of cyclohexane (0.779 g cm3). Xr and Xs can be obtained using Eqs. (4) and (5):

Xs ¼

(8)

3. Results

 

pffiffi p 10 s ¼ þ q qDu1

(2)

where Vr is the volume fraction of the rubber in the swollen sample, c is the interaction constant characteristic between butyl rubber and cyclohexane, which is 0.44 [29], V is the molar volume of cyclohexane (108,03 cm3 mol1) and n is the crosslink concentration (mol cm3). To solve Eq. (2) [30], the rubber volume fraction in the swollen sample, Vr, is calculated by the following Eq. (3):



sþ

(6)

(7)

3.2.1. Effect of radiation dose and temperature Typical stress/strain curves obtained for unaged sample (t0) and samples irradiated at 100 kGy under 25 and 70
C are presented in Fig. 2 and the averaged values of tensile strengths at 50% and 100% of elongation are given in Table 1. The mechanical properties of polymers are considerably modified by high energy radiations. The strains and stresses at break decrease deeply with the irradiation dose, especially at the highest ageing temperature. Hence, the unaged sample and samples irradiated at 100 kGy at 25
and 70
C present stresses of respectively 9.4 MPa, 4.1 MPa and 2.5 MPa. The presence of cracks and fractures revealed by SEM analyses could be one of the explanations for this decrease. Fig. 3 illustrates the variation of the tensile stresses at 50 and 100% of elongation as a function of the irradiation dose for ageing performed at 25
C and 70
C. Between 0 and 5 kGy, the variation of the stresses for both levels of strain are in the standard deviation whatever the temperature of ageing whereas at higher gamma irradiation doses stresses were greatly increased. Hence, at 100 kGy, the stress at 100% of elongation goes up by 137%. At 70
C, the increment of the stress is lower (less than 25%).

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

685

Fig. 2. Tensile stresses vs strains for unaged and 100 kGy irradiated under 25 and 70
C samples.

The main relaxation peak (a relaxation) is associated with the butyl rubber glass transition, while the shoulder corresponds to a subglass relaxation (so-called b relaxation) due to the rotation of methyl groups, directly attached to the polymer backbone [20,33]. The figure revealed a shift of the Ta peak towards higher temperatures (18
C to 15
C) for 100 kGy irradiated sample at 25
C. No significant difference between the glassy G0 values (90
C) of the unaged and 100 kGy/25
C irradiated samples is observed whereas the G0 value of sample irradiated at 70
C is lowered by 40%. At the rubbery state (above 20
C), the rubber modulus does not seem to be affected by ageing. This last observation is consistent with the results of tensile tests done at room temperature which show a superposition of the stress/strain curves at low deformation (<0.5%) for all the samples (Fig. 2).

Fig. 1. SEM micrograph of unaged (A), irradiated at 100 kGy under 25
C (B) and 70
C (C) butyl rubbers.

The dynamic mechanical properties are also influenced by the radiation/temperature ageing, but to a lesser extent than tensile properties. Fig. 4 shows the tan d and G0 versus temperature curves for the unaged sample and for the 100 kGy irradiated samples at 25
C and 70
C. The tan d peak of unaged butyl rubber is pretty wide covering a temperature range between 50 and 50
C, with a maximum at 18
C (Ta) and a shoulder peak near 35
C (Tb).

3.2.2. Failure envelope, stress at breakestrain at break (sreεr) Tensile tests have been widely used by a number of workers in order to study the mechanical properties of rubber materials. From these investigations, Lassiaz [34] proposed the concept of a failure envelope specific to each rubber. In this concept, the locus of the stress at break versus strain at break can be represented as a unique curve independent of the material history. Fig. 5 shows that the locus of the stress and strains at break of the aged samples are superimposed on the tensile curve obtained for the unaged one, that is conform to the failure envelope concept. The failure envelope (sr ¼ f(εr)) shows two distinct zones designed as A and B (Fig. 6) [34]. The A zone represents the incubation phase where the rubber properties do not change significantly compared to the unaged sample. The B zone so-called “the terminal phase” corresponds to samples exposed at higher doses. In this terminal phase, the stresses and strains at break of the material are dramatically lowered. The accumulation of “defects” (fracture, porosity, cracks.) and damages induced by irradiation leads to an earlier failure of rubber that corresponds to a transition from a plastic to a brittle behavior. Radiations transform the flexible samples to brittle and rigid ones especially when high doses are applied. The temperature at which

Table 1 Averaged values of tensile strength at 50% and 100% of elongation (MPa). Unaged 50% 100%

25 70 25 70




C
C
C
C

0.59  0.01 0.79  0.01

5 kGy 0.6 0.6 0.9 0.8

   

0.02 0.01 0.01 0.01

10 kGy 0.7 0.6 1.0 0.8

   

0.01 0.01 0.01 0.03

50 kGy 0.8 0.6 1.6 0.9

   

0.01 0.01 0.02 0.02

100 kGy 0.9 0.5 1.9 1.0

   

0.02 0.01 0.06 0.02

686

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

Fig. 5. Tensile curve of unaged sample and failure envelope (seε). Fig. 3. Evolution of stresses at 50% and 100% of elongation with irradiation dose (kGy) under 25
C (blue: d d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

radiations were performed does not play a major role in this mechanical behavior transition. Similar drop in the mechanical properties were observed by Scagliusi et al. [35] when gamma irradiation were performed on a sulfur cured chloro butyl rubber.

3.3. Network structure studies The calculation of the crosslink density from the swelling test is one of the most used methods to characterize the morphology of a vulcanizated elastomer. Fig. 6 presents the swelling ratios and the resulted crosslink densities obtained by the FloryeRehner equation as a function of the dose for samples irradiated at 25
C and 70
C. The swelling ratios are found to decrease for both studied temperatures with increasing the irradiation dose. Fig. 6 reveals that the swelling ratio of samples irradiated at 25
C decreases sharply up to 10 kGy. Above this absorbed dose, the decreasing trend becomes slower with the additional irradiation dose. Concerning the samples irradiated at 70
C, the swelling ratio decreases gradually with the irradiation dose. The comparison between the swelling ratios evolutions for both temperatures suggests a faster and greater crosslinking phenomenon at 25
C than at 70
C. 3.4. Structure/properties relationships The comparison between Table 1 and Fig. 6 shows an apparent qualitative dependence of the mechanical properties with the crosslinking density and various correlations might be deduced. Hence, Fig. 7 illustrates the relationships between the measured tensile stresses and the crosslink density n. The experimental data could be fitted with a good yield (R2  0.91) by analytical curves of the following form:

s ¼ a  ln n þ b

(9)

where a and b are constants.

Fig. 4. Plot of dynamic mechanical properties (tan d and G0 ) vs temperature for unaged sample (black: d) and 100 kGy irradiated samples at 25
C (blue: d d) and 70
C (red: - - -). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Swelling content and crosslink density (n) measured for samples irradiated at different dose levels under 25
C (blue: d d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

Fig. 7. Evolution of stress at 50% and 100% of elongation vs logarithm of crosslink density (n) at 25
C (blue: d d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

This empirical equation shows that there is a strong correlation between the stresses measured at different levels of elongation and the morphological properties of the rubber. Although correlation does not imply causality, this fact seems to suggest that the increment of the mechanical properties could be partly due to the increment of the crosslink density. The comparison between the “a” values indicates that the effect of the crosslink density is greater at 100% of elongation compared to 50% of elongation Besides, the stresses measured for samples irradiated at 25
C, are higher than those measured for samples irradiated at 70
C if we consider the same crosslinking density. This last observation suggests that the modification of the crosslinking density is not the only explanation for the evolutions of the mechanical properties during ageing (Fig. 2 and Table 1). Other phenomena such as defects accumulation in the sample or increment of the network heterogeneity with ageing could explain the observed differences between the two ageing temperature results. 3.5. CharlesbyePinner theory The scission/crosslinking ratio was analyzed using the CharlesbyePinner equation. The values of p/q could be measured from the CharlesbyePinner plot (Fig. 8): 0.7 and 1.33 were found respectively for ageing at 25
C and 70
C. These values are close to 1 that implies that competitive chain scissions occur in parallel to crosslinking during irradiation. However, the crosslinking phenomenon seems to predominate over chain scission at 25
C whereas the opposite trend seems to occur at 70
C. It is worth to

Fig. 8. CharlesbyePinner plot for ageing at 25
C (blue: d d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

687

note that these results should be considered with precaution due to the shortage of the experimental data’s at high doses for CharlesbyePinner extrapolation. Moreover, these calculations are based on several assumptions, including random spatial distribution of the scission and crosslinking and absence of intramolecular crosslinking or cyclization. According to the CharlesbyePinner results, the density of the elastomeric network seems not to be deeply modified as a result of the strong competition between crosslinking and chains scissions. However, the mechanical properties at high level of deformation are greatly affected by irradiation. As discusses previously (x 2.4), other phenomena could lead to modification of the material mechanical properties. At this stage, it should be reminded that the macroscopic properties of a filled rubber and especially the mechanical properties are not only due to the chemical elastomer network but also to the presence of physical or chemical bonds between fillers and macromolecular chains of elastomer. The high content of carbon blacks in the studied samples (66 phr) allows thinking that these interactions should have an important role. Hence, the modification of the fillers/polymer interactions during irradiation could be one of the major explanations of the observed loss of the mechanical properties. The degradation by creation of cracks, fractures and porosity as shown in SEM images (Fig. 1) is another hypothetic explanation. 3.6. NMR results The NMR spectrum of bromobutyl rubber has been assigned previously by Chu et al. [36]. As the majority of the monomer units are isobutylene, the 13C NMR spectrum (Fig. 9) is practically similar at those of a polyisobutylene and no signal related to isoprene units could be seen in our spectrum. The chemical shifts of the main peaks are reported in Table 2. As presented in Table 2, an increase of the vinyl carbon content (located between 110 and 150 ppm) accompanied by a decrease of the intensities of the signals assigned to quaternary (36 ppm) and methyl carbons (58 ppm) have been measured (Fig. 10). According to the previous reported studies, these evolutions suggest that chain scissions of the isobutylene units occurred during irradiation (Fig. 11) [24]. 4. Discussion The data obtained by mechanical analyses indicate clearly that the irradiation process at room temperature acted effectively in

Fig. 9. NMR spectra of unaged butyl rubber (black: d) and 100 kGy irradiated butyl rubber at 25
C (blue: d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

688

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

Table 2 13 C NMR signal assignments for butyl rubber. Signal (ppm)

Band assignment

150e110 58 36 31.6 30

Unsaturation (C]C) Carbon atom of the methyl group Quaternary carbon atom bearing two methyl groups Vulcanization control band Carbon atom of the methylene group

raising up the tensile stress values of the butyl rubber (Fig. 3) whereas, the evolution of the tensile stresses for samples irradiated at 70
C is less marked. As regards the elongations at break, they are deeply diminished during ageing whatever the temperature considered (Fig. 5). These evolutions of stresses and strains at failure are classically observed when an increment of the crosslink density of the materials is observed. However, the results of NMR presented previously seem to turn in favor of chains scissions. So another explanation must be proposed. The mechanical properties of a filled rubber are strongly dependant not only on the chemical crosslink density ratio but also on the physical and chemical attachments of the molecular chains to the filler surface. Hence, a poor dispersion of particles in a polymeric matrix with large agglomerates leading to low specific surfaces could be responsible of the decrease of mechanical properties such as ultimate stresses [37] and have certain other detrimental effects (reduced product life, poor performance in service.). This hypothesis as an explanation of the mechanical properties drop should be rejected in our study as a good level of dispersion of fillers was observed by SEM for irradiated samples (Fig. 2). However a modification of the number and/or the intensity of the interactions between carbon black particles and rubber is expected under radiation [38e40]. Although there is no complete knowledge of the exact nature of rubberefiller interactions at present, the different types of chemical groups that exist at the carbon black surface such as carboxylic, phenolic, hydroxylic, aldehydic and ketonic, could participate in the formation of physical as well as chemical attachments at the filler/rubber interface. Under high energy irradiation, the surface chemistry of carbon blacks are greatly modified with among others things, the formation of a large number of free radicals, that could induce major changes in the nature of the fillers/rubber bonds [41]. It is worth to note that these bonds play a more important role at high levels of elongation i) where the fillers maybe undergo a damage mechanism leading to carbon black decohesion and ii) because of the better distribution of the carbon black fillers under high

Fig. 10. Evolution of CH2 groups, CH3 groups, quaternary carbon and unsaturation during radiochemical ageing at 25
C (blue: d d) and 70
C (red: d). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 11. Chain scission reaction between two isobutylene units.

deformation levels compared to a more aggregated structure observed without stress [42]. The observed decrease of the elongation at break could hence be explained by an additional bounding between rubber and fillers initiated by the formation of free radicals during radiation. Concerning swelling results, the restriction of the amount of solvent which could be absorbed by an elastomeric network is due to the formation of new crosslinks in the elastomer network or new interactions between rubber and fillers during radiation. The more significant swelling ratio drop for ageing at 25
C is a sign of a higher increase of the crosslink density at this temperature. This result is consistent with the higher tensile stresses measured for samples aged at 25
C than those of samples aged at 70
C (Fig. 2). As regard the dynamic mechanical properties, the Ta shift towards higher temperatures associated with a diminution of a peak amplitude indicates that the macromolecular chains mobility was diminished with irradiation. This phenomenon is consistent with the decrease of the swelling content values (Fig. 6) and could be explained by an increase of the chemical crosslink density or an increase of the physical attachment between matrix and fillers during irradiation. Indeed, crosslinking or bonds between carbon black particles and rubber hamper the segmental motion leading to a higher temperature to initiate movements. At the rubbery plateau, the G0 values for samples irradiated at 25 and 70
C are similar to the one of the unaged sample. In revenge, at the glassy state, a slight decrease of the G0 modulus is observed for samples irradiated at 70
C. The predominance of macromolecular chains scission phenomenon vs crosslinking at this temperature is probably an explanation for this lower value, and consistent with the difference of crosslinking/chains scissions ratio between the two temperatures estimated by CharlesbyePinner equation. Whereas the tensile tests show great effect of the temperature under which the irradiation was carried out, the DMA analyses do not show a great difference between the two ageing temperatures. This could be attributed to the level of solicitation amplitude between the two techniques. A high deformation level (and so a strong solicitation of the bonds between carbon blacks and polymer chains) is imposed to the samples during tensile tests while the deformation applied in DMA is very small (z0.15%). This remark supports the hypothesis that the nature of interactions between carbon black fillers and rubber are greatly modified under radiation whereas the modification of the macromolecular rubber network is less important. The NMR results show an increase of the unsaturated carbon which could be undoubtly attributed to isobutylene chains scissions because a simultaneous decrease of the signals assigned to quaternary carbons and methyl carbons of the isobutylene units was observed. One can note here that the decrease of the methyl groups is in agreement with the b relaxation extinction with ageing observed in DMA results (Fig. 10). These chain scissions are in competition with crosslinks which could occur via the halogen isoprene units. However, this last phenomenon is limited because only 1e2% of isoprene units are present in the butyl rubber. The competition between these two phenomena results in a very slight variation of the mechanical properties at low deformation level. Young modulus values calculated from the tensile tests at the rubbery state are similar for

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

unaged and aged samples that let to think that there is no significant evolution in the chemical network. In contrast, the strong differences observed for the mechanical properties at high deformation levels (increase of the 50 and 100% stresses, decrease of the strains at break, drop of swelling ratios) are the proof that the material has been greatly modified by radiations. Chemical network modifications are not the main reason of these properties evolution; carbon blacks/polymer interactions probably play the major role. Concerning the effect of temperature, two hypotheses could be proposed to explain the greater importance of scission phenomena observed for ageing carried out at the highest temperature (70
C). It is well known that the first chemical stage under irradiation is CeBr or CeH breakdown which could be followed by the scission of the main chain and formed midchain macroradical [23]. This reaction is a reversible one and the formed end chain macroradical may react back with the formed terminal double bond to give the previous structure. An increase of the temperature enhances the mobility of polymer segments and so causes a diminution of the probability for radicals recombination [14]. This could be one potential reason for the chain scission predominance when irradiations are carried out at elevated temperature. Besides, several studies which focused on the thermal degradation of bromobutyl rubber polymers [32,43e45] show that a continuous long thermal exposure is likely to degrade CeH and CeBr bonds resulting in a backbone scission of elastomer chains. The same phenomenon may occur under irradiation. 5. Conclusion The mechanical properties of butyl rubber filled with a high content of carbon blacks, are deeply modified by gamma irradiation ageing. The stresses at 50% and 100% of elongation increase with increasing irradiation doses whereas the strains at break are dramatically reduced. These results are attributed not only to the radiation-induced chemical crosslinking of the rubber but also to the radiationinduced modification of the interactions between carbon blacks and macromolecular chains. From the obtained data, it can be concluded that the two phenomena are involved in the mechanical properties modifications with a major contribution of the last one. The temperature of radiochemical ageing shows also a major role in the degradation process. Whereas crosslinking is mainly observed at 25
C, chains scissions and macroscopic damages as porosity and cracks operates at 70
C. The failure envelope study reveals a complete change of the rubber properties during ageing with a plastic to brittle transition. Acknowledgements This work has been performed with the support of “Centre National des Etudes Spatiales (CNES)” and the “Conseil Régional de la region PACA”. References [1] Brydson JA. Rubbery materials and their compounds. London: Elsevier; 1988. p. 469. [2] Dubey V, Pandey SK, Rao NBSN. Research trends in the degradation of butyl rubber. Journal of Analytical and Applied Pyrolysis 1995;34:111e25. [3] Grossman E, Gouzman I. Space environment effects on polymers in low earth orbit. Nuclear Instruments and Methods in Physics Research B 2003; 208:48e57. [4] Pippin G. Space environments and induced damage mechanisms in materials. Progress in Organic Coatings 2003;47(3e4):424e31. [5] Patrick TJ. Space environment and vacuum properties of spacecraft materials. Vacuum 1981;31(8e9):351e7. [6] Dauphin J. Materials in space: working in a vacuum. Vacuum 1982;32(10e 11):669e73.

689

[7] Zaharescu T, Postolache C, Giurginca M. The structural changes in butyl and halogenated butyl elastomers during gamma irradiation. Journal of Applied Polymer Science 1996;59:969e74. [8] Karaagaç B, Sen M, Deniz V, Guven O. Recycling of gamma irradiated inner tubes in butyl based rubber compounds. Nuclear Instruments and Methods in Physics Research B 2007;265:290e3. [9] Sen M, Uzun C, Kantoglu O, Erdogan SM, Deniz V, Guven O. Effect of gamma irradiation conditions on the radiation-induced degradation of isobutyleneisoprene rubber. Nuclear Instruments and Methods in Physics Research B 2003;208:480e4. [10] Bhagawan SS, Kuriakose B, De SK. Mechanical properties and failure surfaces of gamma-ray irradiated systems based on thermoplastic 1,2-polybutadiene. International Journal of Radiation Applications and Instrumentation. Part C. Radiation Physics and Chemistry 1987;30(2):95e104. [11] Zaharescu T, Cazac C, Jipa S, Setnescu R. Assessment on radiochemical recycling of butyl rubber. Nuclear Instruments and Methods in Physics Research B 2001;185:360e4. [12] Hill DJT, O’Donnell JH, Senake Perera MC, Pomery PJ. Determination of scission and crosslinking in gamma irradiated butyl rubber. Radiation Physics and Chemistry 1992;40(2):127e38. [13] Davenas J, Stevenson I, Celette N, Cambon S, Gardette JL, Rivaton A, et al. Stability of polymers under ionising radiation: the many faces of radiation interactions with polymers. Nuclear Instruments and Methods in Physics Research B 2002;191(1e4):653e61. [14] Schnabel W. Aspects of polymer degradation and stabilisation, Chapter 4 e degradation by high energy radiation. Oxford: Elsevier; 1978. p. 140e190. [15] Rivaton A, Cambon S, Gardette JL. Radiochemical ageing of EPDM elastomers. 2. Identification and quantification of chemical changes in EPDM and EPR films g-irradiation under oxygen atmosphere. Nuclear Instruments and Methods in Physics Research B 2005;227:343e56. [16] Rivaton A, Cambon S, Gardette JL. Radiochemical ageing of EPDM elastomers. 3. Mechanism of radiooxidation. Nuclear Instruments and Methods in Physics Research B 2005;227:357e68. [17] Scagliusi SR, Cardoso ELC, Lugao AB. Effect of gamma radiation on chlorobutyl rubber vulcanized by three different crosslinking systems. Radiation Physics and Chemistry 2012;81(9):1370e3. [18] Maxwell RS, Cohenour R, Sung W, Solyom D, Patel M. The effect of g-radiation on the thermal, mechanical, and segmental dynamics of a silica filled, room temperature vulcanized polysiloxane rubber. Polymer Degradation and Stability 2003;80:443e50. [19] Basfar AA, Abdel-Aziz MM, Mofti S. Stabilization of g-radiation vulcanized EPDM rubber against accelerated aging. Polymer Degradation and Stability 1999;66:191e7. [20] Celette N, Stevenson I, Davenas J, David L, Vigier G. Relaxation behaviour of radiochemically aged EPDM elastomers. Nuclear Instruments and Methods in Physics Research B 2001;185(1e4):305e10. [21] Boiteux G, Chailan JF, Chauchard J, Seytre G. Dielectric and mechanical spectroscopies for the study of thermal and radiochemical ageing of polymers. Nuclear Instruments and Methods in Physics Research B 1997; 131(1e4):172e9. [22] Nair TM, Kumaran MG, Unnikrishnan G, Kunchandy S. Ageing studies of ethylene propylene diene monomer rubber/styrene butadiene rubber blends: effects of heat, ozone, gamma radiation, and water. Journal of Applied Polymer Science 2007;107(5):2923e9. [23] Hill DJT, O’Donnell JH, Perera MCS, Pomery PJ. High energy radiation effects on halogenated butyl rubbers. Polymer 1995;36(22):4185e92. [24] Bremner T, Hill DJT, O’Donnell JH, Senake Perera MC, Pomery PJ. Mechanism of radiation degradation of polyisobutylene. Journal of Polymer Science Part A: Polymer Chemistry 1996;34(6):871e84. [25] Chandra R, Subhash V, Verma AK. Changes in physical properties and molecular structure of butyl rubber during gamma-irradiation. Polymer 1982; 23(10):1457e60. [26] Higgins GMC, Turner DT. Radiolysis of polyisobutene. Part II. Infrared and ultraviolet absorption spectra. Journal of Polymer Science Part A: Polymer Chemistry 1964;2(4):1713e9. [27] Carswell-Pomerantz T, Babanalbandi A, Dong L, Hill DJT, Perera MCS, Pomery PJ, et al. Stability and stabilization of polymers under irradiation, changes in molecular structure and properties of irradiated polymers of different compositions-ESR and NMR study Vienna. Austria: IAEA; 1999. p. 111e128. [28] Munch E. Impact de la déformation sur la mobilité moléculaire des polymères amorphes. Lyon: INSA; 2006. [29] Polymer data handbook. Oxford University Press Inc.; 1999. [30] Najib NN, Ariff ZM, Bakar AA, Sipaut CS. Correlation between the acoustic and dynamic mechanical properties of natural rubber foam: effect of foaming temperature. Materials & Design 2011;32(2):505e11. [31] Olejniczak J, Rosiak J, Charlesby A. Gel/dose curves for polymers undergoing simultaneous cross-linking and scission. International Journal of Radiation Applications and Instrumentation. Part C. Radiation Physics and Chemistry 1991;38(1):113e8. [32] Kannan GK, Gaikewad LV, Nirmala L, Kumar NS. Thermal ageing studies of bromo-butyl rubber used in NBC personal protective equipment. Journal of Scientific and Industrial Research 2010;69:841e9. [33] Dutta NK, Khastgir D, Tripathy DK. The effect of carbon black concentration on the dynamic mechanical properties of bromobutyl rubber. Journal of Materials Science 1991;26:177e88.

690

M. Smith et al. / Polymer Degradation and Stability 98 (2013) 682e690

[34] Lassiaz M, Pouyet J, Verdu J. Effect of photochemical ageing on the tensile properties and behavior law of unstabilized film of low-density polyethylene. Journal of Materials Science 1994;29(8):2177e81. [35] Scagliusi SR, Cardoso ECL, Lugao AB. Radiation-induced degradation of butyl rubber vulcanized by three different crosslinking systems. Radiation Physics and Chemistry 2012;81(8):991e4. [36] Chu CY, Vukov R. Determination of the structure of butyl rubber by NMR spectroscopy. Macromolecules 1985;18:1423e30. [37] Coran AY, Donnet JB. The dispersion of carbon black in rubber. Part I: rapid method for assessing quality of dispersion. Rubber Chemistry and Technology 1992;65:973e97. [38] Fröhlich J, Niedermeier W, Luginsland HD. The effect of filler-filler and fillerelastomer interaction on rubber reinforcement. Composite Part A: Applied Science and Manufacturing 2004;36:449e60. [39] O’Brien J, Cashell E, Wardell GE, McBriety VJ. An NMR investigation of the interaction between carbon black and cis-polybutadiene. Macromolecules 1976;9:653e9.

[40] Sosson F, Belec L, Chailan JF, Carriere P, Crespy A. Highlight of a compensation effect between filler morphology and loading on dynamic properties of filled rubbers. Journal of Applied Polymer Science 2010; 117(5):2715e23. [41] Donnet JB, Vidal A. Carbon black: surface properties and interactions with elastomers. Advances in Polymer Science 1986;76:103e27. [42] Ramier J. Comportement mécanique d’élastomères chargés, influence de l’adhésion charge-polymère, influence de la morphologie; 2004. Lyon. [43] Kannan GK, Simha U, Gaiwad L, Nirmala L, Kumar NS. Bromo-butyl rubber for face piece of a respiratory mask. Defense Science Journal 2009;59(5): 505e11. [44] Jipa S, Giurginca M, Setnescu T, Setnescu R, Ivan G, Mihalcea I. Thermooxidative behaviour of halobutyl and butyl elastomers. Polymer Degradation and Stability 1996;54(1):1e6. [45] Janowska G, Slusarski L, Koch M, Wincel U. Thermal stability and combustibility of butyl and halogenated butyl rubbers. Journal of Thermal Analysis 1997;50(5e6):889e96.