Ionizing radiation induced degradation of poly (2-methoxy-5-(2’-ethyl- hexyloxy) -1,4-phenylene vinylene) in solution E. S. Bronze-Uhle, A. Batagin-Neto, F. C. Lavarda, and C. F. O. Graeff Citation: J. Appl. Phys. 110, 073510 (2011); doi: 10.1063/1.3644946 View online: http://dx.doi.org/10.1063/1.3644946 View Table of Contents: http://jap.aip.org/resource/1/JAPIAU/v110/i7 Published by the AIP Publishing LLC. Additional information on J. Appl. Phys. Journal Homepage: http://jap.aip.org/ Journal Information: http://jap.aip.org/about/about_the_journal Top downloads: http://jap.aip.org/features/most_downloaded Information for Authors: http://jap.aip.org/authors Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions http://jap.aip.org/?ver=pdfcov http://oasc12039.247realmedia.com/RealMedia/ads/click_lx.ads/www.aip.org/pt/adcenter/pdfcover_test/L-37/932441298/x01/AIP-PT/JAP_CoverPg_0513/AAIDBI_ad.jpg/6c527a6a7131454a5049734141754f37?x http://jap.aip.org/search?sortby=newestdate&q=&searchzone=2&searchtype=searchin&faceted=faceted&key=AIP_ALL&possible1=E. S. Bronze-Uhle&possible1zone=author&alias=&displayid=AIP&ver=pdfcov http://jap.aip.org/search?sortby=newestdate&q=&searchzone=2&searchtype=searchin&faceted=faceted&key=AIP_ALL&possible1=A. Batagin-Neto&possible1zone=author&alias=&displayid=AIP&ver=pdfcov http://jap.aip.org/search?sortby=newestdate&q=&searchzone=2&searchtype=searchin&faceted=faceted&key=AIP_ALL&possible1=F. C. Lavarda&possible1zone=author&alias=&displayid=AIP&ver=pdfcov http://jap.aip.org/search?sortby=newestdate&q=&searchzone=2&searchtype=searchin&faceted=faceted&key=AIP_ALL&possible1=C. F. O. Graeff&possible1zone=author&alias=&displayid=AIP&ver=pdfcov http://jap.aip.org/?ver=pdfcov http://link.aip.org/link/doi/10.1063/1.3644946?ver=pdfcov http://jap.aip.org/resource/1/JAPIAU/v110/i7?ver=pdfcov http://www.aip.org/?ver=pdfcov http://jap.aip.org/?ver=pdfcov http://jap.aip.org/about/about_the_journal?ver=pdfcov http://jap.aip.org/features/most_downloaded?ver=pdfcov http://jap.aip.org/authors?ver=pdfcov Ionizing radiation induced degradation of poly (2-methoxy-5-(2’-ethyl-hexyloxy) -1,4-phenylene vinylene) in solution E. S. Bronze-Uhle, A. Batagin-Neto, F. C. Lavarda, and C. F. O. Graeff Department of Physics, FC-UNESP, Av. Eng. Luiz Edmundo Carrijo Coube 14-01, 17033-360 Bauru, Brazil (Received 29 April 2011; accepted 18 August 2011; published online 6 October 2011) In this paper we investigate the causes of the chromatic alteration observed in chloroform solutions of poly (2-methoxy-5-(20-ethyl-hexyloxy)-1,4-phenylene vinylene) (MEH-PPV) after gamma ray irradiation. Structural and chemical changes were analyzed by gel permeation chromatography, fourier transform infrared spectroscopy, and proton nuclear magnetic resonance techniques and complemented by electronic structure calculations. The results indicate chlorine incorporation in the polymer structure and main chain scission after irradiation. Based on our findings we propose that the main mechanism for the blue-shifts, observed in the UV-Vis absorption spectra of MEH-PPV after irradiation, is the result of a radical attack on the polymer main chain. Gamma rays generate radicals, �Cl and �CHCl2 from chloroform radiolysis that attack preferentially the vinyl double bonds of the polymer backbone, breaking the electronic conjugation and eventually the chain. Our results indicate that oxygen does not play a major role in the effect. Electronic spectra simulations were performed based on these assumptions reproducing the UV-Vis experimental results. VC 2011 American Institute of Physics. [doi:10.1063/1.3644946] I. INTRODUCTION Ionizing radiation is commonly used in fields such as medicine and civil construction.1 On the other hand, inspired by the health damages associated with its absorption by the human body, there is a ceaseless interest in the research com- munity in the development of new methods or materials that can measure dose, particularly in what concerns low doses.2 In this regard, polymers and specially semiconducting poly- mers have been described as promising materials for the fab- rication of low cost dosimeters.3–7 In particular, it was reported that poly (2-methoxy-5-(20- ethyl-hexyloxy)-1,4-phenylene vinylene) (MEH-PPV) in halo- genated aliphatic solvents, could be used as low dose dosime- ters, <10 Gy.4 In Ref. 4 it is shown that MEH-PPV in solution after exposure to ionizing radiation, display a blue-shift in its main absorption peak proportional to the applied dose. On the other hand the same material in film form or in other solvents, such as toluene, are insensitive to c-rays up to doses of approxi- mately 100 Gy.4,8 The blue-shifts were explained as a decrease in the polymer conjugation length, induced by irradiation and that the solvent had a major role in the effect. However no detailed polymer degradation mechanisms were proposed. Polymer degradation is a topic of high interest since most organic materials are subject to decay under excitation, like for example under UV illumination.9,10 In the case of semi- conducting polymers this topic is of particular relevance, since devices based on these materials have their performance strongly affected by photo-oxidation.9–13 It is well established in the literature that in these systems oxygen and water plays a crucial role in the polymer degradation.10,11 In the case of PPV derivatives, the main reaction mechanism is believed to be the addition of singlet oxygen in the vinyl group14 or the abstraction of labile hydrogen atom in the a position of the ether. This initiates the chain oxidation by fixation of oxygen, involving peroxyl and alkyl radicals.11 Another possibility is the formation of radical cation of the PPV derivative in the presence of oxygen.11 In this case, (O2 ��) cannot be consid- ered as responsible for the propagation of the oxidation. The main role of oxygen is to favor the formation of the PPV�þ radical cations. The radical cations are likely to initiate a chain oxidation process by fixation of fundamental O2 or by reacting with superoxide anion (O2 ��) as described by Scoponi et al.15 In both cases, the abstraction of labile hydrogen in the ether position occurs. There is a vast literature in these subject, however not when ionizing radiation is used.10–13,16,17 The effects of gamma radiation observed in polymers are diverse and depend on several factors, such as tempera- ture, chemical environment, chemical structure, as well as the polymer physical state.18–20 For example, the irradiation of polyethylene under inert atmosphere results in cross- linking.21 On the other hand, if the atmosphere during irradi- ation contains reactive species, such as oxygen, scission reactions are observed, due to the action of peroxide.18,21–23 In this case photons are absorbed by the polymer creating radicals that react with oxygen or peroxide that induce scis- sion.20 As mentioned in the case of MEH-PPV in solution it was proposed that the solvent had a major role in the poly- mer degradation. It is well known that halide solvents under ionizing irradiation do form radicals,24 however the effect of these radicals on conjugated polymers have not been studied systematically. In this work, we present a detailed investiga- tion on the effects of ionizing radiation on the chemical structure of MEH-PPV in halogenated solvents. II. MATERIALS AND METHODS A. Experimental Poly (2-methoxy-5-(20-ethyl-hexyloxy)-1,4-phenylene vi- nylene) (MEH-PPV) with mean molecular weight from 70000 to 100000 was purchased from ALDRICH (Milwaukee, USA). 0021-8979/2011/110(7)/073510/9/$30.00 VC 2011 American Institute of Physics110, 073510-1 JOURNAL OF APPLIED PHYSICS 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions http://dx.doi.org/10.1063/1.3644946 http://dx.doi.org/10.1063/1.3644946 http://dx.doi.org/10.1063/1.3644946 The main polymer structure is shown in Fig. 1. Chloroform, CHCl3, was purchased from ACROS (New Jersey, USA). Polymer, reagents, and solvents were used as received. MEH-PPV solutions were prepared at room temperature under weak illumination conditions to avoid photo-degrada- tion.25 Solutions were kept in 4 mL glass recipients (Wheaton 13-425) and irradiated with a 60Co gamma ray source (1.25 MeV, SIEMENS GAMATRON S-80) with appropriate acrylic build-ups. The concentrations used ranged from 0.0125 mg/mL to 0.225 mg/mL. All solutions have been prepared at room tem- perature under low illumination conditions to avoid undesir- able photoreactions and stored in a refrigerator (T¼ 283 K). For each concentration five samples have been prepared, one was kept as a reference, while the other four irradiated in pairs in order to test reproducibility. The samples have been irradiated at room temperature in a cobalt therapy unit (CGR - Model Alcyon II) with appropriate acrylic build-ups (thick- ness of 0.5 cm), at a rate of 0.5 Gy/min. The doses used were 30 and 90 Gy. The UV/Vis absorption spectra of MEH-PPV solutions were measured using a SHIMADZU UVMINI 1240 spec- trometer. Infrared spectra were obtained from MEH-PPV films that were deposited directly in the ZSM-5 cells. For the FTIR measurements a Bruker Vertex 70 Fourier Transform spectrometer was used. Gel permeation chromatography (GPC) measurements were performed on an Agilent 1100 se- ries system with PSS-SDV (5 lm, 100 and 1000 Å) columns, THF as eluent (at 35 �C, flow rate 1 mL/min) and calibration using a PSS-ReadyCal Kit of poly(styrene) (PS) (MP¼ 480 - 2500000 Da) standards. For sample preparation, 50 lg/ml solutions of MEH-PPV in chloroform were irradiated with 20 Gy of gamma radiation and subsequently the solvent was completely evaporated. 2 mg of MEH-PPV (irradiated or non-irradiated, from the same batch) was dissolved in 1 mL THF and 100 lL of this solution was injected. Data acquisi- tion and analysis were carried on WinGPC software from Polymer Standards Service GmbH (PSS). 1H and 13C NMR were obtained in using a 400 and 500 MHz spectrometer from Bruker DRX 400. For these meas- urements, the MEH-PPV solutions were evaporated under reduced pressure and then dissolved in deuterated chloro- form (CDCl3) (99.8% chloroform-d – Aldrich) at a concen- tration of 10 mg/mL. The chemical shifts were referenced to tetramethylsilane (TMS) and chloroform-d1 in the case of H- NMR and C-NMR, respectively. Additionally, gradient- selected H, C shift correlation experiments (gs-H, C HSQC) were carried out. B. Electronic structure calculations The structure optimization was performed with a re- stricted Hartree-Fock (RHF) approach using the PM3 semi- empirical method implemented in the MOPAC 2007 software package.26 Initially the steric interaction between the lateral branches of adjacent units were investigated. For that purpose the conformation of these branches were opti- mized by molecular dynamics calculations of tetramers. Larger structures were made using the lowest energy confor- mation obtained from the tetramers study. The solvent was simulated by the COSMO continuum method.27 The reactivity study was performed using Fukui indices analysis. The Fukui function, f ð~rÞ, describes how the elec- tron density varies in response to electrons removal or addi- tion in a system, with the geometry fixed. Within an adiabatic approximation, one can assume that a variation in the electrons number is directly related to a change in the frontier orbitals electron density of the system (HOMO and LUMO). Thus, it carries information about how the frontier orbitals are modified by varying the electrons number, and has been employed in order to understand and predict the reactivity variation in different sites of a molecule.28–30 Practical ways to calculate these indices, proposed by Yang and Mortier,31 are known as the condensed-to-atoms Fukui indices (CAFI). Given a system M with N electrons whose reactivity we want to study, when receiving an elec- tron the system becomes M�1, with (Nþ 1) electrons and losing an electron we have Mþ1, with (N-1) electrons. The CAFI is then expressed by fþk ¼ qk N þ 1ð Þ � qk Nð Þ nucleophilic attack f�k ¼ qk Nð Þ � qk N � 1ð Þ electrophilic attack f 0 k ¼ 1 2 qk N þ 1ð Þ � qk N � 1ð Þ½ � radical attack where qk(Nþ 1), qk(N) and qk(N�1) represent the electron population on the kth atom in the species M-1, M, and Mþ1, respectively. The functions f þ, f,� and f 0 are used for reac- tions analysis involving nucleophilic, electrophilic, and free radical attacks, respectively. The higher the values of these functions the higher the reactivity of the site. It is well known that in general polymers do not present a completely planar structure; the main chain can twist intro- ducing random physical defects, which prevents the p system to extend throughout the structure.32 For MEH-PPV, FIG. 1. (Color online) UV-Vis absorption spectra of 0.0375 mg/mL MEHPPV solutions in chloroform, before (solid line) and after irradiation with 30 Gy (dashed line) and 90 Gy (dotted line). 073510-2 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions especially, evidence shows that the polymer chains can be considered as a collection of approximately planar segments with different conjugation lengths, with an average of 10 repeating units.33 Thus the CAFI study was performed using a MEH-PPV decamer. The indices were calculated using density functional theory (DFT) with Becke’s LYP (B3LYP/VWN5) exchange- correlation functional; for all atoms, 6-31 G basis set was adopted. To calculate the electronic population of the neutral species, we used the restricted Kohn-Sham (RKS) approach. For charged species, with the insertion and removal of an electron, we used the restricted open-shell Kohn-Sham (ROKS) approach, to avoid spin contamination.34,35 The electronic populations were obtained by charge cal- culation with electrostatic potential (ESP) fitting35 and Mul- liken population analysis,30 the solvent was again simulated by the COSMO continuum method. Calculations were per- formed with the Firefly/PCGAMESS computer package.36 The theoretical optical absorption spectra for all structures were calculated with the Orca software package.37 We used a RHF approach with the ZINDO/S semiempirical method in conjunction with the CIS method. Such methods have proved suitable for vertical excitation energies of medium and large molecules, including transition-metals compounds.35 III. RESULTS Figure 1 shows the absorption spectra of samples with 0.0375 mg/mL, before and after irradiation with doses of 30 and 90 Gy. The main peak in the non-irradiated solutions, at 500 nm, is attributed to the transition between the MEH-PPV frontier orbitals of p-p*.35 A second peak at 330 nm is also seen, characteristic of polyphenylene derivatives, and attrib- uted to forbidden transitions allowed by symmetry break- ing.38 After irradiation, a blue-shift in the main peak position is observed as expected.4 As mentioned in the Introduction, the shift in the main peak is assigned to a decrease in conjugation length. One source of the decrease in conjugation length is chain scis- sion, which can be measured using gel permeation chroma- tography (GPC), see Fig. 2. In Fig. 2, the eluogramm of MEH-PPV before and after irradiation is presented. The eluogramm displays a broad peak at around 11 mL which shifts to higher volumes after irradiation, this corresponds to a decrease in molecular weight, from Mn¼ 61000 to about Mn¼ 43000 g/mol. The peak at low molecular weights (higher volumes) comes from FIG. 2. GPC eluogramm of non-irradiated (solid line) and irradiated (dashed line) MEH-PPV solutions. FIG. 3. (Color online) FTIR spectra of MEH-PPV chloroform solution of non- irradiated (solid line), irradiated solution with 30 Gy (dashed line) and irradiated with 90 Gy (dotted line). TABLE I. IR band positions of non-irradiated MEH-PPV. Wavenumber (cm�1) Assignment 3065 CH – stretching vinyl 2956 CH3 asymmetric stretching 2919 CH – stretching 2850 CH2 stretching 1506 Semicircular phenyl stretch (C-C aromatic) 1492 Semicircular phenyl stretch (C-C aromatic) 1464 Anti-symmetric alkyl CH2 stretch 1412 Semicircular phenyl stretch 1379 Symmetric alkyl CH2 deformation 1204 Phenyl oxygen stretch 1078 Alkyl oxygen stretch 1039 Alkyl oxygen stretch 970 trans-double bond CH-wag (vinyl group) 860 1,2,5,6 Tetra-substituted benzene ring bend 073510-3 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions impurities, probably smaller MEH-PPV segments left over from the synthesis. Remember that no purification step was employed, and the MEH-PPV was used as bought. We do not believe that such impurities have a significant role in the modifications observed after gamma irradiation, as will be discussed later. The shift of the main GPC peak to higher volumes is a clear indication that MEH-PPV suffered scis- sion of its backbone chain upon irradiation. Figure 3 shows the FTIR spectra of MEH-PPV solutions before and after irradiation (0.225 mg/mL, 30 and 90 Gy). The main bands and their assignments for the non irradiated polymer are listed in Table I. New bands at 648, 729, and 906 cm�1 can be observed after irradiation, which are clear indications that halogen atoms are added to the polymer chain. The spectra are normalized to the Ph-OR stretching band at 1204 cm�1 to remove the influence of small differen- ces in the baseline.39 This band was chosen since it does not change after irradiation.12,13 Notice that weak bands can be seen at 1680–1720 cm�1. We believe that these bands are due to C¼O and appeared as a result of MEH-PPV manipu- lation and not as a result of gamma irradiation, as will be dis- cussed later on. The oxidation, as a result of sample manipulation, is commonly reported in the literature.40 Figure 4 shows the 1H NMR spectrum of non-irradiated MEH-PPV in chloroform. These spectra match what is reported in the literature with peaks found in 7.52 (2 H, d, CH), 7.19 (2 H, s, CH), 3.94-3.97 (5 H, m, OCH2 and OCH3), 1.15-1.90 (9 H, m, CH2 and CH), and 0.82-1.05 ppm (6 H, m, CH3).14 Signals related to structural defects from the polymer synthesis are also shown at low proportions FIG. 4. 1H NMR spectra of MEH-PPV in CDCl3 solution before irradiation. FIG. 5. 1H NMR spectra of MEH-PPV in CDCl3 solution after irradiation with 30 Gy. 073510-4 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions (2.4 – 3.2, 4.4 – 4.5 e 5.9 – 6.3 ppm) and may be associated with non-conjugated portions of the polymer chain.40,41 After irradiation, as presented in Fig. 5, new signals are observed in different regions of the spectrum with low inten- sity (2.5-3.5, 4.5-5.5, and 5.5-6.5 ppm). The initial structural defects are still present with small alterations. The spectral changes give us a strong indication of structural changes in the polymer chain. For instance, the signals at 4.70, 5.09, and 5.53 ppm are typical of protons adjacent to chlorine atoms, as will be discussed in more detail in the next section. The signals at 9.85 and 10.45 ppm are typical of MEH-PPV oxidation after manipulation of irradiated samples. The analysis of the 1H NMR spectrum suggests that most of the impurities observed in Fig. 2 are due to smaller fragments of the polymer. The signs of these structures can not be resolved in the 1H NMR spectrum since they present the same shifts of larger polymer chains. 13C NMR spectra signals obtained were very weak, due to sample size, even with acquisition times of 13 h. The gs- H, C-HSQC spectrum was also performed, but for the same reasons noted for the 13C NMR, it has not provided a deeper insight into the study (see Supplementary Information for details42). IV. DISCUSSION As mentioned, the effects of gamma radiation on poly- mers are diverse and depend mainly on the polymer structure and the chemical environment in which the irradiation occurs. The action of ionizing radiation is known to modify the struc- ture of conjugated polymers. It may induce cross-linking or scission of backbone chains, changing the number and the na- ture of the double bonds. The cross-linking reactions lead to the formation of inter- and intramolecular bonds, which increase the molecular weight. In contrast, scission reactions lead to polymer chain cleavage, decreasing the molecular weight and the formation of several sub-products.43 In our case we have not observed any direct evidence of cross-linking reactions, however in some cases for very high doses, we could observed that the polymer solution became a viscous solution after mild heat treatments. As the solvent evaporated instead of having simply the polymer powder, a gel was formed. As can be seen in Fig. 2, the decrease in mo- lecular weight, from Mn¼ 61000 to about Mn¼ 43000g/ mol, clearly shows that the polymer backbones did break af- ter irradiation. The scission can be associated with the attack of reactive chemical species or the direct interaction of the polymer chain with the ionizing radiation. We will consider first the latter hypothesis. According to Chambon et al.11 the irradiation can initiate polymer degra- dation by abstraction of the labile hydrogen atom in the a position of the ether function, with subsequent radical reac- tions. Another possibility is the homolytical scission of the –O–CH2 bond (induced by irradiation), which promotes the formation of two radicals that immediately recombine or add on double bonds causing the conjugation reduction.12,13 This hypothesis is not compatible with our IR spectra (see Fig. 3). The IR spectra of our samples do not show a decrease in the bands assigned to the ether at 1205, 1078, and 1035 cm�1, as discussed by Chambon et al.12,13 Moreover, as already men- tioned in the Introduction, MEH-PPV films are insensitive to c-irradiation. More importantly, MEH-PPV dissolved in tol- uene is also insensitive to c-irradiation.8 Thus in the follow- ing discussion we shall not consider the direct absorption of c-rays by the polymer chain as the main mechanism of degradation. We shall now consider the hypothesis of the formation of reactive species as a consequence of irradiation that attacks the polymer. As already mentioned, the most well- known polymer photo-degradation mechanism in conjugated polymers involves chain oxidation. The presence of oxygen and water during irradiation favors scission reactions18 due to the oxidative activity of peroxyl and alcoxyl radicals in PPV derivatives.20 These reactions produce aromatic ketone, aromatic aldehyde, alcohol, and carboxylic acid, as main by- products and frequently are responsible for the conjugation reduction in these systems, promoting a blue-shift in the UV/ Vis absorbance. However, as already mentioned, the effect is dependent on the solvent used, for example, in toluene solu- tions the photo-degradation is absent. Since the amount of oxygen and water should be approximately the same in chlo- roform as well as toluene, we do not believe that oxygen or water is the main reactive species in our experiments. In fact, the solubility of oxygen in these solvents is very similar at room temperature.44,45 Notice also that in our study, we avoided as much as possible exposing our samples to oxy- gen/air and light, in order to avoid the photo oxidation reac- tions, and we employed different solvent batches. However, water and oxygen is certainly present since our samples were prepared and processed in air. For example, for NMR and FTIR measurements, the solvent was evaporated in air, the powder obtained was re-dissolved in deuterated chloroform in the case of NMR, or the manipulation samples for FTIR, as describe in the experimental part. So, as can be seen in Fig. 3 small bands in the region 1680–1720 cm�1 are observed after irradiation which is assigned to carbonyl groups.9,10,40 Notice that MEH-PPV is extremely susceptible to oxidation when exposed to air.10,40 This hypothesis is re- inforced by the fact the bands related to oxidation are similar independent of the dose used 30 and 90 Gy. Therefore we can conclude that oxidation is not the main process responsi- ble for the blue-shift in the absorption bands of MEH-PPV after irradiation. Considering the possibility of reactive species formation from the solvent, it is known that chloroform is unstable under ionizing radiation,24 forming radicals as described by the following reaction: CHCl3! h� �Clþ �CHCl2: (1) Thus let us consider the hypothesis of the attack of these rad- icals to the polymer chain as the main degradation mecha- nism. Considering the MEH-PPV polymer chain (see Scheme 1), there are four regions where the reaction, substi- tution or addition, can occur: the phenylene ring, the vinyl, the methoxy, and the alkoxy groups. Labile hydrogen abstraction is also possible, as already mentioned and in agreement with the literature.46 073510-5 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions Thus in order to evaluate the mechanisms involved in the spectra alterations and identify the most susceptible regions to radical attack in the polymer backbone, electronic structure calculations were performed. Figure 6 illustrates the Fukui indices associated to radical attack on the MEH- PPV decamer using the ESP charges to evaluate the elec- tronic populations. Similar results are obtained using Mul- liken population analysis. The dark and light regions delimit the minimum and maximum indices, respectively. Note that the central vinyl groups are the most suscepti- ble sites to radical attack. In this regard, different mecha- nisms can be considered: i) double bond saturation by radical incorporation or/and ii) scission of the polymer chain induced by addition of radicals in the vinyl double bond. In both cases, a reduction in the polymer conjugation length and a blue-shift in absorption spectra are expected, in agree- ment with the experimental results. From Eq. (1), the main by-products formed from chloro- form irradiation are �CHCl2 and Cl�. However, due to steric factors it is reasonable to assume that chlorine radicals are more likely to react. In fact, it is known that chlorine radicals present higher reaction rates constants than �CHCl2 in liquid phase.47 In this sense, the radical attack was simulated by chlorine incorporation on different vinyl groups in the poly- mer chain. However similar results should be obtained if one considers �CHCl2. Table II shows the main peak position (kmax) and its shifts (Dk), relative to the unmodified structure, for attacked decamers in different regions of the polymer backbone (unmodified polymer; 1: with chlorine incorporated in the first vinyl group, between the first and second repeating units; 2: radical incorporation in the second vinyl group, between second and third repeating units; etc.; for details see the Supplementary Information).42 As expected, the most no- ticeable changes are observed when the central regions of the polymer chain (region 5) are attacked. Remember that the central regions are the most susceptible sites as obtained from the Fukui indices analysis. It is well know that the polymer absorption spectrum is composed of a non-homogeneous superposition of the absorption of planar substructures (chromophores) present in the main chain.32 Our calculations reproduce this effect (see Supplementary Information42). For example, considering the radical incorporation in region 2 (between the second and third units), the main peak shifts to shorter wavelengths (435 nm) and a new weak peak is formed at 325 nm, which is con- sistent with octamer and dimer substructures resulting after the radical attack. The spectrum is dominated by the larger substructures (with more electrons), in this case the octamer units. Similar considerations can be made for incorporations in the other regions. MEH-PPV absorption spectra were simulated following this idea. For that purpose it was considered a polymer main chain with 330 units (82000 g/mol) constituted by chromo- phores with different conjugation length, randomly distrib- uted between 1 and 20 units. As a first approximation, the p system overlap between adjacent chromophores were neglect, thus, the polymer spectra were calculated from the chromophores spectra sum weighted by the number of units present in each substructure. Averages of 10000 distinct polymeric chains, randomly generated, were considered. The main peak position of each substructure were obtained using a linear regression equation using as input data the result of electronic structure calculations of substructures with 4, 5, 6, 8, and 10 units (see Supplementary Information42). The chro- mophores spectrum was simulated using Lorentzian curves. FIG. 6. Fukui indices (f0) calculated using the ESP charges on the atoms of the MEH-PPV decamer. TABLE II. Main peak position of optimized structures after chlorine incor- poration in different regions of the polymer backbone (for details, see Sup- plementary Information42). Attacked region kmax (nm) Dkmax (nm) 1 438.773 4.660 2 435.504 7.929 3 432.924 10.509 4 426.606 16.827 5 422.902 20.531 Non-attacked 443.433 … FIG. 7. Theoretical optical absorption spectra for structures attacked in the regions 1, 2, 3, 4, and 5 of the polymer backbone. 0 represents the unmodi- fied decamer spectrum. 073510-6 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions The radical attacks were simulated by chlorine incorpo- ration in the central vinyl group of each planar substructure. Attacks on the adjacent vinyl groups, were considered less likely to occur, as indicated by Fukui indices. Figure 7 illus- trates the resulting effect. The number 0 represents structures that have not suffered attack. Data regarding the amount of substructure present in each spectrum are shown in the Sup- plementary Information.42 It is found, as expected, that larger structures are more susceptible to radical attack due to the greater number of regions susceptible to attack, or, in other words, the greater number of vinyl groups. The irradiation promotes an increase in the population of smaller structures due to radical incorpo- ration in the polymer substructures. The polymer spectrum is then dominated by these smaller substructures, resulting in a blue-shift in the spectrum, in good agreement with what is seen in Fig. 1. Support to the hypothesis of chlorine incorporation in the main chain is found in the analysis of FTIR. The band positions of the FTIR, Fig. 3, did not change significantly af- ter irradiation, indicating little or no change in the vibrational environment of these bands. After irradiation new absorption bands at 648 cm�1, 729 cm�1, and 906 cm�1 are observed, assigned to C-Cl, –CCl2, –CHCl or –CH2Cl2 stretching. The SCHEME 1. Proposed MEH-PPV degradation by alkyl halide radicals that induce blue-shifts in the absorption spectra. The radicals were formed by c induced dissociation of CHCl3. The products highlighted are related to H-NMR signals observed in Fig. 5. 073510-7 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions clearest change is the band at 729 cm�1 related to –CCl or –CCl2 stretching.48 Another indication of the addition in vinyl double bond after irradiation, is the decrease of the band in 3065 cm�1, which is assigned to C-H stretching from the vinyl group. Notice that the process of chlorine addition is clearly dependent on the irradiation dose. Similar effects were observed in 1H NMR. As men- tioned before, commercial polymers can have structural defects due to impurities or residues of the synthesis. The signals in 2.7–2.8, 6.6, and 6.7–6.8 ppm are attributed to tolane bis-benzyl defects and cis-configurated vinylene moieties.41 Other signals especially in the range of 2.5–3.5 and 5.5–6.5 ppm are assigned to defects. These fragments, initially present as defects, changed after irradiation, which is a clear indication that radical have been created after irra- diation. These structures are more unstable, and thus suscep- tible to radical addition. However, there is no evidence that these defects participate in the chromatic alteration mecha- nism, since these fragments have shorter conjugation lengths, it does not affect the absorption spectrum of the polymer sig- nificantly. In addition the same effect has been observed in different polymer batches, with different defects amount. The most important observation, in what concerns chlo- rine addition, are the signals at 4.7 and 7.2 ppm associated with chloromethylene protons of the –CH2Cl (one singlet as a result of unsymmetrical substitution) and nearby aromatic protons, respectively. The signal in 5.46 ppm is also attrib- uted to the addition of two chlorine atoms in the polymer chain, following what is reported in the literature and 1H NMR simulations using Chem Draw Ultra.49 The presence of chlorine addition/substitution could be confirmed by spin- spin coupling (JHH) analysis. However, given the resolution obtained in the 1H NMR spectrum was not possible to obtain the JHH value. In summary, our results indicate that the origin of the effects observed is the result of chlorine radical attack to the vinyl groups of MEH-PPV. Our theoretical simulations indi- cate that the phenylene ring is less reactive than the vinyl group. Thus we shall not consider the participation of the ring in the effect. The same theoretical simulation shows that substitutions on the methoxy or alkoxy group should not influence the optical properties of the polymer (not shown). From what has been discussed so far, the following reaction mechanism is proposed. Notice however that at this point one cannot give a detailed description of the reaction. The first step is the c induced dissociation of the solvent, Eq. (1). Following this dissociation, the two radicals formed attack the polymer chain. From reactivity and steric consider- ations it is assumed that the addition and substitution reactions are mainly due to �Cl and not �CHCl2 (see Ref. 50) after H abstractions from the polymer chain. An indication in favor of this hypothesis is the formation of CH2Cl2, as seen from H- NMR data published elsewhere.8 As already mentioned, we are only considering here reactions involving the backbone chain. �Cl is added to the vinyl group of MEH-PPV creating a new radical that sits close to Cl, Scheme 1. This secondary radical is highly reactive, so in the following the species avail- able for reaction, �Cl, �CHCl2 are added or substituted in a se- ries of concatenated reactions, some of which are represented in the same scheme. This series of reactions can lead to chain scission, products 6 and 9 in the scheme. Our results are in good agreement with previous reports that shows that the introduction of electron-withdrawing sub- stituent, and in particular halogen atoms, in the conjugated backbone, has profound effect in the bandgap, emission proper- ties, and electron injection properties of PPV polymers.51 For example, PPVs with fluorine atoms in the back bone52 show a blue-shift in the absorption peak, with respect to pure PPV. V. CONCLUSIONS We have investigated the origin of blue-shifts in the MEH-PPV absorption band after c irradiation. Our results indicate that the main mechanism for the blue-shifts is indi- rect. Chloroform absorbs c rays that produce its dissociation into chlorine radicals. These radicals attack the polymer chain preferentially in central vinyl bonds; according to reac- tivity calculations performed using Fukui indices. The radi- cal �Cl is added to C double bonds or substitute H leading to a decrease in conjugation length and eventually to chain scis- sion. Our results indicate that oxygen does not play a major role in the effect. Theoretical electronic spectra simulations were performed based on these assumptions reproducing the UV-Vis experimental results. Our findings on the polymer-solvent radiation interac- tion shall help the development of more sensitive polymer- based dosimeters. ACKNOWLEDGMENTS We would like to thank the Brazilian agencies CAPES, FAPESP (INCTMN) and CNPq for their financial support, Thomas Geiger (EMPA, Switzerland) for the GPC measure- ments, and Marco A. R. Fernandes (UNESP-Botucatu) for the 60Co irradiations. The simulations were done in GRIDU- nesp, with financial support from UNESP. 1G. F. Knoll, Radiation Detection and Measurement, 3rd ed. (Wiley, New York, 2000). 2M. G. Stabin and A. B. Brill, Semin. Nucl. Med. 38, 306 (2008). 3R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes, R. N. Marks, C. Taliani, D. D. C. Bradley, D. A. Dos Santos, J. L. Brédas, M. Lögdlund, and W. Salaneck, Nature 397, 121(1999). 4E. A. B. Silva, J. F. Borin, P. Nicolucci, C. F. O. Graeff, T. G. Netto, and R. F. Bianchi, Appl. Phys. Lett. 86, 131902 (2005). 5J. M. Lobez and T. M. Swager, Angew. Chem. Int. Ed. 48, 95 (2009). 6H. Zhong, Y. Zhao, Y. Li, and Q. Pei, Nanotechnology 19, 505503 (2008). 7Y. S. Zhao, H. Zhong, and Q. Pei, Phys. Chem. Chem. Phys. 10, 1848 (2008). 8E. S. Bronze-Uhle, J. F. Borin, A. Batagin-Neto, and C. F. O. Graeff, “MEH-PPV hypsochromic shifts in halogenated solvents induced by c rays” (unpublished). 9J. C. Scott, J. H. Kaufman, P. J. Brock, R. Di Pietro, J. Salem, and J. A. Goitia, J. Appl. Phys. 79, 2745 (1996). 10T. Tromholt, M. Maceau, Martin Helgesen, J. E. Carlé, and F. C. Krebs, Sol. Energy Mater. Sol. Cells. 95, 1308 (2011). 11S. Chambon, A. Rivaton, J. Gardette, and M. Firon, J. Polym. Sci., Part A: Polym. Chem. 47, 6044 (2009). 12S. Chambon, A. Rivaton, J. Gardette, and M. Firon, Sol. Energy Mater. Sol. Cells. 91, 394 (2007). 13S. Chambon, A. Rivaton, J. Gardette, and M. Firon, Sol. Energy Mater. Sol. Cells. 92, 785 (2008). 14R. D. Scurlock, B. Wang, P. R. Ogilby, J. R. Sheats, and R. L. Clough, J. Am. Chem. Soc. 117, 10194 (1995). 073510-8 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions http://dx.doi.org/10.1053/j.semnuclmed.2008.05.009 http://dx.doi.org/10.1038/16393 http://dx.doi.org/10.1063/1.1891300 http://dx.doi.org/10.1002/anie.200904936 http://dx.doi.org/10.1088/0957-4484/19/50/505503 http://dx.doi.org/10.1039/b801375f http://dx.doi.org/10.1063/1.361096 http://dx.doi.org/10.1016/j.solmat.2010.09.022 http://dx.doi.org/10.1002/pola.23628 http://dx.doi.org/10.1002/pola.23628 http://dx.doi.org/10.1016/j.solmat.2006.10.015 http://dx.doi.org/10.1016/j.solmat.2006.10.015 http://dx.doi.org/10.1016/j.solmat.2007.12.003 http://dx.doi.org/10.1016/j.solmat.2007.12.003 http://dx.doi.org/10.1021/ja00146a004 http://dx.doi.org/10.1021/ja00146a004 15M. Scoponi, F. Pradella, H. Kaczmarek, R. Amadelli, and V. Carassiti, Polymer 37, 903 (1996). 16B. H. Cumpston and K. F. Jensen, J. Appl. Polym. Sci. 69, 2451 (1998). 17I. D. Parker, Y. Cao, and C. Y. Yang, J. Appl. Phys. 85, 2441 (1999). 18A. Chapiro, Radiation Chemistry of Polymeric Systems (Wiley, New York, 1962). 19A. Chapiro, Radiat. Phys. Chem. 63, 207 (2002). 20K. Slimani, L. Moine, C. Aymes-Chodur, A. Laurent, D. Labarre, and N. Yagoubi, Polym. Degrad. Stab. 94, 584 (2009). 21A. Charlesby, Atomic Radiation and Polymers (Elsevier, New York, 1960). 22C. L. Moad and D. J. Winzor, Prog. Polym. Sci. 23, 759 (1998). 23T. Seguchi, S. Hashimoto, K. Arakawa, N. Hayakawa, W. Kawakami, and I. Kuriyama, Radiat. Phys. Chem. 17, 195 (1981). 24H. R. Werner and R. F. Firestone, J. Phys. Chem. 69, 840 (1965). 25L. Ma, X. Wang, B. Wang, J. Chen, J. Wang, K. Huang, B. Zhang, Y. Cao, Z. Han, S. Qian, and S. Yao, Chem. Phys. 285, 85 (2002). 26James J. P. Stewart, MOPAC 2007: Molecular Orbital Package, Stewart Computational Chemistry, 2007. 27A. Klamt, and G. Schüürmann, J. Chem. Soc., Perkin Trans. 2, 5, 799 (1993). 28T. Mineva, J. Mol. Struct.: THEOCHEM 762, 79 (2006). 29J. Cruz, L. M. R. Martı́nez-Aguilera, R. Salcedo, and M. Castro, Int. J. Quantum Chem. 85, 546 (2001). 30E. G. Lewars, Computational Chemistry: Introduction to the Theory and Applications of Molecular and Quantum Mechanics, 2nd ed. (Springer, Berlin, 2010). 31W. Yang and W. J. Mortier, J. Am. Chem. Soc. 108, 5708 (1986). 32B. J. Schwartz, Annu. Rev. Phys. Chem. 54, 141 (2003). 33R. Traiphol, P. Sanguansat, T. Srikhirin, T. Kerdcharoen, and T. Osotchan, Macromolecules 39, 1165 (2006). 34A. Szabo and N. S. Ostlund, Modern Quantum Chemistry: Introduction to Advanced Electronic Structure Theory (Dover, New York, 1996). 35I. N. Levine, Quantum Chemistry, 5th ed. (Prentice-Hall, New York, 1999). 36A. A. Granovsky, see http://classic.chem.msu.su/gran/firefly/index.html for more information about Firefly 7.1. G, 2008. 37F. Neese, see http://www.thch.uni-bonn.de/tc/orca/ for more information about ORCA: An ab initio, DFT and semiempirical SCF-MO, 2007. 38Y. Gartstein, M. Rice, and E. Conwell, Phys. Rev. B 51, 5546 (1995). 39B. D. Yang, K. H. Yoon, and K. W. Chung, Mater. Chem. Phys. 83, 334 (2004). 40H. Becker, H. Spreitzer, K. Ibrom, and W. Kreuder, Macromolecules 32, 4925 (1999). 41T. Schwalm, J. Wiesecke, S. Immel, and M. Rehahn, Macromol. Rapid Commun. 30, 1295 (2009). 42See Supplementary Material at http://dx.doi.org/10.1063/1.3644946 for details about the HSQC spectrum and computational simulations. 43M. Marinovic-Cincovic, Z. Kacarevic-Popovic, and D. Babic, Radiat. Phys. Chem. 67, 425 (2003). 44S. L. Murov, I. Carmichael, and G. L. Hug, Handbook of Photochemistry, 2nd ed. (CRC Press, Boca Raton, 1993). 45B. M. Monroe, Photochem. Photobiol. 35, 863 (1982). 46J. Davenas, I. Stevenson, N. Celette, S. Cambon, J. L. Gardette, A. Riva- ton, and L. Vignoud, Nucl. Instrum. Methods Phys. Res. B 191, 653 (2002). 47N. E. Bibler, J. Phys. Chem. 75, 2436 (1971). 48G. Socrates, Infrared Characteristic Group Frequencies, 2nd ed. (Wiley, New York, 1994). 49Chem Draw Ultra: Chemical Structure Drawling Standard, Cambridge Soft, 2006. 50F. M. Salih and A. E. Pillay, J. Radioanal. Nucl. Chem. 261, 665 (2004). 51F. Babudri, G. M. Farinola, F. Naso, and R. Ragni, Chem. Commun. 2007, 1003. 52M. Losurdo, M. M. Giangregorio, P. Capezzuto, G. Bruno, F. Babudri, A. Cardone, C. Martinelli, G. M. Farinola, F. Naso, and M. Büchel, Polymer. 49, 4133 (2008). 073510-9 Bronze-Uhle et al. J. Appl. Phys. 110, 073510 (2011) Downloaded 11 Jul 2013 to 186.217.234.138. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://jap.aip.org/about/rights_and_permissions http://dx.doi.org/10.1016/0032-3861(96)87272-7 http://dx.doi.org/10.1002/(SICI)1097-4628(19980919)69:12<>1.0.CO;2-N http://dx.doi.org/10.1063/1.369564 http://dx.doi.org/10.1016/S0969-806X(01)00621-1 http://dx.doi.org/10.1016/j.polymdegradstab.2009.01.011 http://dx.doi.org/10.1016/S0079-6700(97)00041-5 http://dx.doi.org/10.1016/0146-5724(81)90331-9 http://dx.doi.org/10.1021/j100887a023 http://dx.doi.org/10.1016/S0301-0104(02)00691-2 http://dx.doi.org/10.1039/P29930000799 http://dx.doi.org/10.1016/j.theochem.2005.08.044 http://dx.doi.org/10.1002/qua.v85:4/5 http://dx.doi.org/10.1002/qua.v85:4/5 http://dx.doi.org/10.1021/ja00279a008 http://dx.doi.org/10.1146/annurev.physchem.54.011002.103811 http://dx.doi.org/10.1021/ma052512+ http://classic.chem.msu.su/gran/firefly/index.html http://www.thch.uni-bonn.de/tc/orca/ http://dx.doi.org/10.1103/PhysRevB.51.5546 http://dx.doi.org/10.1016/j.matchemphys.2003.10.003 http://dx.doi.org/10.1021/ma990347q http://dx.doi.org/10.1002/marc.200900104 http://dx.doi.org/10.1002/marc.200900104 http://dx.doi.org/10.1063/1.3644946 http://dx.doi.org/10.1016/S0969-806X(03)00079-3 http://dx.doi.org/10.1016/S0969-806X(03)00079-3 http://dx.doi.org/10.1111/j.1751-1097.1982.tb02660.x http://dx.doi.org/10.1016/S0168-583X(02)00628-6 http://dx.doi.org/10.1021/j100685a004 http://dx.doi.org/10.1023/B:JRNC.0000037111.93234.c8 http://dx.doi.org/10.1039/b611336b http://dx.doi.org/10.1016/j.polymer.2008.07.024