Unravelling the Photodegradation Mechanisms of a Low Bandgap

Article pubs.acs.org/JPCC

Unravelling the Photodegradation Mechanisms of a Low Bandgap Polymer by Combining Experimental and Modeling Approaches Isabel Fraga Domínguez,†,‡,§ Paul D. Topham,§ Pierre-Olivier Bussière,‡,∥ Didier Bégué,*,⊥ and Agnès Rivaton*,†,‡ Université Blaise Pascal, Institut de Chimie de Clermont-Ferrand, Équipe Photochimie, BP 10448, F-63000 Clermont-Ferrand, France ‡ CNRS, UMR 6296, ICCF, Équipe Photochimie, BP 80026, F-63171 Aubière, France § Chemical Engineering and Applied Chemistry, Aston University, Birmingham, B4 7ET, United Kingdom ∥ Clermont Université, École Nationale Supérieure de Chimie, ICCF, Équipe Photochimie, BP 10448, F-63000 Clermont-Ferrand, France ⊥ Université de Pau et des Pays de l’Adour, IPREM (ECP, CNRS-UMR 5254), 2 Avenue Président Angot, 64053 Pau, France †

S Supporting Information *

ABSTRACT: Large-scale introduction of Organic Solar Cells (OSCs) onto the market is currently limited by their poor stability in light and air, factors present in normal working conditions for these devices. Thus, great efforts have to be undertaken to understand the photodegradation mechanisms of their organic materials in order to find solutions that mitigate these effects. This study reports on the elucidation of the photodegradation mechanisms occurring in a low bandgap polymer, namely, Si-PCPDTBT (poly[(4,4′-bis(2ethylhexyl)dithieno[3,2-b:2′,3′-d]silole)-2,6-diyl-alt-(4,7-bis(2-thienyl)-2,1,3benzothiadiazole)-5,5′-diyl]). Complementary analytical techniques (AFM, HS-SPME-GC-MS, UV−vis and IR spectroscopy) have been employed to monitor the modification of the chemical structure of the polymer upon photooxidative aging and the subsequent consequences on its architecture and nanomechanical properties. Furthermore, these different characterization techniques have been combined with a theoretical approach based on quantum chemistry to elucidate the evolution of the polymer alkyl side chains and backbone throughout exposure. Si-PCPDTBT is shown to be more stable against photooxidation than the commonly studied p-type polymers P3HT and PCDTBT, while modeling demonstrated the benefits of using silicon as a bridging atom in terms of photostability.

1. INTRODUCTION The field of organic electronics has strongly evolved over the last two decades mainly toward the development of practical applications for ultrathin, large-area, and/or flexible devices.1−4 Since organic solar cells (OSCs) can be produced by low-cost and low-energy consumption processes, these devices have been expected to offer a powerful alternative to the conventional electronics technology landscape relying on inorganic materials. However, despite high expectation and very large interest, OSCs are still limited by their low performances and short lifetimes.5−10 To overcome the first problem, numerous research groups have been involved in the development of new kinds of materials, particularly focusing on the so-called low bandgap polymers, based on the “push−pull” concept.5,11,12 It has been demonstrated that very small structural variation can lead to significant modification in material properties. For instance, it has been demonstrated that replacing the bridging carbon atom in the cyclopentadithiophene unit of low bandgap C-PCPDTBT (poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b′]-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)]; © 2014 American Chemical Society

see Figure S1 for chemical structure) with a silicon atom (to give Si-PCPDTBT (poly[(4,4′-bis(2-ethylhexyl)dithieno[3,2b:2′,3′-d]silole)-2,6-diyl-alt-(4,7-bis(2-thienyl)-2,1,3-benzothiadiazole)-5,5′-diyl]); chemical structure in Figure 1) improves properties of the material, such as higher crystallinity, enhanced

Figure 1. Chemical structure of Si-PCPDTBT. Received: October 13, 2014 Revised: December 16, 2014 Published: December 19, 2014 2166

DOI: 10.1021/jp5103065 J. Phys. Chem. C 2015, 119, 2166−2176

Article

The Journal of Physical Chemistry C

irradiated in a fixed position, and UV irradiance under 300 nm and IR irradiation are filtered. Photooxidation experiments using both irradiation setups resulted in identical results, the main difference being the velocity of the degradation process (Figures S2 and S3). 2.4. Spectroscopy Analysis. UV−vis absorption spectra were registered using a Shimadzu UV-2600 spectrophotometer provided with an integration sphere. Infrared spectra were recorded by a Nicolet 760 Magna spectrophotometer working in transmission mode and purged with dry air. The parameters used were 32 acquisitions and 4 cm−1 resolution. 2.5. Identification of Photoproducts. The identification of film oxidation products was performed using chemical derivatization treatments with sulfur tetrafluoride (SF4) and ammonia (NH3) in combination with IR spectroscopy, as previously described.24 HS-SPME-GC-MS was employed to collect and analyze the volatile compounds created upon photooxidation of Si-PCPDTBT films, irradiated in sealed vials for 70 h. The procedure has been detailed for studies concerning other polymeric matrices.25 2.6. Atomic Force Microscopy. A Nanoscope IIIa atomic force microscope from Veeco Instruments was used to obtain surface topography measurements and perform nanoindentations. The applied methodology has been previously described.26 2.7. Computational Details. The near-infrared (NIR) spectra were fully interpreted by applying a method developed for efficient automatic computation of both the infrared wavenumbers and intensities. The employed procedure uses Parallel Variational Multiple Window Configuration Interaction wave functions, the so-called P_VMWCI2 algorithm,27 which incorporates both mechanical and electrical anharmonic effects. It has been shown that inclusion of anharmonicities is crucial to correctly assigning the fundamental, combination, and overtone vibrational frequencies in the infrared spectrum of the target system. Molecular geometries of model compounds were fully optimized within density functional theory (DFT) method, using two double-ζ polarized basis-sets (6-31G* and +631+G**).28−30 DFT calculations used a B3LYP exchangecorrelation function.31 Where required, the open-shell wave function was set to an unrestricted type (UHF/UKS). Modeling of hydrogen-abstracted geometries was performed with a doublet multiplicity in a neutral state. BDMs were calculated with respect to the ground state. All calculations were performed using Gaussian 09 software.32 Spin distribution was extracted from the Löwdin partition.33,34

charge transport characteristics, reduced bimolecular recombination, and reduced formation of charge transfer complexes when blended with a fullerene derivative.13−15 Indeed, SiPCPDTBT:PCBM solar cells gave efficiencies of 5.2%, making Si-PCPDTBT the first low-bandgap semiconducting polymer to have a certified efficiency of over 5%.15 Moreover, it has also been found that blend films of SiPCPDTBT:PCBM are less prone to photooxidation than CPCPDTBT:PCBM films.16 The lifetime of polymer-based solar cells appears to be limited by a number of factors, with one of the most crucial being the photochemical degradation of the active layer components.9 Indeed, the employed polymers, as organic materials, are unstable and degrade when exposed to solar irradiation, this effect being more enhanced when UV−vis radiation is combined with oxygen ingress into the device.17−21 Therefore, the elucidation of the mechanism of polymer photodegradation is a key point to develop strategies to decrease or prevent the loss of the functional properties of the material.22 The work herein describes the photochemical behavior of low bandgap Si-PCPDTBT thin films under photooxidative conditions, artificially obtained using a SEPAP 12/24 device. The evolution of polymer films under exposure is monitored using a range of analytical techniques, namely UV−vis and IR spectroscopy, atomic force microscopy (AFM), and profilometry. Furthermore, headspace solid-phase microextraction (HSSPME) coupled with gas chromatography/mass spectrometry (GC-MS) has been exploited to elucidate any gas phase low molecular weight byproducts released during degradation. Finally, the results obtained with these analytical techniques are combined with molecular modeling, allowing for a complete elucidation of the photochemical processes responsible for the degradation of Si-PCPDTBT under the exposure conditions.

2. METHODS 2.1. Materials. Si-PCPDTBT (Mw = 52.8 kg mol−1, PD = 2.85) was provided by BELECTRIC OPV GmbH and was used without further purification. Ortho-dichlorobenzene was obtained from Aldrich, HPLC grade (99%). 2.2. Thin Films Preparation. The films (∼200 nm thick) were prepared via spin coating using a G3P-8 Cookson Electronics Equipment on KBr, CaF2 and glass substrates. The thickness of the samples was determined with a KLA Tencor Alpha Step IQ profilometer. 2.3. Aging Experiments. Photodegradation of the samples was performed in a SEPAP 12/24 device, consisting on a square reactor containing a rotating carousel to hold the samples, and four medium-pressure mercury vapor lamps (Mazda MA 400) located in each corner of the chamber. Wavelengths below 300 nm are filtered by the borosilicate glass envelope of the lamps. The carousel has capacity for 24 samples, placed at 20 cm distance from the lamps axis, and turns four times per minute. Chamber temperature is maintained at 60 °C during all experiments, controlled by a platinum probe in contact with a polyethylene film placed on the carousel. The humidity level, a parameter known to play a role in the photooxidation rate and product formation,23 was set at 7% (60 °C) for all of the experiments performed in this study. The degradation of the polymer using SEPAP 12/24 was compared to that using a Suntest CPS/XLS Atlas device (see Supporting Information). The latter is provided with a xenon lamp from ATLAS (NXE1700) configured at 750 W m−2; a cryostat maintains the black standard temperature (BST) at 80 °C. Samples are

3. RESULTS AND DISCUSSION 3.1. Morphological Analysis. The morphology of pristine Si-PCPDTBT films spin-coated on glass substrates and their corresponding evolution upon aging was characterized by AFM. Figure 2 shows the AFM surface images and profiles obtained for pristine films and films exposed to photooxidation in SEPAP 12/24 for 150 h. Prior to irradiation, the surface of the film has a relatively even and homogeneous appearance. After 40 h of aging, although the topographic image does not appear to have significantly changed, when plotting the surface profile (Figure 2b), a variation in the surface height is already noticeable. A dramatic modification is observed in both the surface image and the corresponding profile after 100 h of irradiation, characterized by the formation of a multitude of small spherical features. Finally, 50 h later, these features 2167


Article


Figure 3. (a) Roughness and thickness evolution of Si-PCPDTBT thin films upon 150 h photooxidation, extracted from AFM profiles and profilometric analysis, respectively. (b) Nanohardness evolution upon 100 h of photooxidation, as calculated using AFM nanoindentations. After 100 h of exposure, it was not possible to obtain accurate reproducible hardness values due to the high increase in roughness and heterogeneity of the sample surface.

Figure 3a, show a rapid decrease after 60 h of exposure, finally reaching approximately half of its initial value after 150 h. The increased error associated with each measurement in the later stages of degradation reveals once again the augmentation of surface heterogeneity with degradation time. The thickness values have been employed to set up a bar in the surface profile evolution in Figure 2b, indicating the position of the initial surface height in comparison with the final surface level. This gives account for the important loss of materials that occurs during light exposure. It must be noted that surface degradation can highly influence the active layer interfaces in terms of mechanical, adhesive, or charge extraction properties, all of which ultimately having drastic consequences on the performance of the device.36 AFM nanoindentations were subsequently performed using a constant load or force, allowing any variation in mechanical properties of the sample to be monitored. During the whole irradiation period, particular attention was paid to the employed force to avoid taking into account the mechanical properties of the glass substrate. Quantitative values of the nanohardness of the material were estimated using the Oliver and Pharr procedure,37 showing a steady increase in stiffness with oxidation time (Figure 3b). As previously suggested for PCDTBT, the most relevant hypothesis is that radicals involved in the chain radical oxidation process can recombine, provoking cross-linking of the sample, and therefore increasing stiffness. 3.2. Chemical Evolution. The observed changes in thickness, morphology, and stiffness originate from modification of the chemical structure of the polymer macromolecular chains and the formation of degradation products. Accordingly, these chemical changes have been monitored using infrared spectroscopy (IR) in combination with chemical derivatization treatments and analysis of the gas phase using HS-SPME-GCMS, in order to identify the oxidation products and ultimately propose a degradation mechanism.

Figure 2. (a) Height images obtained by AFM, indicating the surface evolution of Si-PCPDTBT films at different stages of photooxidation. (b) Surface profiles corresponding to the height images displayed in panel a. Thickness data, shown in Figure 3, were used to set up a bar (−) comparing the final surface profile to the initial surface position and show that the there is a 100 nm average decrease in thickness after 150 h.

coalesce and create a smaller number of larger aggregates. For previously studied conjugated polymers, such as PCDTBT (poly[N-9′-hepta-decanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2thienyl-2′,1′,3′-benzothiadiazole)]), this type of evolution with the formation of spherical features has been related to a new arrangement of the polymer macromolecules, namely πstacking, indicated by changes in the UV−vis spectra.35 However, no such evidence has been observed in the case of Si-PCPDTBT films. Consistent with the described evolution, Figure 3a shows that roughness values extracted from AFM topographic images remain constant and around 4−6 nm during the first 60 h of exposure. Following this exposure time, the roughness dramatically increases together with the heterogeneity of the surface, the latter indicated by the augmentation in the error bars (z values). Usually, this evolution of surface roughness can be related to the formation and subsequent loss of low molecular weight species from the polymer. Additionally, thickness values obtained by profilometry, also reported in 2168


Article

The Journal of Physical Chemistry C As displayed in Figure 1, Si-PCPDTBT is composed of a silicon-cyclopentadithiophene (Si-CPDT) donor group bearing two alkyl solubilizing side-chains linked to the silicon atom and a benzothiadazole (BT) unit as a withdrawing group. Due to the complexity of the IR spectrum, the different structures in the polymer were identified by matching the experimental IR data with theoretical predictions in solution based on B3LYP/ 6-31G* anharmonic calculations (see Table 1).38,39 It is noteworthy that experimental and computational spectra are in good agreement. Table 1. Attribution of the IR Bands of Si-PCPDTBTa ν̃/cm−1 experimental

B3LYP/6-31G*

assignment

613

610 (29.3)

749

741 752 788 799

ω thiophene ring + ω benzothiadiazole ω benzothiadiazole

794 881

870 (21.0)

1105

1096 (43.3) 1109 (12.9) 1180 (37.4) 1183 (37.1) 1231 (19.4) 1262 (30.4) 1263 (17.0) 1375 (39.0) 1395 (60.2) 1494 (125.2) 1495 (13.6) 1532 (77.1) 1599 (24.0) 2923−2961 (45.5) (37.7) 2972−3034 (84.6) (95.8) 3100−3173 (26.5)

1184 1245

1360 1498 1568 2858 2869 2926 2956 3000−3060 a

(29.2) (29.3) (56.0) (17.9)

Figure 4. IR spectra of Si-PCPDTBT thin films during photooxidation in the SEPAP device: before irradiation (black); after 50 h (blue) and 205 h (red) of photooxidation.

side chain and its corresponding mechanism, before evolution of the units of the conjugated backbone (Si-CPDT and BT) was analyzed. 3.2.1. Alkyl Side Chain Photooxidation. Monitoring the Alkyl Side Chain Evolution. The IR bands associated with the alkyl side chain (2956, 2926, 2869, 2858 cm−1) steadily decrease from the beginning of light exposure, indicating that the moiety is being consumed or modified. It has been repeatedly reported that the side-chains needed to render conjugated polymers soluble in organic solvents are actually the Achilles’ heel of these structures (in terms of photooxidation), being the first moiety to oxidize and subsequently start a chain radical oxidation that leads to the degradation of the conjugated backbone.24,40−42 More specifically, it has been demonstrated that the process starts by hydrogen abstraction on the alkyl chain, notably in the α position to the conjugated backbone, as demonstrated for P3HT.24 Low Molecular Weight Oxidation Products. HS-SPME-GC-MS analysis was performed in order to ascertain the migration of volatile photoproducts and identify their structures. The total ion count (TIC) chromatograms (Figure S5) corroborate the presence of low molecular weight oxidation products in the gas phase for samples irradiated for 70 h, a time that corresponds to a significant loss in thickness and increase in roughness, vide supra. Identification of the main products, displayed in Table 2,

benzothiadiazole and thiophene rings deformations + C4Si ω benzothiadiazole δ thiophene ring + υCS thiophene + υC−Si αCCC rings + δCH alkyl chains νCC rings + δCH alkyl chains δCC thiophene ring + δC−H alkyl chain “breathing” C4Si cycle + δC−H alkyl chain υa,CC C4Si ring + υCC all rings υCC, υCN, ∝CCC benzothiadiazole υCC benzothiadiazole υs υs υa υa

CH2 alkyl side CH3 alkyl side CH2 alkyl side CH3 alkyl side υ C−H

chain chain chain chain

IR intensities (km/mol) are included as values in parentheses.

Exposing the polymer to photooxidative conditions provoked a gradual disappearance of the different chemical functionalities of Si-PCPDTBT, as well as the appearance of new bands corresponding to the oxidation products. Figure 4 shows the IR spectrum in transmission mode of a Si-PCPDTBT film before and after two different irradiation times (50 and 205 h). Subtraction of the initial spectrum (i.e., before irradiation) from the spectra recorded after irradiation permits observation of the global shape of the absorption bands that are formed or consumed (Figure S4). The changes in the initial spectrum include a decrease of the bands related to the alkyl side chain (2858−2956 cm−1): thiophene (1184 cm−1, 881 cm−1), benzothiadiazole (1568 cm−1, 1498 cm−1), and C−Si bond (1245, 881 cm−1). In parallel, new bands appear in the carbonyl (1800−1600 cm−1) and hydroxyl (3600−3000 cm−1) regions in addition to the fingerprint region (1500−500 cm−1). In order to facilitate a complete understanding of the degradation mechanism, the different parts of the polymer have been individually addressed. First, particular attention was paid to the evolution of the alkyl

Table 2. Main Volatile Organic Compounds Released from Irradiated Si-PCPDTBT Films, Identified by HS-SPME-GCMS number

RT/min

identification

1 2 3 4 5

2.5 13.7 15.2 18.6 22.0

3-methylene-heptane 3-heptanone 2-ethylhexanal 3-heptanol 2-ethyl-hexan-1-ol

was carried out by comparison with the software spectral library. No products containing the backbone heteroatoms (S, N) were identified, and the structures of the determined species clearly suggest that their origin is the oxidation of the alkyl chain of the polymer. Chain Oxidation Products. In order to identify the chain oxidation products that remained in the films, we focused on the evolution of the IR spectra in the carbonyl and hydroxyl regions (Figure 5). From the beginning of exposure, the most 2169


Article


linkage (Table S1). During the first 100 h of exposure, there is an increase in the band at 1215 cm−1 that fits the calculated value (1216 cm−1) for the structure XLβ (structure shown as an inset in Figure 6), which involves a C−O−C cross-link through

Figure 6. Evolution of nanohardness values and the IR band at 1215 cm−1 assigned to XLβ structure (inset) during the first 100 h of irradiation.

Figure 5. (a) IR spectral changes in the carbonyl region (1820−1600 cm−1) and (b) in the hydroxyl region (3500−2900 cm−1) region of SiPCPDTBT thin films, caused by photooxidative aging.

the tertiary atom of the alkyl chain. The other two calculated modes associated with this structure are located at 1014 and 784 cm−1, regions that are difficult to monitor due to the convoluted growth of bands. Additionally, there is the appearance of low intensity bands at 1251 and 1267 cm−1, more in accordance with the calculated modes for XLα (Table S1), and suggesting the formation of C−O−C bonds involving carbons with less substituents. Figure 6 displays the increase in the band at 1215 cm−1 (attributed to the formation of the XLβ structure within the polymer matrix) together with the increase in nanohardness, highlighting their good agreement and providing strong evidence of the relationship between the evolution of the chemical structure (cross-linking via ether linkages) upon irradiation and the effect on nanomechanical properties. Alkyl Side Chain Photooxidation Mechanism. As aforementioned, the first step of the degradation process is hydrogen abstraction at the alkyl side chain. It is well-known that the lability of hydrogen atoms depends on the stability of the corresponding radical created, and this is strongly linked to the substitution of the involved carbon atom, i.e., the attached substituents create a stabilizing inductive effect on the formed radical.47 As shown in Figure 1, the α position in Si-PCPDTBT is a secondary carbon (Cα), whereas the beta position is a tertiary carbon (Cβ). Thus, the tertiary carbon of the polymer (Cβ) is expected to be more susceptible to hydrogen abstraction than the secondary (Cα) one. The C−H bond dissociation energies at the Cα and Cβ positions (in relation to Si) were calculated within B3LYP/6-31+G** theory level, showing values of 100.4 and 97.4 kcal/mol for Cα and Cβ, respectively. As expected, the tertiary beta position requires a slightly inferior energy value to dissociate the C−H bond; however, on proposing a degradation mechanism, the formation of radicals at the α position is also possible and should not be excluded. After hydrogen abstraction, and according to the fundamental aspects of the photooxidation of polymers, the prime oxidation products created, namely hydroperoxides, rapidly decompose (thermally or photochemically) to give hydroxyl radicals and macroalkoxy radicals. The latter can further evolve

significant spectral changes were noted in the 1600−1800 cm−1 region, characteristic of CO vibrations. The carbonyl band is centered on 1714 cm−1, usually ascribed to carboxylic acids in the dimer form. In the hydroxyl region, a broad band around 3200 cm−1 again suggests the formation of carboxylic OH groups, and no band attributed to alcohols (usually observed at 3400 cm−1) is observed. Photooxidized samples were then chemically treated with NH3 and SF4 to better identify the nature of the carbonylated products.43,44 NH3 treatment permitted the identification of carboxylic acid species in the sample, as their reaction with NH3 yields ammonium salts with distinctive bands at 1575 cm−1 (υCO asymmetric stretch) and 1475 cm−1 (υCO symmetric stretch). Additionally, the formation of derivative bands around 1660 cm−1 (υCO) and around 3215 cm−1 (υNH) indicated the existence of esters and/or anhydrides that react with NH3 to yield primary amides. The presence of carboxylic acids was also revealed by SF4 treatment. The reaction of saturated carboxylic acids (1710 cm−1) with SF4 provoked the formation and subsequent increase of the band at 1840 cm−1 stemming from the formation of acyl fluorides. Correspondingly, the signals in the hydroxyl region at 3200 cm−1, assigned to carboxylic acids, disappeared after SF4 treatment. Finally, after this same treatment, the existence of a remaining band at 1720 cm−1 indicates that the sample still contains ketones that do not react with SF4. Cross-Linking Reactions. The reported increase in nanohardness (see Figure 3) indicates an increase in stiffness of the polymer, usually associated with cross-linking of the polymer.26,35 To further confirm this hypothesis, modeling was employed to identify possible C−O−C links, which have been shown to be more prevalent than the formation of direct C−C bonds in photooxidative degradation.26,45,46 Theoretical calculations were undertaken to identify the frequencies associated with cross-linking at the alkyl chain positions, Cα and Cβ. The results indicated that C−O−C bonds show frequencies in the range of 1200 to 1260 cm−1 depending on the number of alkyl substituents directly linked to the carbon involved in the ether 2170


Article


Scheme 1. Main Routes Involved in the Oxidation of Si-PCPDTBT Alkyl Side Chain: (a) Mechanisms Starting by Hydrogen Abstraction at Cβ [ΔE(Cβ−H) = 97.4 kcal mol−1]; (b) Mechanisms Starting by Hydrogen Abstraction at Cα [ΔE(Cα−H) = 100.4 kcal mol−1]

through three main routes: (i) formation of alcohols by hydrogen abstraction, (ii) cage reaction to produce chain ketones, and (iii) beta scission processes. It is noteworthy that cage reactions comprise the main route of evolution when oxidation involves secondary carbon atoms, whereas for tertiary carbon atoms, the main route is beta scission, which leads to the formation of acids.17−21

Combining these basic rules of polymer photodegration and the results obtained in this study, we are able to identify the main steps of the Si-PCPDTBT alkyl side-chain photooxidative mechanism (Scheme 1). The oxidation of the alkyl side-chain starting at the most labile position, the tertiary carbon (Cβ), can proceed through cross-linking to yield product XLβ (IR band at 1215 cm−1) or go via chain scission to produce chain end 2171


Article


Table 3. Theoretical Predictions (based on B3LYP/6-31G*) for the Most Characteristic IR Bands for Model Oxidized Backbone Structures, Confronted to Experimental IR Bands Increasing in the Interval of the Predicted Values.a

a

Values marked with an * indicate experimental bands that can be attributed to either BT-sulfoxide or sulfinic ester.

thiophene units of PCDTBT.35 Based on this, modeling was employed to identify the bands associated with these different stages of sulfur oxidation in the two sulfur-containing structures of the polymer, BT and Si-CPDT. Table 3 shows theoretical and experimental data for the selected oxidized structures. Oxidation of the sulfur atom in the BT unit (BT-sulfoxide) was calculated to show one band in the interval 1115−1133 cm−1, and an additional feature at 1317 cm−1 for the sulfonic derivative. Experimental infrared spectra (Figure 7) show bands in these regions, specifically at 1122, 1107, and 1297 cm−1. Good agreement was again observed for the first and second oxidation of the sulfur atoms in the thiophene units, showing the growth of one band at 1051 cm−1 (corresponding to SiCPDT-sulfoxide formation) and bands at 1089 cm−1 and 1215−1234 cm−1 (attributed to Si-CPDT-sulfone). As for P3HT, bands associated with sulfinic esters (617 cm−1) appear in a later stage of degradation, in this case after 70 h of irradiation. It must be noted that the second band related to this structure (1122 cm−1) also fits with the calculated theoretical wavenumber of oxidation of the sulfur in the BT unit. This could explain its growth from the beginning of irradiation (instead of after 70 h like the band at 617 cm−1). According to these data, it seems likely that the same step-bystep sulfur oxidation mechanism proposed for P3HT and PCDTBT is prevalent in Si-PCPDTBT.

carboxylic acids (identified by an IR band centered at 1714 cm−1 and confirmed by SF4 and NH3 treatments) and low molecular weight oxidation products (namely, products 2 and 4 in Table 2, as identified by HS-SPME-GC-MS). On the other hand, oxidation of the secondary carbon atom (Cα) produces chain ketones (IR band at 1720 cm−1) and low molecular weight oxidation products (1, 3, and 5 in Table 2), analyzed by HS-SPME-GC-MS. 3.2.2. Conjugated Backbone Photooxidation. Conjugated Backbone Evolution. Radicals involved in the side-chain photooxidative process can also react with moieties of the conjugated backbone, initiating their oxidative degradation. As observed in Figure 4 (and Figure S4), the characteristic bands of the polymer backbone [benzothiadiazole (1498 cm−1), thiophene (1184 cm−1), C4−Si (1245 cm−1), and the signal at 881°cm−1, corresponding to both Si−C and thiophene moieties] decrease during the irradiation period. In parallel, the appearance of new bands in the fingerprint region is also observed, indicating the transformation of the chemical structures present in the polymer. The oxidation of the thiophene ring in P3HT has been proposed to occur through oxidation of the sulfur atom to produce sulfoxides (R−SO−R), sulfones (R−SO2−R), and finally sulfinic esters (R−SO−OR), by decomposition of the latter.24 The same steps were observed upon oxidation of the 2172


Article


corresponding to the oxidation of the sulfur atoms in BT and DT have been identified from the beginning of the irradiation. These bands indicate that the sulfur steadily evolves leading to sulfoxide, sulfone, and sulfinic species (the latter appearing after 70 h of oxidation). Oxidation of the bridging silicon atom is only observed at very late stages of degradation (360 h of irradiation). Finally, although the appearance of low intensity bands at 1430, 1515, and 1540 cm−1 suggests that nitrogen could also be susceptible to oxidation, the intensity and general evolution of the ensemble of IR bands indicates that the main degradation of the backbone is due to oxidation of thiophene sulfur atoms, as already reported for P3HT and PCDTBT.24,35 3.3. Evolution of Optical and Photophysical Properties. Figure 8a shows the UV−vis spectra recorded upon

Figure 7. Evolution of the IR spectrum of Si-PCPDTBT during photooxidation in the fingerprint region [(a) 1500−1000 cm−1 region; (b), 632−600 cm−1 region], with (i) 0 h, (ii) 20 h, (iii) 50 h, (iv) 110 h, and (v) 205 h of irradiation duration.

Oxidation of the nitrogen of the BT unit (BT-nitroxide) was also investigated in combination with theoretical calculations (Table 3). Accordingly, the band associated with the oxidation of this atom is theoretically located toward 1460 cm−1 if just one nitrogen atom is oxidized, and in the range 1436−1520 cm−1 when oxidation occurs in both nitrogen atoms. In the IR spectra, three low intensity bands can be distinguished at 1430, 1515, and 1540 cm−1. This same IR region was investigated for P3HT and PCDTBT, the latter also containing a benzothiadiazole unit. Similar low intensity bands were also observed in PCDTBT infrared spectra, but not for P3HT. This further suggests that these bands are not associated with thiophene ring oxidation, but to the oxidation of a different structure. However, their intensity is far away from the pronounced evolution of the bands associated with the thiophene rings. In fact, the IR spectra in the late stages of oxidation are clearly similar for these three polymers, showing bands in three main areas (1390, 1110, and 617 cm−1) attributed to the oxidation of the sulfur atoms of the thiophene rings. Finally, modeling was employed to identify oxidation of the silicon atom in the polymer (Si-CPDT-silanol). The characteristic bands of bending and deformation of the Si−OH group are localized between 780 and 900 cm−1, with two principal components at 816 and 822 cm−1. The development of a sharp band at 828 cm−1 was only observed at very late stages of degradation of the polymer, after 360 h of irradiation. Photooxidation Mechanism of the Conjugated Backbone. The decrease in intensity of the characteristic IR bands (see Figure 4 and Figure S4) indicates that all of the moieties are oxidized in parallel. Indicative model oxidized molecules (Table 3) were employed to predict the oxidation of the polymer by combining the evolution of the bands in the IR recorded spectra with theoretical calculations. These proposed structures may differ from the actual products, as the different moieties of the polymer are all oxidized at the same time. That said, good agreement between the calculated modes and the recorded bands indicates that the heteroatoms present in the backbone are indeed susceptible to oxidation. More specifically, vibrations

Figure 8. (a) Evolution of Si-PCPDTBT UV−visible spectra upon photooxidation for (i) 0 h, (ii) 20 h, (iii) 40 h, (iv) 70 h, (v) 90 h, (vi) 110 h, (vii) 130 h, (viii) 160 h, (ix) 205 h, and (x) 280 h. (b) UV−Vis maximum absorbance decay (normalized to the corresponding initial values) for P3HT, PCDTBT, and Si-PCPDTBT exposed to photooxidation.

irradiation of Si-PCPDTBT thin films in the SEPAP 12/24 device under ambient air conditions. Prior to irradiation, the absorption spectrum of the polymer shows a main band centered at 700 nm, characteristic of the π-electronic system, with two maxima located at λ = 690 and 760 nm. Three more energetic bands are observed at λ = 425, 320, and 250 nm. The two different maxima around 700 nm stem from different vibrational transitions and their relative ratio correlates with the crystallinity of the polymer. The initially well-pronounced lowenergy band at 760 nm indicates a very high degree of crystallinity in Si-PCPDTBT films, which is fully in line with the literature.15 Irradiation in the presence of air provokes a continuous decrease of the UV−vis absorption bands indicating that the π-system of the polymer is diminished or destroyed, which finally leads to the complete bleaching of the sample. Remarkably, only a slight blue-shift of the absorbance bands is observed in the latter stages of the degradation. We attribute this evolution to two factors; on the one hand, the crystalline character of the polymer (together with its high molecular 2173


Article

The Journal of Physical Chemistry C weight) creates long conjugation lengths, and thereby minimizes chain scission at random positions of the backbone, as oxygen tends to attack at points where the conjugation is broken (i.e., chain ends or kinks).23,48−50 Consequently, extensive π-conjugation is preserved for a long time during the degradation progress, which in turn delays a blue-shift of the absorption bands. On the other hand, the existence of a depth-dependent degradation may keep intact lower layers of the material as photooxidation proceeds, i.e., upper parts of the layer facing the incident light and oxygen are more strongly oxidized than the lower buried parts closer to the substrate. This assumption is supported by the observed thickness evolution, according to which, after 100 h of exposure (corresponding to marginally less than 50% absorbance loss) the thickness of the sample has already decreased by approximately 25%. Nevertheless, the shape of the absorption spectrum does not change significantly (besides a faster decrease of the band at 760 nm, indicating a reduction in crystallinity), which suggests that intact, long π-conjugated material is still present in the lower part of the sample, after the upper polymer layers have already completely degraded. Finally, and related to this, the thickness decrease cannot solely account for the decrease in absorbance, as the main absorption band at 700 nm decreases faster than the film thickness (see Figure 3 and Figure 8b). The absorbance loss kinetics of Si-PCPDTBT upon photooxidative conditions was then compared to that of previously studied P3HT and PCDTBT.35 Deposits of the three materials exhibiting the same initial absorbance were irradiated under identical conditions, and the absorbance was periodically recorded throughout exposure at the respective maximum absorbance wavelengths. As shown in Figure 8b, the initial velocities, calculated from t = 0 until 50% of absorbance loss, are 0.0256 h−1 for P3HT, 0.0120 h−1 for PCDTBT, and 0.00446 h−1 for Si-PCPDTBT. Accordingly, approximately 50 and 150 h of irradiation were needed to achieve complete bleaching of P3HT and PCDTBT, respectively, whereas SiPCPDTBT was only fully degraded after 280 h. Interestingly, these three polymers share common structural features (thiophene and benzothiadiazole units), with a notable difference being the way in which the alkyl side chain is attached to the polymer backbone in each case (see Figure 1 and Figure S1). In P3HT, for which the hexyl chain is directly linked to the thiophene, degradation has been reported to start by hydrogen abstraction on the α carbon (to the thiophene ring).24 The great lability of this hydrogen atom can be explained in terms of the stabilization of the corresponding radical by delocalization through the conjugated backbone. On the other hand, in PCPDTBT, it has been proven that irradiation provokes direct scission of the bond between the N atom of the carbazole unit and the pendant alkyl side-chain.35 Finally, the alkyl side-chain of Si-PCPDTBT is attached to the CPDT unit through a silicon bridging atom. Thus, our interest focused on understanding the role of Si on the oxidation of the polymer, and more precisely, on the oxidation of the alkyl sidechain identified as the starting point of the degradation. It should be noted that the relative stabilities of the polymers are also influenced by the structural order obtained upon deposition. In this regard, both P3HT and Si-PCPDTBT exhibit rather crystalline properties, which are known to increase stability toward photooxidation, 23,48,49 whereas PCDTBT forms mainly amorphous films.51

Calculations were performed for both Si-PCPDTBT and the analogous polymer C-PCPDTBT, solely differing on the bridging atom, Si or C (Figures 1 and S1). Initially, the relative hydrogen labilities at the Cα and Cβ positions, in relation to the respective bridging atoms, were determined in terms of the C− H bond dissociation energies. Subsequently, the corresponding macroradicals, formed after hydrogen abstraction at the most labile position identified for each polymer, were evaluated in terms of their thermodynamic stability and electron delocalization. The bond dissociation energies reported in Table 4 first indicate that, independent of the bridging atom (Si or C), Cβ− Table 4. Calculated Bond Dissociation Energies and Thermodynamic Stabilities in Si-PCPDTBT and CPCPDTBT within the B3LYP/6-31+G** Theory Level bond dissociation energy (kcal mol‑1) C∝−H Cβ−H thermodynamic stability of R• (kcal mol‑1)

Si-PCPDTBT

C-PCPDTBT

100.4 97.4 −15.3

101.9 98.7 −18.3

H is more labile than Cα−H, as expected, due to their tertiary character. Higher bond dissociation energies, and thus stronger C−H bonds, are observed in the presence of carbon in comparison to the silicon counterpart. Nevertheless, variation in the bond dissociation energies between both polymers is small (+1.5 and +1.3 kcal/mol for Cα and Cβ, respectively). These differences were further explored by examining the stability of macroradicals created after abstraction of the most labile hydrogens, i.e., those in the β position. The corresponding macromolecular stabilities were assessed using DFT calculations for overall energy gains on performing the reaction RH + OH• → R• + H2O.52,53 According to the data reported in Table 3, C-PCPDTBT gives rise to the most stable structure (−18.3 kcal mol−1), thus significantly favoring radical formation. Finally, the spatial delocalization of unshared electrons in the created macroradicals was evaluated. In CPCPDTBT, the unpaired electron is localized with a partition coefficient of 0.59 at the native carbon (from which the hydrogen was abstracted) and 0.16 over the C5 ring. In SiPCPDTBT, the coefficient on the native carbon increases up to 0.74, and sharply decreases to 0.06 over the cycle. This indicates that hydrogen abstraction in C-PCPDTBT produces a more delocalized macroradical, and therefore more stabilized than the equivalent abstraction in Si-PCPDTBT. In fact, the silicon atom blocks the unpaired electron from moving toward the C5 ring54,55 producing a less stable macroradical, and thus disfavoring the general process of hydrogen abstraction. All of these computational data converge to the conclusion that introducing a silicon atom as a bridge to the solubilizing alkyl side-chains is a good strategy to warrant improved photostability of the resulting polymer. This is, in turn, supported by the comparative absorbance decay shown in Figure 8b, which clearly shows an increased resistance to photooxidation for Si-PCPDTBT in relation with other relevant conjugated polymers.

3. CONCLUSIONS The evolution of Si-PCPDTBT thin films under photooxidation conditions has been monitored using several analytical techniques in order to obtain a multiscale analysis 2174



■

from the molecular to the nanomechanical properties. Moreover, combining IR experimental analysis with modeling has allowed accurate elucidation of the photooxidation mechanism of Si-PCPDTBT. The results indicate that the polymer sidechains that continuously degrade from the beginning of irradiation, evolve through the formation of chain oxidation products, cross-linking reactions, and formation and loss of low molecular weight oxidation products. This loss of products results in a high degree of surface modification, a decisive feature for solar cell devices at an interfacial level. On the other hand, the overall photodegradation process in the polymer backbone takes place via oxidation of the sulfur units of the thiophene rings, although, bands related to the oxidation of the N and Si atoms could also be identified. The degradation rate of Si-PCPDTBT measured in terms of its UV−vis absorbance decay is found to be slower than for other commonly employed p-type polymers, such as P3HT and PCDTBT. In order to explain this, the pertinence of employing silicon as a bridging atom for the solubilizing alkyl chains was approached by comparing the macroradicals created upon hydrogen abstraction in the alkyl chain of both Si-PCPDTBT and the analogous C-PCPDTBT. The calculations revealed that the inclusion of Si as a bridging atom appears to be an interesting structural strategy when aiming for polymers with improved photostabilities.

■

REFERENCES

(1) Lewis, J. Material Challenge for Flexible Organic Devices. Mater. Today 2006, 9, 38−45. (2) Nickel, F.; Haas, T.; Wegner, E.; Bahro, D.; Salehin, S.; Kraft, O.; Gruber, P. A.; Colsmann, A. Mechanically Robust, ITO-Free, 4.8% Efficient, All-Solution Processed Organic Solar Cells on Flexible PET Foil. Sol. Energy Mater. Sol. C 2014, 130, 317−321. (3) Carlé, J. E.; Helgesen, M.; Madsen, M. V.; Bundgaard, E.; Krebs, F. C. Upscaling from Single Cells to Modules - Fabrication of Vacuumand ITO-Free Polymer Solar Cells on Flexible Substrates with Long Lifetime. J. Mater. Chem. C 2014, 2, 1290−1297. (4) Andersen, T. R.; et al. Scalable, Ambient Atmosphere Roll-to-Roll Manufacture of Encapsulated Large Area, Flexible Organic Tandem Solar Cell Modules. Energy Environ. Sci. 2014, 7, 2925−2933. (5) Bundgaard, E.; Krebs, F. C. Low Band Gap Polymers for Organic Photovoltaics. Sol. Energy Mater. Sol. C 2007, 91, 954−985. (6) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Enhanced Power-Conversion Efficiency in Polymer Solar Cells Using an Inverted Device Structure. Nat. Photonics 2012, 6, 591−595. (7) Liu, S.; Zhang, K.; Lu, J.; Zhang, J.; Yip, H.-L.; Huang, F.; Cao, Y. High-Efficiency Polymer Solar Cells via the Incorporation of an Amino-Functionalized Conjugated Metallopolymer as a Cathode Interlayer. J. Am. Chem. Soc. 2013, 135, 15326−15329. (8) You, J.; et al. A Polymer Tandem Solar Cell with 10.6% Power Conversion Efficiency. Nat. Commun. 2013, 4, 1446. (9) Jørgensen, M.; Norrman, K.; Krebs, F. C. Stability/Degradation of Polymer Solar Cells. Sol. Energy Mater. Sol. C 2008, 92, 686−714. (10) Grossiord, N.; Kroon, J. M.; Andriessen, R.; Blom, P. W. M. Degradation Mechanisms in Organic Photovoltaic Devices. Org. Electron. 2012, 13, 432−456. (11) Chen, J.; Cao, Y. Development of Novel Conjugated Donor Polymers for High-Efficiency Bulk-Heterojunction Photovoltaic Devices. Acc. Chem. Res. 2009, 42, 1709−1718. (12) Su, Y.-W.; Lan, S.-C.; Wei, K.-H. Organic Photovoltaics. Mater. Today 2012, 15, 554−562. (13) Mühlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C. High Photovoltaic Performance of a LowBandgap Polymer. Adv. Mater. 2006, 18, 2884−2889. (14) Scharber, M. C.; et al. Influence of the Bridging Atom on the Performance of a Low-Bandgap Bulk Heterojunction Solar Cell. Adv. Mater. 2010, 22, 367−370. (15) Morana, M.; et al. Nanomorphology and Charge Generation in Bulk Heterojunctions Based on Low-Bandgap Dithiophene Polymers with Different Bridging Atoms. Adv. Funct. Mater. 2010, 20, 1180− 1188. (16) Distler, A.; Kutka, P.; Sauermann, T.; Egelhaaf, H.-J.; Guldi, D. M.; Di Nuzzo, D.; Meskers, S. C. J.; Janssen, R. A. J. Effect of PCBM on the Photodegradation Kinetics of Polymers for Organic Photovoltaics. Chem. Mater. 2012, 24, 4397−4405. (17) McKellar, J. F.; Allen, N. S. Photochemistry of Man-Made Polymers; Elsevier Applied Science Publishers, Ltd: London, 1979. (18) Grassie, N.; Scott, G. Polymer Degradation and Stabilization; Cambridge University Press: Cambridge, 1985. (19) Hamid, S. H.; Amin, M. B.; Maadhah, A. G. Handbook of Polymer Degradation; Marcel Dekker, Inc.: New York, 1992. (20) Rabek, J. F. Photodegradation of Polymers; Springer: Berlin, 1996. (21) Gardette, J.-L. Fundamental and Technical Aspects of the Photo-Oxidation of Polymers. Handbook of Polymer Degradation; Marcel Dekker, Inc.: New York, 2000. (22) Rivaton, A.; Tournebize, A.; Gaume, J.; Bussière, P.-O.; Gardette, J.-L.; Therias, S. Photostability of Organic Materials Used in Polymer Solar Cells. Polym. Int. 2014, 63, 1335−1345. (23) Hintz, H.; Egelhaaf, H. J.; Lüer, L.; Hauch, J.; Peisert, H.; Chassé, T. Photodegradation of P3HT−A Systematic Study of Environmental Factors. Chem. Mater. 2010, 23, 145−154. (24) Manceau, M.; Rivaton, A.; Gardette, J.-L.; Guillerez, S.; Lemaître, N. The Mechanism of Photo- and Thermooxidation of Poly(3-Hexylthiophene) (P3HT) Reconsidered. Polym. Degrad. Stab. 2009, 94, 898−907.

ASSOCIATED CONTENT

* Supporting Information S

Structures of the other p-type polymers under discussion, comparison between data sets corresponding to photooxidation of Si-PCPDTBT films using SEPAP 12/24 and Suntest AM 1.5, subtracted IR spectra of Si-PCPDTBT films after exposure, total ion count (TIC) chromatogram of the gas phase of SiPCPDTBT films after irradiation, and structures and IR bands for proposed products of cross-linking. This material is available free of charge via the Internet at http://pubs.acs.org.

■

Article

AUTHOR INFORMATION

Corresponding Authors

*Agnès Rivaton: E-mail: [email protected]. Phone: +33 473 40 77 43. Fax: +33 473 40 77 00. *Didier Bégué: E-mail: [email protected]. Phone: +33 559 40 78 52. Fax: +33 559 40 78 62. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Computer time for this study was provided by the computing facilities at the Mésocentre de Calcul Intensif Aquitaine (MCIA) of the Université de Bordeaux and at the Université de Pau et des Pays de l′Adour. IFD thanks Pascal Wong Wah Chung for the HS-SPME-GC-MS experiments and Aurélien Tournebize for P3HT and PCDTBT photooxidation data. The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2011 under Grant Agreement ESTABLIS No. 290022). 2175


Article

The Journal of Physical Chemistry C (25) Gaume, J.; Wong-Wah-Chung, P.; Rivaton, A.; Therias, S.; Gardette, J.-L. Photochemical Behavior of PVA as an Oxygen-Barrier Polymer for Solar Cell Encapsulation. RSC Adv. 2011, 1, 1471−1481. (26) Bussière, P.-O.; Rivaton, A.; Thérias, S.; Gardette, J.-L. Multiscale Investigation of the Poly(N-Vinylcarbazole) Photoageing Mechanism. J. Phys. Chem. B 2011, 116, 802−812. (27) Bégué, D.; Gohaud, N.; Pouchan, C.; Cassam-Chenaï, P.; Liévin, J. A Comparison of Two Methods for Selecting Vibrational Configuration Interaction Spaces on a Heptatomic System: Ethylene Oxide. J. Chem. Phys. 2007, 127, 164115. (28) Hehre, W. J.; D, R.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XII. Further Extensions of Gaussian-Type Basis Sets for Use in Molecular Orbital Studies of Organic Molecules. J. Chem. Phys. 1972, 56, 2257−2261. (29) Dill, J. D.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XV. Extended Gaussian-Type Basis Sets for Lithium, Beryllium, and Boron. J. Chem. Phys. 1975, 62, 2921−2923. (30) Francl, M. M. P.; W, J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XXIII. A Polarization Type Basis Set for Second-Row Elements. J. Chem. Phys. 1982, 77, 3654−3665. (31) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652. (32) Frisch, M. J. et al. Gaussian 09, revision D.01; Gaussian Inc.: Wallingford, CT, 2009. (33) Löwdin, P.-O. On the Non-Orthogonality Problem Connected with the Use of Atomic Wave Functions in the Theory of Molecules and Crystals. J. Chem. Phys. 1950, 18, 365−375. (34) Löwdin, P.-O. On the nonorthogonality problem. Adv. Quantum Chem. 1970, 5, 185−199. (35) Tournebize, A.; Bussière, P.-O.; Wong-Wah-Chung, P.; Thérias, S.; Rivaton, A.; Gardette, J.-L.; Beaupré, S.; Leclerc, M. Impact of UVVisible Light on the Morphological and Photochemical Behavior of a Low-Bandgap Poly(2,7-Carbazole) Derivative for Use in HighPerformance Solar Cells. Adv. Eng. Mater. 2013, 3, 478−487. (36) Rosch, R.; et al. Investigation of the Degradation Mechanisms of a Variety of Organic Photovoltaic Devices by Combination of Imaging TechniquesThe ISOS-3 Inter-Laboratory Collaboration. Energy Environ. Sci. 2012, 5, 6521−6540. (37) Oliver, W. C.; Pharr, G. M. An Improved Technique for Determining Hardness and Elastic Modulus Using Load and Displacement Sensing Indentation Experiments. J. Mater. Res. 1992, 7, 1564−1583. (38) Bégué, D.; Elissalde, S.; Pere, E.; Iratcabal, P.; Pouchan, C. New Theoretical and Experimental Infrared Results on Formaldehyde in Solution. J. Phys. Chem. A 2006, 110, 7793−7800. (39) Bégué, D.; Qiao, G. G.; Wentrup, C. Nitrile Imines: Matrix Isolation, IR Spectra, Structures, and Rearrangement to Carbodiimides. J. Am. Chem. Soc. 2012, 134, 5339−5350. (40) Chambon, S.; Rivaton, A.; Gardette, J.-L.; Firon, M.; Lutsen, L. Aging of a Donor Conjugated Polymer: Photochemical Studies of the Degradation of Poly[2-Methoxy-5-(3′,7′-Dimethyloctyloxy)-1,4-Phenylenevinylene]. J. Polym. Sci., Polym. Chem. 2007, 45, 317−331. (41) Manceau, M.; Gaume, J.; Rivaton, A.; Gardette, J.-L.; Monier, G.; Bideux, L. Further Insights into the Photodegradation of Poly(3Hexylthiophene) by Means of X-Ray Photoelectron Spectroscopy. Thin Solid Films 2010, 518, 7113−7118. (42) Rivaton, A.; Chambon, S.; Manceau, M.; Gardette, J.-L.; Lemaître, N.; Guillerez, S. Light-Induced Degradation of the Active Layer of Polymer-Based Solar Cells. Polym. Degrad. Stab. 2010, 95, 278−284. (43) Carlsson, D. J.; Brousseau, R.; Zhang, C.; Wiles, D. M. Identification of Products from Polyolefin Oxidation by Derivatization Reactions. Chemical Reactions on Polymers; American Chemical Society: Washington, DC, 1988, 364, 376−389. (44) Wilhelm, C.; Gardette, J.-L. Infrared Identification of Carboxylic Acids Formed in Polymer Photooxidation. J. Appl. Polym. Sci. 1994, 51, 1411−1420.

(45) Rivaton, A.; Mailhot, B.; Derderian, G.; Bussiere, P. O.; Gardette, J. L. Investigation of the Photophysical Processes and Photochemical Reactions Involved in PVK Films Irradiated at λ > 300 nm. Macromolecules 2003, 36, 5815−5824. (46) Bussière, P. O.; Mailhot, B.; Rivaton, A.; Barthe, M. F.; Gardette, J. L.; Baba, M. Photocrosslinking of Poly(N-vinylcarbazole): Implementing a Complementary Set of Techniques to Characterize the Three-Dimensional Network Formation. Polym. Degrad. Stab. 2008, 93, 1376−1382. (47) Lowry, T. H. R., K, S. Mechanism and Theory in Organic Chemistry; Harper & Row: New York, 1987. (48) Madsen, M. V.; Tromholt, T.; Böttiger, A.; Andreasen, J. W.; Norrman, K.; Krebs, F. C. Influence of Processing and Intrinsic Polymer Parameters on Photochemical Stability of Polythiophene Thin Films. Polym. Degrad. Stab. 2012, 97, 2412−2417. (49) Dupuis, A.; Wong-Wah-Chung, P.; Rivaton, A.; Gardette, J.-L. Influence of the Microstructure on the Photooxidative Degradation of Poly(3-Hexylthiophene). Polym. Degrad. Stab. 2012, 97, 366−374. (50) Hintz, H.; Egelhaaf, H. J.; Peisert, H.; Chassé, T. PhotoOxidation and Ozonization of Poly(3-Hexylthiophene) Thin Films as Studied by UV/Vis and Photoelectron Spectroscopy. Polym. Degrad. Stab. 2010, 95, 818−825. (51) Alem, S.; et al. Effect of Mixed Solvents on PCDTBT:PC70BM Based Solar Cells. Org. Electron. 2011, 12, 1788−1793. (52) Bolland, J. L.; Gee, G. Kinetic Studies in the Chemistry of Rubber and Related Materials. II. The Kinetics of Oxidation of Unconjugated Olefins. Trans. Faraday Soc. 1946, 42, 236−243. (53) Audouin, L.; Langlois, V.; Verdu, J.; de Bruijn, J. C. M. Role of Oxygen Diffusion in Polymer Ageing: Kinetic and Mechanical Aspects. J. Mater. Sci. 1994, 29, 569−583. (54) Mark, J. E.; Allcock, H. R.; West, R. Inorganic Polymers; Prentice Hall: Upper Saddle River, NJ, 1992. (55) Miller, R. D.; Michl, J. Polysilane High Polymers. Chem. Rev. 1989, 89, 1359−1410.

2176


Unravelling the Photodegradation Mechanisms of a Low Bandgap

Recommend Documents