Polymer Degradation and Stability 121 (2015) 340e347
Contents lists available at ScienceDirect
Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab
Development of high-throughput chemiluminescence imaging instrument for parallel evaluation of polymer lifetime Naoki Aratani, Ikki Katada, Koyuru Nakayama, Minoru Terano, Toshiaki Taniike* School of Materials Science, Japan Advance Institute of Science and Technology, 1-1 Asahidai, Nomi, Ishikawa 923-1292, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 10 September 2015 Received in revised form 29 September 2015 Accepted 30 September 2015 Available online 9 October 2015
A high-throughput chemiluminescence imaging instrument is developed to realize facile and systematic data accumulation in degradation and stabilization of polymer. The instrument works based on imaging chemiluminescence of 100 oxidatively degrading polymer samples that are placed in a multi-cell with arrayed wells being flushed with atmospheric gas, and thereby enables simultaneous acquisition of 100 chemiluminescence curves. The present study reports validation and demonstrative application of the instrument for the oxidative degradation of stabilized polypropylene with a variety of hindered phenol anti-oxidants. Based on quantitatively accurate and large-volume data on oxidative induction time, not only the feasibility of the new instrument but also a number of useful aspects of stabilizers are demonstrated. © 2015 Elsevier Ltd. All rights reserved.
Keywords: Chemiluminescence imaging High-throughput Oxidative degradation Anti-oxidants Structureeperformance relationship Polypropylene
1. Introduction With the continuing expansion of the polymer industry, growing attention has been paid not only on developments of renewable feedstock-based polymer [1e3] but also on reuse/ recycle of fossil-based polymer with a dominant share [4e6]. One of major concerns for the latter is to maximize the stability of polymer so as to minimize degradation during its processing and service [5,7e11], in which stabilizers (such as anti-oxidants, light stabilizers, anti-hydrolysis agents, and etc.) play a pivotal role. A variety of stabilizers have been developed and more or less commercially available for each type of polymer and its application. However, if one aims at enormous stabilization beyond average, only the way is to explore stabilizers and their combination specific to the polymer and application of interests, otherwise, it is inevitable to merely increase the amount of the addition in cost of concomitant drawbacks [12e14]. The problem lies in the difficulty to accumulate systematic knowledge on detailed structureeactivity relationships of stabilizers [15e19] and their combination [20e23], whose main causes are summarized below:
* Corresponding author. E-mail address:
[email protected] (T. Taniike). http://dx.doi.org/10.1016/j.polymdegradstab.2015.09.025 0141-3910/© 2015 Elsevier Ltd. All rights reserved.
i) The practical activity of a stabilizer is not solely determined by its intrinsic activity, but by many other factors such as mixing uniformity, solubility, migration, extraction/volatilization resistance in polymer, and so on [24e29]. Consequently, the activity of a stabilizer largely depends on the choice of polymer, and the processing and degradation methods [22,30,31], which makes it difficult to accumulate literature data in a systematic way. ii) In general, the degradation of polymer initiates and spreads in a spatially heterogeneous manner [32e38]. For example, the degradation of polypropylene (PP) initiates at the proximity of metal catalyst residues [33,36], and subsequently spreads along physicochemically weaker regions such as spherulite interfaces [38] and regions that contain lower concentrations of stabilizers due to incomplete mixing [37]. Therefore, the lifetime of polymer greatly relies on the weakest region present in a sample piece [34]. This fact suggests that measures for the polymer lifetime (such as oxidative induction time) are greatly distributed, since the lifetime is determined by accidental inclusion of the weakest regions in a sample piece. Consequently, the number of sample pieces must be large enough for acquiring quantitative trends. Thus, it can be concluded that knowledge accumulation on stabilizers is not realized without systematic lifetime measurements for each of stabilizers and their combination under constant
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
preparation and degradation protocols. Nonetheless, such systematic lifetime measurements are not practically easy, facing multidimensional and time-consuming aspects of experiments: A variety of stabilizers or an enormous number of their combination need to be tested, for each of which repetitive measurements are essential. Lifetime measurements can be greatly shortened through accelerated aging, but the extent of the acceleration is limited for reasonable extrapolation to the reality [39,40]. Likewise, in polymer stabilization, the major bottleneck has been definitely at poor throughput in lifetime measurements in contrast to the required number of the measurements. Since the last two decades, high-throughput screening (HTS) is becoming increasingly popular as a powerful tool to attain quick screening and combination optimization in materials science [41e43]. The lifetime of polymer can be measured by various methods such as the carbonyl index by IR spectroscopy, the molecular weight measurement using size-exclusion chromatography, the oxidative induction time (OIT) by differential scanning calorimetry (DSC), and so on. However, their extension to highthroughput evaluation of polymer lifetime has been scarcely attempted. Though Wroczynski et al. reported the first application of the HTS concept to the polymer degradation based on facile measurements of the melt flow index (as a measure of chain scission), the method is limited to the degradation in melt processing [44]. The chemiluminescence (CL) method is recognized as one of the most sensitive methods for in-situ detection of auto-oxidative degradation of polymer. Due to its mechanical simplicity and high sensitivity, high-throughput evaluation of polymer lifetimes would be realized by applying CL imaging to arrayed polymer samples under oxidation degradation. To date, most of previous efforts relevant to CL imaging have focused on studying space-resolved degradation within a single polymer sample. As was stated by Ahlblad, Gijsman, and their coworkers (who have firstly applied CL imaging to simultaneous acquisition of CL curves for 11 polymer pieces) [45e47], the following major technical issues must be resolved: iii) It is well established that oxidative polymer degradation propagates in an infectious manner, in which a degrading sample promotes the degradation of neighboring samples through emission and migration of volatile organic compounds [48,49]. Therefore, arrayed samples under a degradation test need to be physically and atmospherically isolated among each other. iv) As the polymer oxidation follows an Arrhenius behavior in terms of temperature, high temperature uniformity is required among arrayed polymer samples [50]. In this contribution, we report a home-made high-throughput CL imaging instrument (HTP-CLI) for realizing lifetime determination of 100 polymer samples in a single measurement. Quantitative accuracy and mutual independence of 100 parallel degradation tests were successfully proven using thermooxidative degradation of stabilized polypropylene (PP) samples. In addition, the efficacy of 10 different hindered phenols was systematically compared based on results of 300 degradation tests, which were acquired within 1 month using HTP-CLI in contrast to ca. 3 years for conventional methods.
341
104, mmmm ¼ 98 mol%) was obtained by bulk polymerization of propylene using a 4th-generation ZieglereNatta catalyst. Hindered phenol anti-oxidants of AO and IRGANOX series were supplied from ADEKA Corporation and Toyotsu Chemiplas Corporation, respectively (Fig. 1). 2.2. Sample preparation Test samples for the developed instrument were prepared as follows: The pristine polypropylene powder was first melt mixed with 1.0 wt% of AO-50 using a two-roll mixer at 20 rpm and at 185 C for 15 min. Thus obtained masterbatches were further melt mixed with additional pristine powder at 185 C for 15 min to adjust the content of AO-50 at either 0.07 or 0.10 wt%. The products were hot-pressed into 100-mm-thick films at 230 C and 10 MPa, and then quenched at 100 C. Thus obtained sample films were termed as PP-0.07 and PP-0.10 according to the anti-oxidant content. Sample films that contained 0.07 wt% of different antioxidants were prepared in a similar manner. 2.3. Instrumental The developed HTP-CLI instrument is illustrated in Fig. 2. As was mentioned, infectious spreading of degradation through volatile degradation products is a technical bottleneck for making parallel degradation tests in a small area. We have successfully eliminated the problem by designing a multi-cell that consists of three layers [51]: A well plate with equally-spaced 10 10 columnar wells (D ¼ 6.0 mm, H ¼ 3.0 mm), a flow plate to distribute the air flow into each well, and a chimney plate with open cylinders to eject volatile degradation products with the distributed air flow, thus prohibiting the infectious spreading. The multi-cell is placed in a constant-temperature oven which equips cartridge heaters and a circulating fan. Two glass plates are placed at the top of the oven for heat insulation as well as for chemiluminescence detection. A tricolor CCD camera (BU-51C, BITRAN) is loaded at the top of the system. Dry air is supplied at a controlled flow volume to the gas inlets of the multi-cell after heated to a target temperature using an external heater. At the set temperature of 150 C and the flow volume of 8.0 L/min, the temperature distribution over the multicell was measured to be sufficiently within ±0.5 C. Chemiluminescence was acquired by continuously capturing images with appropriate exposure time (Fig. 2b). These continuous images are analyzed by ImageJ software, where the intensity of the chemiluminescence was calculated based on (R þ G þ B)/3 for each cell. Likewise, chemiluminescence curves for 10 10 samples are simultaneously acquired (Fig. 2c), which are used to determine the oxidation induction time (OIT). Thermo-oxidative degradation using the developed instrument was performed as follows: 100 sample pieces of 5.0 f were cut out from the above-mentioned films and were placed in the multi-cell at room temperature. The multi-cell was heated by the oven to the specific temperature under the fixed flow volume of dry air that was pre-heated at the target temperature. Once the target temperature was attained, the images were continuously acquired and stored with the acquisition time of 20 min until all the pieces finished the chemiluminescence emission. 3. Results and discussion
2. Experimental 2.1. Materials Additive-free powder of isotactic polypropylene (iPP, Mn ¼ 4.6
In order to evaluate the feasibility of the instrument, parallel degradation tests were implemented: 100 sample pieces that were cut out from PP-0.07 and PP-0.10 films were subjected to thermooxidative degradation at 150 C under the dry air flow of 8.0 L/
342
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
Fig. 1. Hindered phenol anti-oxidants employed in this study.
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
343
Fig. 2. a) Developed high-throughput chemiluminescence imaging instrument; b) Snapshots of chemiluminescence images during the thermoxidative degradation of stabilized polypropylene samples at 150 C; c) Acquisition of chemiluminescence curves from image sequences.
min. Fig. 3 summarizes the obtained OIT values. In Fig. 3a, it was found that 100 pieces exhibited the dispersion of OIT values (maximum 30% with respect to the average values) even though they were from the same sample film. Since the dispersion pattern did not show any regularity among different measurements, possibilities of instrumental problems (such as non-uniformity of the ejecting flow, a temperature gradient in the oven and so on) were obviously excluded. Instead, the cumulative distribution of the OIT values was well fit to that of the normal distribution with suitable average and variation values (Fig. 3b). This fact suggested that the dispersion of the OIT values was caused by random sampling from a sample film which contained internal heterogeneity. Such heterogeneity is relatively well known in polymer oxidative degradation, and its origin could be the location of catalyst residues, the dispersion level of stabilizers in polymer, and so on [37]. When successive OIT acquisition was performed for the same sample pieces using a usual chemiluminescence analyzer (CLA-ID2-HS,
TOHOKU ELECTRONIC INDUSTRIAL), similar dispersion was observed. The conclusive OIT values were derived as 23 ± 4.5 h for PP-0.07 and 60 ± 7.0 h for PP-0.10 (1st). The comparison between the first and second measurements for PP-0.10 has successfully demonstrated quantitative reproduction of the OIT values by the developed instrument. It must be stressed that quantitative accuracy of the polymer lifetime was assured by testing a relatively large number of sample pieces, which is hardly accessible for conventional one-by-one measurements. The absence of unfavorable infectious degradation among neighboring wells, i.e. independence of 100 parallel measurements, was examined by the following experiments: Each 50 pieces of PP0.07 and PP-0.10 were alternatively placed in the multi-cell as shown in Fig. 4a, and subjected to the degradation at 150 C under the dry air flow of either 8.0 or 2.0 L/min. In the presence of infectious degradation, emission of volatile degradation products from less stable PP-0.07 would shorten the lifetime of PP-0.10. The
344
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
Fig. 3. Results of thermo-oxidative degradation tests using 100 sample pieces at 150 C under the dry air flow of 8.0 L/min: a) Heatmap visualization of OIT values and b) cumulative distribution of OIT values. The solid lines in b) represent cumulative curves for the best-fit normal distribution.
results are summarized in Fig. 4. From the heatmap of the OIT values, it appeared that the OIT values clearly represented the alternate placement of PP-0.07 and PP-0.10 at 8.0 L/min, which suggested the independence of the parallel measurements. On the contrary, at 2.0 L/min, the OIT values for PP-0.10 became much
shorter at the middle area of the multi-cell, which was the farthest from the gas inlets. It was clear that insufficient ejecting flow caused infectious spreading of degradation from PP-0.07 to PP0.10 at the middle area of the cell, while it was not the case at the upper and lower areas. The cumulative distribution further
Fig. 4. Results of thermo-oxidative degradation tests at 150 C under the dry air flow of either 8.0 or 2.0 L/min, where each 50 pieces of PP-0.07 and PP-0.10 were alternatively placed in the multi-cell: a) Heatmap visualization of OIT values and b) cumulative distribution of OIT values.
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
demonstrates the significance of the ejecting flow on the parallel OIT measurements (Fig. 4b). At 8.0 L/min, the OIT values for the 50 sample pieces of PP-0.10 nicely reproduced cumulative distribution obtained for 100 sample pieces of PP-0.10 in Fig. 3b. Meanwhile, nearly half of the sample pieces showed shorter OIT values at 2.0 L/ min, thus broadening the left half of the cumulative distribution. Thus, the critical role of the ejecting flow has been found for parallel degradation tests: At a sufficient flow volume, independence of parallel OIT measurements can be assured. The instrumental feasibility for 100 parallel degradation tests had been above established in a quantitative manner. Demonstration experiments were conducted for PP samples containing 10 kinds of hindered phenol anti-oxidants at 0.07 wt% (Fig. 1). First, 10 sample pieces for each stabilizer were degraded at 140 C under the dry air flow of 8.0 L/min. The OIT values are summarized in Fig. 5a.
345
At 140 C, the OIT values were roughly correlated with the molecular weight of the stabilizers, e.g. low molecular-weight stabilizers such as AO-40, AO-50, and AO-30 provided smaller OIT values. Irganox 1098 led to a notably large OIT value at the corresponding molecular weight, while a notably short OIT value was found for AO-20. In general, the efficacy of hindered phenol antioxidants is determined by many factors: The number of trappable radicals per molecule, the compatibility, the molecular mobility and the retention in polymer matrices, and so on. Further, the durability of stabilizers in melt processing is also important, since stabilizers are more or less consumed in processing, whose condition is usually harsher than that for accelerated aging. Likewise, the efficacy of stabilizers is dependent not only on the condition of accelerated aging but also on the processing history in a complicated manner through the interplay of the above-mentioned
Fig. 5. a) Average OIT values at 140 C for different hindered phenol anti-oxidants, b) correlation between the OIT values and T10 (the temperature at 10 wt% weight loss of pristine stabilizers), and c) activation energies for 1/OIT in the range of 140e160 C (R2 > 0.99). All the degradation tests were performed under the dry air flow of 8.0 L/min using the developed instrument.
346
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
factors. Though we have considered potentially important parameters such as the molecular weight, the concentration of active phenolic hydroxyl groups, the volatility (evaluated as T10 of pristine stabilizers in thermogravimetric analysis), and the solubility of stabilizers in model solvent, a relatively good correlation with the OIT values was found only for T10 values. As shown in Fig. 5b, the OIT values tended to be larger when stabilizers were less volatile (or less decomposable) under air. It was considered that both/either in the accelerated aging at 140 C and/or in the melt processing, the retention of stabilizers was most important. Second, similar degradation tests were performed at 150 and 160 C in order to acquire the temperature dependence of the stabilization efficacy for 10 kinds of the stabilizers. The activation energies were acquired based on the Arrhenius plot of 1/OIT in the range of 140e160 C. It can be seen in Fig. 5c that the activation energies widely varied depending on the stabilizers: The maximum value reached 223 kJ/mol (AO-80), while the minimum value was 96 kJ/mol (AO-40). Considering that a greater activation energy results in larger temperature dependence of OIT values, it is clear that the order of the stabilization efficacy can be reversed at different temperature (known as a crossover phenomenon). For instance, the best stabilization was achieved by Irganox 1330 at 140 C and Irganox 1098 at 150 C. Thus, the wide variation in the activation energy sensitively affected relative performance of the stabilizers even at an 10 C interval. It appeared still difficult to fully conclude the structureeperformance relationships of hindered phenol anti-oxidants based on the given variety of molecular structures. However, it must be stressed that 300 degradation tests (corresponding to 2.8 y in sequential measurements) were completed within 1 month using the developed instrument, which has enabled us to withdraw useful conclusions in a quantitatively validated manner. 4. Conclusions In order to improve the throughput of polymer lifetime measurements, a high-throughput chemiluminescence imaging (HTPCLI) instrument was successfully developed, in which oxidative induction time (OIT) of 100 polymer samples can be simultaneously acquired in a single measurement. In thermo-oxidative degradation of stabilized polypropylene, it was found that the dispersion of OIT values followed the Gauss distribution plausibly due to the microscopic mixing non-uniformity and the HTP-CLI instrument enabled quantitative reproduction of not only the average but also standard deviation of OIT. Furthermore, it was revealed that the ejection of volatile degradation products was essential to prevent infectious degradation spreading among neighboring samples and thus to assure implementation of 100 degradation tests in an independent manner. Employing 10 kinds of hindered phenol antioxidants, a structureeperformance relationship study was demonstrated, where 300 OIT values amounting to 2.8 y were acquired to clarify the importance of the thermo-oxidative stability of stabilizer molecules themselves. Though the full description of structureeperformance relationships requires further data accumulation, the significance of HTP-CLI is at the realization of fast systematic data accumulation on polymer degradation and stabilization. Acknowledgments The Authors are grateful to ADEKA Corporation, Japan Polychem Corporation, and Toyotsu Chemiplas Corporation for the donation of reagents. This work was conducted under cooperation of Tohoku Electronic Industrial Co., Ltd.
References [1] L. Yua, K. Deana, L. Li, Polymer blends and composites from renewable resources, Prog. Polym. Sci. 31 (6) (2006) 576e602. [2] M. Okada, Chemical syntheses of biodegradable polymers, Prog. Polym. Sci. 27 (1) (2002) 87e133. glise, M.F. Lacrampe, P. Krawczak, Thermosetting (bio) [3] J.M. Raquez, M. Dele materials derived from renewable resources: A critical review, Prog. Polym. Sci. 35 (4) (2010) 487e509. [4] R. Siddique, J. Khatib, I. Kaur, Use of recycled plastic in concrete: A review, Waste Manag. 28 (10) (2008) 1835e1852. [5] J. Hopewell, R. Dvorak, E. Kosior, Plastics recycling: challenges and opportunities, Phil. Trans. R. Soc. B 364 (2009) 2115e2126. [6] S. Serranti, A. Gargiulo, G. Bonifazi, Characterization of post-consumer polyolefin wastes by hyperspectral imaging for quality control in recycling processes, Waste Manag. 31 (11) (2011) 2217e2227. rov [7] S. Luzuriaga, J. Kova a, I. Fortelný, Degradation of pre-aged polymers exposed to simulated recycling: properties and thermal stability, Polym. Degrad. Stab. 91 (6) (2006) 1226e1232. pez, D. García, M.D. Salvador, Recycling of ABS and PC from [8] R. Balart, J. Lo electrical and electronic waste. Effect of miscibility and previous degradation on final performance of industrial blends, Eur. Polym. J. 41 (9) (2005) 2150e2160. €ller, T. Gevert, Degradation of post-consumer polypropylene [9] A. Jansson, K. Mo materials exposed to simulated recyclingdmechanical properties, Polym. Degrad. Stab. 82 (1) (2003) 37e46. [10] R. Pfaendner, H. Herbst, K. Hoffmann, F. Sitek, Recycling and restabilization of polymers for high quality applications. An overview, Angew. Makromol. Chem. 232 (1) (1995) 193e227. [11] F. Vilaplana, S. Karlsson, Quality concepts for the improved use of recycled polymeric materials: a review, Macromol. Mater Eng. 293 (4) (2008) 274e297. [12] N.C. Billingham, P.D. Calvert, I.W. Okopi, A. Uzuner, The solubility of stabilizing additives in polypropylene, Polym. Degrad. Stab. 31 (1) (1991) 23e36. [13] K. Kech-Antoine, E. Lievens, J. Bayer, J. Mara, D.S. Jung, S.L. Jung, in: J.R. Wagner Jr. (Ed.), Multilayer Flexible Packaging, Elsevier, New York, 2009, p. 37. [14] C. Homkhiew, T. Ratanawilai, W. Thongruang, Composites from recycled polypropylene and rubberwood flour: effects of composition on mechanical properties, J. Thermoplast. Compos Mater 28 (2) (2015) 179e194. [15] F. Catalina, C. Peinado, N.S. Allen, T. Corrales, Chemiluminescence of polyethylene: the comparative antioxidant effectiveness of phenolic stabilizers in low-density polyethylene, J. Polym. Sci. Part A Polym. Chem. 40 (19) (2002) 3312e3326. [16] I. Vulic, G. Vitarelli, J.M. Zenner, Structure-property relationships: phenolic antioxidants with high efficiency and low colour contribution, Polym. Degrad. Stab. 78 (1) (2002) 27e34. [17] E. Richaud, B. Fayolle, J. Verdu, Polypropylene stabilization by hindered phenols - Kinetic aspects, Polym. Degrad. Stab. 96 (1) (2011) 1e11. [18] T. Zaharescu, M. Giurginca, I. Mihalcea, Effects of some antioxidants on the stability of ethylene-propylene elastomers, Polym. Degrad. Stab. 48 (1) (1995) 151e154. [19] E.N. Step, N.J. Turro, M.E. Gande, P.P. Klemchuk, Mechanism of Polymer Stabilization by Hindered-Amine Light Stabilizers (HALS). Model Investigations of the Interaction of Peroxy Radicals with HALS Amines and Amino Ethers, Macromolecules 27 (9) (1994) 2529e2539. [20] N.S. Allen, J.L. Gardette, J. Lemaire, Interaction between antioxidants and hindered piperidine compounds in the photostabilisation of polypropylene: influence of processing history, Polym. Photochem 1 (2) (1981) 111e121. [21] S. Jipa, T. Zaharescu, R. Setnescu, T. Setnescu, M.J.S. Brites, A.M.G. Silva, M.J. Marcelo-Curto, B. Gigante, Chemiluminescence study of thermal and photostability of polyethylene, Polym. Int. 48 (5) (1999) 414e420. Chmela, B. Pawelke, G. Theumer, W.D. Habicher, New combined sa, S. [22] C. Ko hindered phenol/hindered amine stabilizers for polymers based on diphenylmethane-4,40 -diisocyanate, Polym. Degrad. Stab. 81 (3) (2003) 453e461. [23] T. Zaharescu, S. Jipa, R. Setnescu, J. Brites, M.A. Esteves, B. Gigante, Synergistic effects on thermal stability of ethyleneepropylene elastomers stabilized with hindered phenols and secondary amines, Polym. Test. 21 (2) (2002) 149e153. [24] D. Park, D. Kobayashi, H. Suh, Y. Cho, A. Lee, Synthesis of (3-tert-butyl-4hydroxy-5-methylphenyl) propionate derivatives and their thermal antioxidation behavior for POM, J. Appl. Polym. Sci. 124 (2) (2012) 1731e1736. [25] G.A. George, M. Ghaemy, The effect of stabilisers on integrated chemiluminescence in the early stages of polypropylene photo-oxidation, Polym. Degrad. Stab. 34 (1e3) (1991) 37e53. €ldes, Study of the effects influencing additive migration in polymers, [26] E. Fo Angew. Makromol. Chem. 261e262 (1) (1998) 65e76. [27] J. Tocha cek, Effect of secondary structure on physical behaviour and performance of hindered phenolic antioxidants in polypropylene, Polym. Degrad. Stab. 86 (2) (2004) 385e389. [28] N. Haider, S. Karlsson, Kinetics of migration of antioxidants from polyolefins in natural environments as a basis for bioconversion studies, Biomacromolecules 1 (3) (2000) 481e487. [29] S. Beißmann, M. Stiftinger, K. Grabmayer, G. Wallner, D. Nitsche, W. Buchberger, Monitoring the degradation of stabilization systems in polypropylene during accelerated aging tests by liquid chromatography combined
N. Aratani et al. / Polymer Degradation and Stability 121 (2015) 340e347
[30] [31] [32] [33] [34]
[35]
[36] [37]
[38]
[39]
[40]
[41]
with atmospheric pressure chemical ionization mass spectrometry, Polym. Degrad. Stab. 98 (9) (2013) 1655e1661. R. Pfaendner, How will additives shape the future of plastics? Polym. Degrad. Stab. 91 (9) (2006) 2249e2256. H. Zweifel, Stabilization of Polymeric Materials, Springer, 1998. P. Richters, Initiation process in the oxidation of polypropylene, Macromolecules 3 (2) (1970) 262e264. J.B. Knight, P.D. Calvert, N.C. Billingham, Localization of oxidation in polypropylene, Polymer 26 (11) (1985) 1713e1718. M. Celina, G.A. George, N.C. Billingham, Physical spreading of oxidation in solid polypropylene as studied by chemiluminescence, Polym. Degrad. Stab. 42 (3) (1993) 335e344. M. Celina, G.A. George, D.J. Lacey, N.C. Billingham, Chemiluminescence imaging of the oxidation of polypropylene, Polym. Degrad. Stab. 47 (2) (1995) 311e317. I. Blakey, G.A. George, Raman spectral mapping of photo-oxidised polypropylene, Polym. Degrad. Stab. 70 (2) (2000) 269e275. B. Fayolle, L. Audouin, G.A. George, J. Verdu, Macroscopic heterogeneity in stabilized polypropylene thermal oxidation, Polym. Degrad. Stab. 77 (3) (2002) 515e522. T. Taniike, M. Umemori, F. Kobayashi, M. Terano, Critical role of spherulite structure on behavior of stabilizers in poly(propylene) stabilization, Macromol. Symp. 260 (1) (2008) 42e48. M. Celina, K.T. Gillen, R.A. Assink, Accelerated aging and lifetime prediction: review of non-Arrhenius behaviour due to two competing processes, Polym. Degrad. Stab. 90 (3) (2005) 395e404. M. Celina, Review of polymer oxidation and its relationship with materials performance and lifetime prediction, Polym. Degrad. Stab. 98 (12) (2013) 2419e2429. R. Hoogenboom, M.A.R. Meier, U.S. Schubert, Combinatorial methods,
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49] [50]
[51]
347
automated synthesis and high-throughput screening in polymer research: past and present, Macromol. Rapid Commun. 24 (1) (2003) 16e32. A. Hagemeyer, B. Jandeleit, Y.M. Liu, D.M. Poojary, H.W. Turner, A.F. Volpe, H.W. Weinberg, Applications of combinatorial methods in catalysis, Appl. Catal. A Gen. 221 (1e2) (2001) 23e43. R. Potyrailo, K. Rajan, K. Stoewe, I. Takeuchi, B. Chisholm, H. Lam, Combinatorial and high-throughput screening of materials libraries: review of state of the art, ACS Comb. Sci. 13 (6) (2011) 579e633. R.J. Wroczynski, M. Rubinsztajn, R.A. Potyrailo, Evaluation of process degradation of polymer formulations utilizing high-throughput preparation and analysis methods, Macromol. Rapid Commun. 25 (1) (2004) 264e269. G. Ahlblad, B. Stenberg, B. Terselius, T. Reitberger, Imaging chemiluminescence instrument for the study of heterogeneous oxidation effects in polymers, Polym. Test. 16 (1) (1997) 59e73. M. Hamskog, G. Ahlblad, G. F€ arnert, P. Gijsman, B. Terselius, New multi-cell imaging-chemiluminescence (ICL) instrument to evaluate the stability of polymers, Polym. Test. 22 (4) (2003) 363e369. M. Hamskog, B. Terselius, P. Gijsman, Multi-cell imaging chemiluminescence to map heterogeneous degradation of polypropylene plaques, Polym. Degrad. Stab. 82 (2) (2003) 181e186. P. Eriksson, T. Reitberger, B. Stenberg, Gas-phase contribution to the spreading of oxidation in polypropylene as studied by imaging chemiluminescence, Polym. Degrad. Stab. 78 (1) (2002) 183e189. M. Celina, R.L. Clough, G.D. Jones, Initiation of polymer degradation via transfer of infectious species, Polym. Degrad. Stab. 91 (5) (2006) 1036e1044. €hnke, Chemiluminescence as an industrial test method for D.R. Kohler, C. Kro antioxidant effectiveness in polyolefinsd1. Fundamental considerations, Polym. Degrad. Stab. 62 (2) (1998) 385e393. T. Taniike, M. Terano, T. Sato, JP Patent 114212; 2015.