- Email: [email protected]
Suppressing the energy transfer in polymer blends films upon addition of a co-solvent
Materials Letters 175 (2016) 248–251
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Suppressing the energy transfer in polymer blends films upon addition of a co-solvent J. Pereira a, J. Farinhas a, J. Morgado a,b,n a b
Instituto de Telecomunicações, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal
art ic l e i nf o
a b s t r a c t
Article history: Received 12 December 2015 Received in revised form 1 April 2016 Accepted 3 April 2016 Available online 6 April 2016
Energy transfer in conjugated polymer blends is an important process in optoelectronics, with its efficiency very sensitive to the details of phase separation in films. Modification of the relative composition of the blend has been the usual strategy to tune the energy transfer efficiency. In this letter we report on a new approach to switch-off that process: the use of a co-solvent. In particular, we find that the addition of 1,8-diiodooctane to xylene solutions of poly(9,9-dioctylfluorene): poly(9,9-dioctylfluorene-alt-benzothiadiazole) (PFO:F8BT), blends, 99:1 by weight, suppresses the energy transfer in films. This leads to a change from green light-emitting diodes into whitish emitting diodes. This energy transfer suppression is attributed to a stronger phase segregation, with the concomitant formation of larger domains of the two polymers, thereby increasing the distance between energy donor (PFO) and acceptor (F8BT) sites. & 2016 Elsevier B.V. All rights reserved.
Keywords: Conjugated polymer blends Phase separation Additives Luminescence properties Optoelectronics
1. Introduction Polymer blends have been used to extend the absorption range of organic photovoltaic cells [1] and to tune the emission color of polymer light-emitting diodes (PLEDs) [2]. The use of conjugated polymer blends in PLEDs has also been used to improve the emission efficiency by combining polymers with complementary charge transport properties. Blends of poly(9,9-dioctylfluorene) (PFO) and poly(9,9-diotylfluorene-alt-benzothiadiazole) (F8BT) are among the best studied polymer blends in this context, [3] as these two polymers combine complementary hole and electron mobilities with an efficient energy transfer, due to a large overlap of PFO emission and F8BT absorption spectra, leading to green emitting devices [4]. High electroluminescence, EL, efficiencies were reported for blends consisting of 95:5, PFO to F8BT, weight ratio [5]. There was a report showing that the exposure of such blend to a non-solvent (acetone) was able to reduce the efficiency of the energy transfer and induce the appearance of a residual PFO emission. This was a surface directed process, amplified by the device structure, as recombination was essentially occurring close to the surface of the active layer film, which had been in contact with acetone [6]. In the context of the organic photovoltaic cells based on blends n Corresponding author at: Instituto de Telecomunicações, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal. E-mail address: [email protected] (J. Morgado).
http://dx.doi.org/10.1016/j.matlet.2016.04.047 0167-577X/& 2016 Elsevier B.V. All rights reserved.
of conjugated polymers and fullerenes, the importance of the domains size formed upon spontaneous phase separation was early recognized. In such bulk heterojunction solar cells, too large domains of either component lead to exciton loss associated to inefficient charge generation via exciton dissociation. Therefore the preparation details of such films and the use of post-film formation processes (temperature and solvent annealing) were shown to have a profound impact on the device efficiency [7]. In particular, the use of additives has been explored, [8] such as alcohols and, probably the most commonly used additive currently, 1,8-diiodooctane (DIO) and is still routinely used [9]. DIO, though not always leading to efficiency increase, appears to reduce the size of the domains, leading to finer morphologies, with a material-dependent change of the polymer phase crystallinity [10]. These effects are related to DIO's high boiling point and dissimilar interactions with the blends components. In this letter we report on recent findings about the role of DIO addition to xylene solutions of PFO:F8BT blends. At variance with the general DIO-promoted finer phase separation in blends of conjugated polymers and fullerenes, DIO promotes the suppression of the very efficient energy transfer form PFO to F8BT, that we find in films prepared by spin coating from regular xylene solutions, and which we attribute to the formation of larger domains of the polymers. We find also that this coarsen phase separation is not specific of this particular polymer pair. We believe this effect has broad implications for polymer-based optoelectronic devices in general and in particular for PLEDs.
J. Pereira et al. / Materials Letters 175 (2016) 248–251
249
Fig. 2. A. Current (I) and luminance (L) of the devices prepared with the PFO:F8BT blends as a function of the applied voltage, and B. corresponding luminous and external quantum efficiency (EQE).
Fig. 1. Normalized absorbance (A) and photoluminescence (PL) and electroluminescence (EL) spectra (B) of the PFO:F8BT, 99:1, blend without and with 2% and 20% DIO added to the solution. PL spectra of neat films of F8BT and PFO (prepared from a xylene solution with 20% DIO content) are also shown. PL spectra of the blends were obtained upon excitation at 385 nm, while the PL of PFO film prepared from a solution with DIO was obtained upon excitation at 407 nm (absorption maximum). The inset shows the CIE coordinates in the chromaticity diagram of the EL spectra.
2. Materials and methods PFO and F8BT were obtained from ADS and used as received. Solutions were prepared inside the globe box. A starting solution of PFO:F8BT, 99:1, by weight, were prepared in xylene at 0.8%, by weight. Two additional solutions were prepared by adding DIO in 2% and 20% by weight. PLEDs with ITO/PEDOT:PSS/blend/LiF/Al structure were prepared and tested as described in Ref. [11]. PEDOT:PSS (40 nm thick) was deposited by spin coating on glass/ITO (Visiontek) substrates from an aqueous dispersion (Clevios P VP.AI 4083, from Heraeus) and annealed in air at 120 °C for 10 min. These glass/ITO/
PEDOT:PSS substrates were then transferred inside a glove box to finalize device fabrication. Films of PFO:F8BT were deposited by spin coating at 1800 rpm to give films with thicknesses in the range 80–90 nm. PLEDs were finalized by depositing 1.5 nm of LiF followed by a top layer of ca. 80 nm aluminum, defining pixel areas of 4 mm2. Electroluminescence (EL) spectra were recorded with a ScanSci CCD spectrograph. Absorption and photoluminescence (PL) spectra of films deposited on spectrosil substrates were obtained using a Cecil 7200 UV/VIS spectrophotometer and a SPEX fluorolog 212I fluorimeter, respectively. AFM images were obtained with a Nano Observer from Concept Scientific Instruments (Les Ulis, France) using non-contact mode with cantilivers having resonant frequencies between 200 kHz and 400 kHz and silicon tips of under 10 nm radii. Gwydion (version 2.26) software was used for data processing.
3. Results and discussion Fig. 1 shows the absorption, PL and EL spectra for the various blends. F8BT contribution to the blends absorption spectrum is negligible, which is consistent with the very low F8BT content. The PL spectrum of the 99:1 blend has contributions from both materials. Despite the very low content of F8BT, which is not noticeable in the absorption this dominates the emission spectrum. Additionally, we find that the emission maximum of FBT in the
250
J. Pereira et al. / Materials Letters 175 (2016) 248–251
Fig. 3. AFM images of films of PFO:F8BT, 99:1, by weight, prepared by spin coating from xylene (A-topography, B-phase image) and from xylene with 20% DIO (C-topography, D-phase) solutions.
blend occurs at 530 nm, which is blue-shifted with respect to the emission of neat F8BT (549 nm), a shift that is related to a decrease of the F8BT inter-fluorophore interactions when in the blend. The emission from PFO peaks at about 440 nm, with a small shoulder at 420 nm. The sharp spectrum, with a maximum at 440 nm, is attributed to the so-called PFO β-phase [12], while the shoulder at 420 nm is related to the emission of the amorphous PFO phase. The film prepared from the solution with 2% DIO, shows a strong suppression of F8BT emission. In addition, F8BT emission spectrum is red-shifted with respect to that of the film prepared from only xylene solution, suggesting an increase of F8BT inter-fluorophore interactions. The quenching of F8BT emission is even more pronounced in the film prepared with 20% DIO, with the spectrum dominated by the PFO β-phase. Comparing with the PL spectrum of a film of neat PFO prepared from a solution with 20% DIO, which is entirely due to the PFO β-phase, in agreement with a previous report by Peet et al., [13] we conclude that the contribution from F8BT in the blend prepared from the solution with 20% DIO, is very small. The EL spectra of the LEDs prepared with the PFO:F8BT blend show a similar effect of the DIO addition to the solutions used to prepare the films, though the EL spectra show a higher contribution of the F8BT emission when comparing with PL. This we attribute to an additional contribution resulting from electronhole recombination occurring preferentially at F8BT sites. The EL spectrum of the PLEDs based on the PFO:F8BT blend has CIE color coordinates, x ¼0.31 and y¼0.59, that are typical of a green emission color. Upon addition of 2% DIO they change to x ¼0.30 and y ¼0.50, and PLEDs based on the blend film prepared from the
solution with 20% DIO have an emission spectrum with x¼ 0.24 and y¼ 0.36, that correspond to a whitish color. The CIE coordinates for the corresponding PL spectra are (0.25, 0.38), (0.34, 044) and (0.21, 0.23), which reflect the higher contribution of PFO blue emission (0.17, 0.11), when comparing with the EL spectra, discussed above. Fig. 2 compares the performance of the LEDs. Luminance reaches values around 7600 cd/m2 for devices based on the blend prepared from xylene solutions, decreasing for the films prepared from the DIO-containing solutions, down to 2000 cd/m2 for the films prepared from the solution with 20% DIO. The effect on the current follows the same trend, though without significant differences between the films prepared from solutions with different DIO content. As shown in Fig. 2B, the EL efficiency decreases when DIO is added to the blend solution, attributed to a decrease of the F8BT role in improving the electron:hole current balance and increased quenching of excitons in richer F8BT domains. Isolated F8BT chains, when in the regular blend, are expected to confine the excitons transferred from PFO and this confinement increases the PL efficiency. The addition of DIO, by promoting the formation of larger domains of F8BT, as discussed below, attenuates that confinement effect [14]. Overall, we find that the addition of DIO to the blend xylene solution suppresses the energy transfer from PFO to F8BT, under both photo and electrical excitation. The energy transfer processes in PFO:F8BT blends were studied by Buckley et al. [4] The excited state energy transfer from PFO to F8BT was rationalized in terms of a Förster resonance energy transfer, which relies on the spectral
J. Pereira et al. / Materials Letters 175 (2016) 248–251
overlap between PFO (donor) emission and F8BT (acceptor) absorption spectra and on the distance between donor and acceptor. It was concluded that at high F8BT concentration there is a nearest neighbour energy transfer, while at low concentrations this process is accompanied by exciton migration within PFO domains prior to its transfer to F8BT. Within the Förster model, a key parameter is the Förster radius, R0, which corresponds to the donor–acceptor (D–A) distance at which the probability of energy transfer equals the probability of exciton decay in the donor. This can be taken as the critical D–A distance for efficient energy transfer to occur. This theory has been developed assuming that D and A correspond to point dipoles. However, in view of the delocalized nature of the conjugated polymers excited states, the theory cannot be strictly used in such cases. An additional difficulty arises when defining the molar extinction coefficient of the acceptor. In fact, Buckley et al. [4] found that the Förster radius for PFO:F8BT would be 9 nm if the entire F8BT macromolecule is used or 4.3 nm if a 5 monomer units is considered. However, from the fluorescence time dependent studies in PFO:F8BT they obtained a Förster radius of 0.8 nm. This large difference in values points to the inadequacy of the simple model to describe the energy transfer in conjugated polymer systems. However, by considering the fundamentals of the model, we may conclude that the suppression of the energy transfer from PFO to F8BT upon addition of DIO is related to the phase separation in the blend film. In fact, as the materials are the same, only the distance between the two can explain the reduction of the energy transfer efficiency. As DIO has a much higher boiling point than xylene (365–140 °C), it is likely that, after xylene evaporation, DIO remains for a longer time in the film and promotes a higher degree of phase segregation, leading to larger domains of either polymer. The high quenching efficiency points even to the existence of domains of essentially pure polymers. This is quite unexpected as complete phase segregation in polymers is not usually found, as shown, for instance, in blends of conjugated polymers and perfluorinated saturated polymers [15]. We performed AFM studies of the films surfaces and found, as shown in Fig. 3, that the presence of DIO in the starting solutions leads to rougher films, consistent with the formation of larger phase separated domains.
4. Conclusions In conclusion, we find that the addition of DIO to the xylene solution of the PFO:F8BT blend leads to films with largely quenched energy transfer from PFO to F8BT which indicates that DIO leads to the formation of larger domains of the two polymers resulting from enhanced phase segregation. This approach is also
251
demonstrated to change the emission color of the PLEDs based on the blend, from green to a whitish emission. We believe this finding is relevant for the broad area of the polymer-based optoelectronics devices and can be further explored to fine tune the emission color of PLEDs.
Acknowledgements We thank FCT-Portugal for financial support under the contract UID/EEA/50008/2013.
References [1] (a) T. Ameri, P. Khoram, J. Min, C.J. Brabec, Adv. Mater. 25 (2013) 4245; (b) V. Gupta, V. Bhati, M. Kumar, S. Chand, A.J. Heeger, Adv. Mater. 27 (2015) 4398. [2] (a) C.R. McNeill, N.C. Greenham, Adv. Mater. 21 (2009) 3840; (b) E. Moons, J. Phys: Condens. Matter 14 (2002) 12235; (c) A. Charas, J. Morgado, J.M.G. Martinho, A. Fedorov, L. Alcácer, F. Cacialli, J. Mater. Chem. 12 (2002) 3523; (d) J.F. De Deus, G.C. Faria, E.T. Iamazaki, R.M. Faria, T.D.Z. Atvars, L. Akcelrud, Org. Electron. 12 (2011) 1493. [3] (a) C.I. Wilkinson, D.G. Lidzey, L.C. Palilis, R.B. Fletcher, S.J. Martin, X.H. Wang, D.D.C. Bradley, Appl. Phys. Lett. 79 (2001) 171; (b) J. Morgado, R.H. Friend, F. Cacialli, Appl. Phys. Lett. 80 (2002) 2436. [4] A.R. Buckley, M.D. Rahn, J. Hill, J. Cabanillas-Gonzalez, A.M. Fox, D.D.C. Bradley, Chem. Phys. Lett. 339 (2001) 331. [5] A.J. Campbell, D.D.C. Bradley, H. Antoniadis, Appl. Phys. Lett. 79 (2001) 2133. [6] J. Morgado, E. Moons, R.H. Friend, F. Cacialli, Adv. Mater. 13 (2001) 810. [7] (a) G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat. Mater. 4 (2005) 864; (b) H. Chen, J. Miao, J. Yan, Z. He, H. Wu, IEEE J. Sel. Top. Quant. Electron. 22 (2016) 4100107. [8] (a) H. Zhou, Y. Zhang, J. Seifter, S.D. Collins, C. Luo, G.C. Bazan, T.-Q. Nguyen, A. J. Heeger, Adv. Mater. 25 (2013) 1513; (b) J. Peet, J.Y. Kim, N.E. Coates, W.L. Ma, D. Moses, A.J. Heeger, G.C. Bazan, Nat. Mater. 6 (2007) 497. [9] A.K.K. Kyaw, D.H. Wang, D. Wynands, J. Zhang, T.-Q. Nguyen, G.C. Bazan, A. J. Heeger, Nano Lett. 13 (2013) 3796. [10] (a) A.K.K. Kyaw, D.H. Wang, C. Luo, Y. Cao, T.-Q. Nguyen, G.C. Bazan, A. J. Heeger, Adv. Energy Mater. 4 (2014) 1301469; (b) H.-C. Liao, C.-C. Ho, C.-Y. Chang, M.-H. Jao, S.B. Darling, W.-F. Su, Mater. Today 16 (2013) 326. [11] J. Morgado, A. Charas, J.A. Fernandes, I.S. Gonçalves, L.D. Carlos, L. Alcácer, J. Phys. D: Appl. Phys. 39 (2006) 3582. [12] M. Grell, D.D.C. Bradley, G. Ungar, J. Hill, K.S. Whitehead, Macromolecules 32 (1999) 5810. [13] J. Peet, E. Brocker, Y. Xu, G. Bazan, Adv. Mater. 20 (2008) 1882. [14] A. Charas, J. Morgado, J.M.G. Martinho, A. Fedorov, L. Alcácer, F. Cacialli, J. Mater. Chem. 12 (2002) 3523. [15] (a) C.R. McNeill, K. Asadi, B. Watts, P.W.M. Blom, D.M. De Leeuw, Small 6 (2010) 508; (b) T. Braz, A.L. Mendonça, R.E. di Paolo, J. Morgado, Submitted for publication.
Contents lists available at ScienceDirect
Materials Letters journal homepage: www.elsevier.com/locate/matlet
Suppressing the energy transfer in polymer blends films upon addition of a co-solvent J. Pereira a, J. Farinhas a, J. Morgado a,b,n a b
Instituto de Telecomunicações, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal
art ic l e i nf o
a b s t r a c t
Article history: Received 12 December 2015 Received in revised form 1 April 2016 Accepted 3 April 2016 Available online 6 April 2016
Energy transfer in conjugated polymer blends is an important process in optoelectronics, with its efficiency very sensitive to the details of phase separation in films. Modification of the relative composition of the blend has been the usual strategy to tune the energy transfer efficiency. In this letter we report on a new approach to switch-off that process: the use of a co-solvent. In particular, we find that the addition of 1,8-diiodooctane to xylene solutions of poly(9,9-dioctylfluorene): poly(9,9-dioctylfluorene-alt-benzothiadiazole) (PFO:F8BT), blends, 99:1 by weight, suppresses the energy transfer in films. This leads to a change from green light-emitting diodes into whitish emitting diodes. This energy transfer suppression is attributed to a stronger phase segregation, with the concomitant formation of larger domains of the two polymers, thereby increasing the distance between energy donor (PFO) and acceptor (F8BT) sites. & 2016 Elsevier B.V. All rights reserved.
Keywords: Conjugated polymer blends Phase separation Additives Luminescence properties Optoelectronics
1. Introduction Polymer blends have been used to extend the absorption range of organic photovoltaic cells [1] and to tune the emission color of polymer light-emitting diodes (PLEDs) [2]. The use of conjugated polymer blends in PLEDs has also been used to improve the emission efficiency by combining polymers with complementary charge transport properties. Blends of poly(9,9-dioctylfluorene) (PFO) and poly(9,9-diotylfluorene-alt-benzothiadiazole) (F8BT) are among the best studied polymer blends in this context, [3] as these two polymers combine complementary hole and electron mobilities with an efficient energy transfer, due to a large overlap of PFO emission and F8BT absorption spectra, leading to green emitting devices [4]. High electroluminescence, EL, efficiencies were reported for blends consisting of 95:5, PFO to F8BT, weight ratio [5]. There was a report showing that the exposure of such blend to a non-solvent (acetone) was able to reduce the efficiency of the energy transfer and induce the appearance of a residual PFO emission. This was a surface directed process, amplified by the device structure, as recombination was essentially occurring close to the surface of the active layer film, which had been in contact with acetone [6]. In the context of the organic photovoltaic cells based on blends n Corresponding author at: Instituto de Telecomunicações, Av. Rovisco Pais, P-1049-001 Lisboa, Portugal. E-mail address: [email protected] (J. Morgado).
http://dx.doi.org/10.1016/j.matlet.2016.04.047 0167-577X/& 2016 Elsevier B.V. All rights reserved.
of conjugated polymers and fullerenes, the importance of the domains size formed upon spontaneous phase separation was early recognized. In such bulk heterojunction solar cells, too large domains of either component lead to exciton loss associated to inefficient charge generation via exciton dissociation. Therefore the preparation details of such films and the use of post-film formation processes (temperature and solvent annealing) were shown to have a profound impact on the device efficiency [7]. In particular, the use of additives has been explored, [8] such as alcohols and, probably the most commonly used additive currently, 1,8-diiodooctane (DIO) and is still routinely used [9]. DIO, though not always leading to efficiency increase, appears to reduce the size of the domains, leading to finer morphologies, with a material-dependent change of the polymer phase crystallinity [10]. These effects are related to DIO's high boiling point and dissimilar interactions with the blends components. In this letter we report on recent findings about the role of DIO addition to xylene solutions of PFO:F8BT blends. At variance with the general DIO-promoted finer phase separation in blends of conjugated polymers and fullerenes, DIO promotes the suppression of the very efficient energy transfer form PFO to F8BT, that we find in films prepared by spin coating from regular xylene solutions, and which we attribute to the formation of larger domains of the polymers. We find also that this coarsen phase separation is not specific of this particular polymer pair. We believe this effect has broad implications for polymer-based optoelectronic devices in general and in particular for PLEDs.
J. Pereira et al. / Materials Letters 175 (2016) 248–251
249
Fig. 2. A. Current (I) and luminance (L) of the devices prepared with the PFO:F8BT blends as a function of the applied voltage, and B. corresponding luminous and external quantum efficiency (EQE).
Fig. 1. Normalized absorbance (A) and photoluminescence (PL) and electroluminescence (EL) spectra (B) of the PFO:F8BT, 99:1, blend without and with 2% and 20% DIO added to the solution. PL spectra of neat films of F8BT and PFO (prepared from a xylene solution with 20% DIO content) are also shown. PL spectra of the blends were obtained upon excitation at 385 nm, while the PL of PFO film prepared from a solution with DIO was obtained upon excitation at 407 nm (absorption maximum). The inset shows the CIE coordinates in the chromaticity diagram of the EL spectra.
2. Materials and methods PFO and F8BT were obtained from ADS and used as received. Solutions were prepared inside the globe box. A starting solution of PFO:F8BT, 99:1, by weight, were prepared in xylene at 0.8%, by weight. Two additional solutions were prepared by adding DIO in 2% and 20% by weight. PLEDs with ITO/PEDOT:PSS/blend/LiF/Al structure were prepared and tested as described in Ref. [11]. PEDOT:PSS (40 nm thick) was deposited by spin coating on glass/ITO (Visiontek) substrates from an aqueous dispersion (Clevios P VP.AI 4083, from Heraeus) and annealed in air at 120 °C for 10 min. These glass/ITO/
PEDOT:PSS substrates were then transferred inside a glove box to finalize device fabrication. Films of PFO:F8BT were deposited by spin coating at 1800 rpm to give films with thicknesses in the range 80–90 nm. PLEDs were finalized by depositing 1.5 nm of LiF followed by a top layer of ca. 80 nm aluminum, defining pixel areas of 4 mm2. Electroluminescence (EL) spectra were recorded with a ScanSci CCD spectrograph. Absorption and photoluminescence (PL) spectra of films deposited on spectrosil substrates were obtained using a Cecil 7200 UV/VIS spectrophotometer and a SPEX fluorolog 212I fluorimeter, respectively. AFM images were obtained with a Nano Observer from Concept Scientific Instruments (Les Ulis, France) using non-contact mode with cantilivers having resonant frequencies between 200 kHz and 400 kHz and silicon tips of under 10 nm radii. Gwydion (version 2.26) software was used for data processing.
3. Results and discussion Fig. 1 shows the absorption, PL and EL spectra for the various blends. F8BT contribution to the blends absorption spectrum is negligible, which is consistent with the very low F8BT content. The PL spectrum of the 99:1 blend has contributions from both materials. Despite the very low content of F8BT, which is not noticeable in the absorption this dominates the emission spectrum. Additionally, we find that the emission maximum of FBT in the
250
J. Pereira et al. / Materials Letters 175 (2016) 248–251
Fig. 3. AFM images of films of PFO:F8BT, 99:1, by weight, prepared by spin coating from xylene (A-topography, B-phase image) and from xylene with 20% DIO (C-topography, D-phase) solutions.
blend occurs at 530 nm, which is blue-shifted with respect to the emission of neat F8BT (549 nm), a shift that is related to a decrease of the F8BT inter-fluorophore interactions when in the blend. The emission from PFO peaks at about 440 nm, with a small shoulder at 420 nm. The sharp spectrum, with a maximum at 440 nm, is attributed to the so-called PFO β-phase [12], while the shoulder at 420 nm is related to the emission of the amorphous PFO phase. The film prepared from the solution with 2% DIO, shows a strong suppression of F8BT emission. In addition, F8BT emission spectrum is red-shifted with respect to that of the film prepared from only xylene solution, suggesting an increase of F8BT inter-fluorophore interactions. The quenching of F8BT emission is even more pronounced in the film prepared with 20% DIO, with the spectrum dominated by the PFO β-phase. Comparing with the PL spectrum of a film of neat PFO prepared from a solution with 20% DIO, which is entirely due to the PFO β-phase, in agreement with a previous report by Peet et al., [13] we conclude that the contribution from F8BT in the blend prepared from the solution with 20% DIO, is very small. The EL spectra of the LEDs prepared with the PFO:F8BT blend show a similar effect of the DIO addition to the solutions used to prepare the films, though the EL spectra show a higher contribution of the F8BT emission when comparing with PL. This we attribute to an additional contribution resulting from electronhole recombination occurring preferentially at F8BT sites. The EL spectrum of the PLEDs based on the PFO:F8BT blend has CIE color coordinates, x ¼0.31 and y¼0.59, that are typical of a green emission color. Upon addition of 2% DIO they change to x ¼0.30 and y ¼0.50, and PLEDs based on the blend film prepared from the
solution with 20% DIO have an emission spectrum with x¼ 0.24 and y¼ 0.36, that correspond to a whitish color. The CIE coordinates for the corresponding PL spectra are (0.25, 0.38), (0.34, 044) and (0.21, 0.23), which reflect the higher contribution of PFO blue emission (0.17, 0.11), when comparing with the EL spectra, discussed above. Fig. 2 compares the performance of the LEDs. Luminance reaches values around 7600 cd/m2 for devices based on the blend prepared from xylene solutions, decreasing for the films prepared from the DIO-containing solutions, down to 2000 cd/m2 for the films prepared from the solution with 20% DIO. The effect on the current follows the same trend, though without significant differences between the films prepared from solutions with different DIO content. As shown in Fig. 2B, the EL efficiency decreases when DIO is added to the blend solution, attributed to a decrease of the F8BT role in improving the electron:hole current balance and increased quenching of excitons in richer F8BT domains. Isolated F8BT chains, when in the regular blend, are expected to confine the excitons transferred from PFO and this confinement increases the PL efficiency. The addition of DIO, by promoting the formation of larger domains of F8BT, as discussed below, attenuates that confinement effect [14]. Overall, we find that the addition of DIO to the blend xylene solution suppresses the energy transfer from PFO to F8BT, under both photo and electrical excitation. The energy transfer processes in PFO:F8BT blends were studied by Buckley et al. [4] The excited state energy transfer from PFO to F8BT was rationalized in terms of a Förster resonance energy transfer, which relies on the spectral
J. Pereira et al. / Materials Letters 175 (2016) 248–251
overlap between PFO (donor) emission and F8BT (acceptor) absorption spectra and on the distance between donor and acceptor. It was concluded that at high F8BT concentration there is a nearest neighbour energy transfer, while at low concentrations this process is accompanied by exciton migration within PFO domains prior to its transfer to F8BT. Within the Förster model, a key parameter is the Förster radius, R0, which corresponds to the donor–acceptor (D–A) distance at which the probability of energy transfer equals the probability of exciton decay in the donor. This can be taken as the critical D–A distance for efficient energy transfer to occur. This theory has been developed assuming that D and A correspond to point dipoles. However, in view of the delocalized nature of the conjugated polymers excited states, the theory cannot be strictly used in such cases. An additional difficulty arises when defining the molar extinction coefficient of the acceptor. In fact, Buckley et al. [4] found that the Förster radius for PFO:F8BT would be 9 nm if the entire F8BT macromolecule is used or 4.3 nm if a 5 monomer units is considered. However, from the fluorescence time dependent studies in PFO:F8BT they obtained a Förster radius of 0.8 nm. This large difference in values points to the inadequacy of the simple model to describe the energy transfer in conjugated polymer systems. However, by considering the fundamentals of the model, we may conclude that the suppression of the energy transfer from PFO to F8BT upon addition of DIO is related to the phase separation in the blend film. In fact, as the materials are the same, only the distance between the two can explain the reduction of the energy transfer efficiency. As DIO has a much higher boiling point than xylene (365–140 °C), it is likely that, after xylene evaporation, DIO remains for a longer time in the film and promotes a higher degree of phase segregation, leading to larger domains of either polymer. The high quenching efficiency points even to the existence of domains of essentially pure polymers. This is quite unexpected as complete phase segregation in polymers is not usually found, as shown, for instance, in blends of conjugated polymers and perfluorinated saturated polymers [15]. We performed AFM studies of the films surfaces and found, as shown in Fig. 3, that the presence of DIO in the starting solutions leads to rougher films, consistent with the formation of larger phase separated domains.
4. Conclusions In conclusion, we find that the addition of DIO to the xylene solution of the PFO:F8BT blend leads to films with largely quenched energy transfer from PFO to F8BT which indicates that DIO leads to the formation of larger domains of the two polymers resulting from enhanced phase segregation. This approach is also
251
demonstrated to change the emission color of the PLEDs based on the blend, from green to a whitish emission. We believe this finding is relevant for the broad area of the polymer-based optoelectronics devices and can be further explored to fine tune the emission color of PLEDs.
Acknowledgements We thank FCT-Portugal for financial support under the contract UID/EEA/50008/2013.
References [1] (a) T. Ameri, P. Khoram, J. Min, C.J. Brabec, Adv. Mater. 25 (2013) 4245; (b) V. Gupta, V. Bhati, M. Kumar, S. Chand, A.J. Heeger, Adv. Mater. 27 (2015) 4398. [2] (a) C.R. McNeill, N.C. Greenham, Adv. Mater. 21 (2009) 3840; (b) E. Moons, J. Phys: Condens. Matter 14 (2002) 12235; (c) A. Charas, J. Morgado, J.M.G. Martinho, A. Fedorov, L. Alcácer, F. Cacialli, J. Mater. Chem. 12 (2002) 3523; (d) J.F. De Deus, G.C. Faria, E.T. Iamazaki, R.M. Faria, T.D.Z. Atvars, L. Akcelrud, Org. Electron. 12 (2011) 1493. [3] (a) C.I. Wilkinson, D.G. Lidzey, L.C. Palilis, R.B. Fletcher, S.J. Martin, X.H. Wang, D.D.C. Bradley, Appl. Phys. Lett. 79 (2001) 171; (b) J. Morgado, R.H. Friend, F. Cacialli, Appl. Phys. Lett. 80 (2002) 2436. [4] A.R. Buckley, M.D. Rahn, J. Hill, J. Cabanillas-Gonzalez, A.M. Fox, D.D.C. Bradley, Chem. Phys. Lett. 339 (2001) 331. [5] A.J. Campbell, D.D.C. Bradley, H. Antoniadis, Appl. Phys. Lett. 79 (2001) 2133. [6] J. Morgado, E. Moons, R.H. Friend, F. Cacialli, Adv. Mater. 13 (2001) 810. [7] (a) G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat. Mater. 4 (2005) 864; (b) H. Chen, J. Miao, J. Yan, Z. He, H. Wu, IEEE J. Sel. Top. Quant. Electron. 22 (2016) 4100107. [8] (a) H. Zhou, Y. Zhang, J. Seifter, S.D. Collins, C. Luo, G.C. Bazan, T.-Q. Nguyen, A. J. Heeger, Adv. Mater. 25 (2013) 1513; (b) J. Peet, J.Y. Kim, N.E. Coates, W.L. Ma, D. Moses, A.J. Heeger, G.C. Bazan, Nat. Mater. 6 (2007) 497. [9] A.K.K. Kyaw, D.H. Wang, D. Wynands, J. Zhang, T.-Q. Nguyen, G.C. Bazan, A. J. Heeger, Nano Lett. 13 (2013) 3796. [10] (a) A.K.K. Kyaw, D.H. Wang, C. Luo, Y. Cao, T.-Q. Nguyen, G.C. Bazan, A. J. Heeger, Adv. Energy Mater. 4 (2014) 1301469; (b) H.-C. Liao, C.-C. Ho, C.-Y. Chang, M.-H. Jao, S.B. Darling, W.-F. Su, Mater. Today 16 (2013) 326. [11] J. Morgado, A. Charas, J.A. Fernandes, I.S. Gonçalves, L.D. Carlos, L. Alcácer, J. Phys. D: Appl. Phys. 39 (2006) 3582. [12] M. Grell, D.D.C. Bradley, G. Ungar, J. Hill, K.S. Whitehead, Macromolecules 32 (1999) 5810. [13] J. Peet, E. Brocker, Y. Xu, G. Bazan, Adv. Mater. 20 (2008) 1882. [14] A. Charas, J. Morgado, J.M.G. Martinho, A. Fedorov, L. Alcácer, F. Cacialli, J. Mater. Chem. 12 (2002) 3523. [15] (a) C.R. McNeill, K. Asadi, B. Watts, P.W.M. Blom, D.M. De Leeuw, Small 6 (2010) 508; (b) T. Braz, A.L. Mendonça, R.E. di Paolo, J. Morgado, Submitted for publication.