Nonradiative Energy Transfer between Porphyrin and Copolymer in

Feb 20, 2018 - This process requires a strong overlap between host emission and guest absorption without orbital alignment. In OPV, each .... For inst...
0 downloads 15 Views 1MB Size
Subscriber access provided by UNIV OF DURHAM

Article

Non-Radiative Energy Transfer Between Porphyrin and Copolymer in Films Processed by Organic Solvent and WaterDispersible Nanoparticles with Photovoltaic Applications Luana Cristina Wouk de Menezes, Wesley Renzi, Cleber Fabiano N. Marchiori, Camilla K. B. Q. M. de Oliveira, Fredrik von Kieseritzky, José L Duarte, and Lucimara Stolz Roman J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b00390 • Publication Date (Web): 20 Feb 2018 Downloaded from http://pubs.acs.org on February 21, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Non-Radiative Energy Transfer Between Porphyrin and Copolymer in Films Processed by Organic Solvent and Water-Dispersible Nanoparticles with Photovoltaic Applications

Luana Cristina Wouk de Menezes1, Wesley Renzi2, Cleber Fabiano do Nascimento Marchiori13, Camilla Karla Brites Queiroz Martins de Oliveira1, Frederick von Kieseritzky4, José Leonil Duarte2, and Lucimara Stolz Roman*1

1

Department of Physics, Federal University of Paraná, PO Box 19044, 81531-990 Curitiba,

Paraná, Brazil

2

Department of Physics, State University of Londrina, PO Box 10011, 86057-970 Londrina,

Paraná, Brazil

3

Materials Theory Division, Department of Physics and Astronomy, Uppsala University, Box

516, 75121 Uppsala, Sweden

4

Arubedo AB, Slottsbacken 8, 111 30 Stockholm, Sweden

* Corresponding Author: Prof Dr Lucimara Stolz Roman Email: [email protected] Tel: +55 41 33613275

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract In organic photovoltaic (OPV) devices, one way to increase light harvesting is to combine materials with complementary absorptions. However, the physical properties behind this process, such as Förster Resonance Energy Transfer (FRET), remain elusive. In this paper, a mixture of the metalloporphyrin Zn(5BrTTP ) and the donor-acceptor copolymer PSiF-DBT in films processed by organic solvent and water-soluble nanoparticles was investigated, and the energy-transfer rate was correlated to the bi-layer of an OPV device with a fullerene derivative (C70).

Using steady-state and time-resolved emission

studies, the FRET from the porphyrin to the copolymer was observed and found to be highest in the film processed by organic-solvent treatment at 100 °C. The devices processed by organic solvents showed superior performance to blended materials when treated at 100 °C, increasing the current without reducing open-circuit voltage. In nanoparticle systems, we observed that, with a smaller distance between the materials, higher FRET is performed. The device’s performance showed higher current as the materials were closer together. To go beyond materials with complementary absorption, the optimization of energy transfer between them might be a promising way to increase charge generation in photovoltaic devices with different morphologies.

2 ACS Paragon Plus Environment

Page 2 of 34

Page 3 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Introduction

Conjugated

organic

materials

have

been

widely

studied

as

semiconductor layers in optoelectronic devices. Their relatively low-cost manufacturing over inorganics, as well the chemical versatility in their molecular design, allows a variety of organic materials with tailored properties. In addition, features such as easy manufacturing and film flexibility for large-area coating make such materials attractive for devices with various designs and applications1,2. The high absorption coefficient of the organic semiconductor enables this technology to be used to harvest solar energy. Progress on organic photovoltaic devices (OPV) has been achieved by several research groups in recent years3–9. The power-conversion efficiency (PCE) has been increasing as a result of a better understanding of photocurrent generation in OPV devices, pushing

the

efficiency

over

13%4.

Therefore,

comprehending

the

photoconversion mechanism and the influence of physical properties upon the performance of solar cells has been adopted as a strategy to achieve more efficient devices3. One way to improve photovoltaic performance is to increase the number of excitons generated in the active layer. This can be achieved by broadening the absorption spectra of organic semiconductors, for example by using the donor-acceptor

(D-A)-copolymerization

strategy10,11.

In

D-A

copolymers,

intramolecular charge transfer (ICT) is observed, resulting in a low-band-gap material (LBG) together with a more planar configuration, enhancing the πelectron delocalization along the conjugated backbone11. ICT consequently influences the absorption profile, promoting a redshift in the π-π* band and

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

giving rise to a new absorption band corresponding to ICT between the moieties12–14. Another successful strategy to improve photon harvesting is the mixture of different materials with complementary absorptions. Besides the use of nonfullerene acceptors15, ternary-device structures have also been intensively deployed to improve solar-spectral absorption5,8,16. The ternary approach uses two-electron donors and one acceptor, or one donor and two electronacceptors, which compose a single absorber layer. However, the addition of a more electronically active component implies that the system will operate in a more complicated mechanism turning a challenge to obtain the simultaneous increase in device parameters. The system can be explained through cascade charge transfer (i.e., an orbital alignment leading to a hop of electrons and holes in the opposite direction after exciton dissociation at a donor-acceptor interface), parallel-like charge transfer (whereby the materials act like isolated binary cells), or energy transfer (whereby the exciton moves forward to the material with a smaller band gap)16. Enhancement of photovoltaic performance can be achieved through energy- transfer mechanism that improves the PCE of organic solar cells in different device architectures17–19. Mixing two donor materials, for instance, can enable Förster resonance energy transfer (FRET), improving the efficiency of bi-layer devices using a homopolymer and fullerene as an active layer20,21. FRET was originally described by Förster in 194822 as a long-range interaction with a non-radiative transfer process in donor-acceptor oscillators. These coupled oscillators perform like dipoles, and their Coulomb interaction dominates the energy-transfer rate. This process requires a strong overlap

4 ACS Paragon Plus Environment

Page 4 of 34

Page 5 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

between host emission and guest absorption without orbital alignment. In OPV, each interface with the energy cascade can lead to exciton dissociation. Thus, resonance-energy transfer is an alternative mode of interaction between different materials that is capable of broadening solar-spectral absorption without causing losses in Voc23. Optimizations of energy transfer, for example during exciton harvesting or transport until dissociation, can improve photoelectron-conversion efficiency6,23. Furthermore, FRET also depends on the interaction distance, usually around 10 nm, and, for this reason, small morphological changes can lead to different energy-transfer rates. Besides the above-mentioned works, there have been few studies addressing the effect of FRET on the interaction between electron-donor materials. Therefore, in this paper, we present two complementary absorbing materials based on a D-A copolymer (poly[2,7-(9,9-dioctyl-dibenzosilole)-alt-4,7bis(thiophene-2-yl)benzo-2,1,3-thiadiazole] - PSiF-DBT) and a metalloporphyrin (5,10,15,20-tetrakis(5-bromothien-2-yl)porphyrin zinc(II) - Zn(5BrTTP)). The energy-resonance features of this system are reported for two types of film processing: that using conventional organic solvents and that using aqueous nanoparticles. Furthermore, the films were studied as electron-donor materials in OPV devices. Our analysis here enables us to find the energy-transfer rate from porphyrin to copolymer, as well as the morphological properties of films correlated with the photovoltaic devices.

Experimental Section

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The

electron-donor

materials

used

included

Page 6 of 34

Zn(5BrTTP)

(from

Arubedo24) and the D-A copolymer PSiF-DBT (purchased from Sigma Aldrich). The chemical structures of the materials used in this study are represented in Figure 1a and 1b. The Zn(5BrTTP):PSiF-DBT blend (named following host and guest in the energy-transfer system), as well as the pristine-porphyrin and pristine-polymer active layers, were processed from a binary organic-solvent mixture (for the detailed experimental procedure, see the supporting information). For the blend processed by the organic solvent, a proportion of 1:3 w/w was considered (we also studied the proportion 3:1 w/w; for the results, see the Figure SI1, SI7 and SI8 of supporting information). Optical characterization was performed for thin films with same thickness spin coated onto a pre-cleaned quartz substrate. The UV-Vis spectra were obtained using a Shimadzu Spectrophotometer model NIR 3101. The photoluminescence spectra were recorded at room temperature on a Fluorolog– 3

Jobin

Yvon–Spex.

The

samples

were

excited

at

440

m.

The

photoluminescence-decay time was measured using a Fluo Time 200, from PicoQuant, based on the time-correlated single-photon-counting (TCSPC) technique equipped with an MCP detector. The excitation was performed with a 440-nm diode-pulsed laser. The aqueous nanostructures of

the PSiF-

DBT:Zn(5BrTTP) blend were synthesized by a mini-emulsion technique described in a previous paper25. As in the other steps, the synthesis was performed under an environmental atmosphere. Two different nanostructures were considered. In the first case, a solution containing the PSiFDBT:Zn(5BrTTP) mixture was processed, resulting in nanoparticles containing both materials; in the second case, the solutions containing the pristine

6 ACS Paragon Plus Environment

Page 7 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

materials were processed, resulting in nanoparticles containing only one material. For the second case, after preparation of the nanoparticles, a mixture of this solution was prepared, forming a blend containing both kinds of nanoparticles. Photovoltaic devices were built onto a pre-cleaned glass/ITO substrate containing a 40-nm-thick layer of PEDOT:PSS (Sigma-Aldrich), spin coated and annealed at 100 °C for 15 min in vacuum. The active layer was processed using the same solutions described above. After the deposition, films were treated at 100°C for 15 min in the vacuum. The as-cast device was built following the same procedure, except for the thermal process. The fullerene C70 from Sigma-Aldrich (Figure 1c) was sublimated into a vacuum chamber onto the donor layer after which a layer of aluminum was evaporated (the device’s structure is illustrated in Figure 1d). Photovoltaic characterization was performed using a Keithley picoamperimeter with power supply (model 6487) and a monochromator/spectrometer (1/4 m Oriel). The solar simulation was performed using an illumination power of 100 mWcm-2 from a 150-W Oriel Xenon lamp with an air-mass (AM1.5) filter. To understand the interaction between the two materials and the morphology formation, atomic-forcemicroscopy measurements (AFM) in the intermittent mode (Shimadzu SPM9700).

Results and Discussions

Optical Analysis The porphyrin with Zn as a metallic base and thiophene rings containing Br in their meso-positions [Zn(5BrTTP)] has a chemical structure similar to that

7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of a natural absorber24. The absorption spectra for the thin film shown in Figure 2a exhibits a peak at 450 nm named the Soret band, as well as another (Quiral) band with a smaller energy. Due to the excited state having a different geometry from ground state, the porphyrin emission under Soret-band excitation showed a shoulder at 625 nm and an intensity peak at 675 nm. After thermal annealing at 100 °C, no significant changes in absorption spectra were observed. Nevertheless, photoluminescence emission changed – intensifying the first peak and shifting it into the blue region (Figure 2a) – suggesting a lower aggregates formation after thermal treatment. For instance, the presence of aggregates (“ face-to-face” or “tilted-deck-of-cards”) in free-base porphyrin has been found to red-shift the photoluminescence peak26. A broad absorption spectrum with two defined peaks (Figure 2b) is observed for the PSiF-DBT copolymer. The peak at 390 nm in the absorption spectrum can be attributed to the π-π* transition, while the second at 560 nm, can be assigned to the ICT transition12–14 owing to the thiophene-donor and benzothiadiazole-acceptor units of the copolymer. On the other hand, the photoluminescence, peaked in the 690 nm, is dominated by emission from the ICT transition. Under thermal treatment at 100 °C, the optical absorption of the copolymer does not change, but a small reduction in the emission intensity is observed. The copolymer after thermal treatment showed an increase in non-radiative quenching above that in as-cast film. This suggests a change in the structural order of the copolymer film following thermal treatment. This trend has been found in different conjugated polymers27,28. For instance, the ordered polymeric chains of F8T2 with DIO treatment had a higher photoluminescence intensity due to a reduction in nonradiative quenching28.

In the case of PSiF-DBT copolymer, the thermal

8 ACS Paragon Plus Environment

Page 8 of 34

Page 9 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

treatment decreased the structural order, leading to a reduction in the photoluminescence efficiency. Although PSiF-DBT has a broad absorption spectrum ranging from 300 nm up to 650 nm, there is a lower absorption-intensity region close to 450 nm, coinciding with the Soret absorption band of Zn(5BrTTP). Moreover, the emission peak of porphyrin overlaps with the 560-nm absorption band of PSiFDBT. In this case, Zn5BTTp might act as the host and PSiF-DBT as the guest in an energy-transfer system. Figure 2c shows the absorption and emission spectra of Zn(5BrTTP):PSiF-DBT with and without treatment. The absorption profile of blend fits the absorption of the neat material, indicating that the presence of porphyrin in the copolymer host matrix does not affect the interchain transitions between the D-A components of PSiF-DBT. After annealing at 100 °C, a reduction in copolymer-absorption intensity was observed, especially for the second peak. This suggests that after thermal treatment, the porphyrin is closer to copolymer chain loading to torsion on benzothiadiazole units reducing the LUMO and HOMO interaction of copolymer14. This phenomenon can also be observed when there is a higher amount of porphyrin in the film, as reported in the supporting information (Figure SI1). The emission spectra of Zn(5BrTTP):PSiF-DBT without treatment present higher emission intensities of both pristine characteristic bands than in the isolated materials. However, the porphyrin-characteristic shoulder vanishes when the film is annealed. There are two possible explanations for this behavior: firstly, the amount of porphyrin in the blend film is lower than that in the pristine film, consequently lowering the emission intensity. This peak was

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 34

observed before treatment, and hence the hypothesis is not plausible. Another probable explanation is that the observed quenching is ruled by resonanceenergy transfer between the two chromophores, which is favored after annealing. The energy-transfer rate,  , is dependent on the excited-state lifetime of the host ( ), the distance between host-guest dipoles ( ), and the Forster Radius (R0), as expressed in Equation (1)22:





 =  





(1)



R0 – the limit distance at which non-radiative and radiative emission have the same probability – is given by:

 =

 9,78  10  " !

*

.

# $% (')Ɛ) (')' +', ,

(2)

where k, QD, and n are the alignment factor between the host and guest dipoles, the host photoluminescence quantum efficiency and the environmental refractive index, respectively. FD is the host emission spectrum, and ƐA is the molar extinction coefficient of the guest, which indicates the orbital overlap requirement between chromophores states. The energy-transfer rate has an inversely proportional dependence upon the host-guest distance and direct dependence upon dipole orientation (intrinsically introduced by R0). These parameters seem to change when reorganization energy is applied to the

10 ACS Paragon Plus Environment

Page 11 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

system; for example, k increases by realignment or reducing the distance between dipoles . According to FRET theory22, as the transfer presents an additional nonradiative decay path for the host, the host experiences a decrease in its excitedstate lifetime as guest concentration increases. TCSPC analysis was performed to investigate energy transfer between the porphyrin and the copolymer on mixed films treated at different temperatures (Figure 2d). The average lifetime of the porphyrin excited state decreased from 0.54 ns to 0.48 ns with the treatment at 100 °C. Further investigation was performed; during treatment at 200 °C, the lifetime was reduced to 0.46 ns (Figure SI2 of supporting information).

The decrease in the lifetime suggests that, after thermal

annealing, there is an increase in the energy-transfer rate. The efficiency of energy transfer can be calculated using:

E = 1 − τ23 /τ2

(3)

where τ23 and τ2 are the lifetimes of the host with and without the guest, respectively. Therefore, by maintaining the host-guest proportion and only changing the system’s organization through thermal treatment, the energytransfer rate is improved by 10%. As discussed above, the energy-transfer rate depends on the distance (r) among the resonant dipoles. Although FRET is a long-range interaction, the Förster radius is the upper limit to the radiative interaction between the materials. This suggests that after thermal treatment, the distance between the materials may change.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 34

To gain insight concerning energy transfer between the porphyrin and the copolymer, a morphological model comprising water-soluble nanoparticle synthesis was used. The nanoparticle system allows the distance between two materials to be controlled by encapsulation. During the process

of

nanoparticle’s synthesis, following the miniemulsion technique, there are two different solutions: the “oil phase” composed of organic materials (copolymer or porphyrin or copolymer+porphyrin diluted in the organic solvent) and the aqueous phase composed of surfactant and water. After the synthesis steps, the material in “oil phase” will be inside of the nanostructure delimited by the quantity of surfactant. As represented in Figure 2e, two different solutions containing nanoparticles were synthesized. In the first case, two components (host and guest) are present inside the same nanoparticle. In the synthesis, the organic solvent is dissolving the blend of Zn5(BrTTP):PSiF-DBT. So, both materials are inside of nanostructure. Thus, there is a limit to the interaction range due to the materials’ dispersion dependence upon nanoparticle diameter. This structure permits a maximum non-radiative interaction between these resonant materials. Dynamic light-scattering measurements gives an average diameter of 80 nm. The second case consists of a mixture containing single-component nanoparticles. The single-component nanoparticle means that the film has the nanoparticles with pure materials. In this case, the films are made with the mixture of Zn(5BrTTP) and PSiF-DBT nanoparticles, but there is only one material (single) inside of nanostructure, whereas the two-component system, the Zn(5BrTTP):PSiF-DBT are inside of the same nanoparticle. This structure (single-component) allows an interaction beyond the limit of non-radiative

12 ACS Paragon Plus Environment

Page 13 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

transfer. Aiming this mixing behavior, the synthesis is performed separately: the synthesis of only polymer, and the synthesis of only porphyrin. After these steps, both copolymer and porphyrin nanoparticles are mixed in order to acquire the pristine materials isolated inside of nanoparticle. All nanoparticles were made under the same conditions to obtain an average diameter of 80 nm. Hence, there is no energy transfer between the chromophores due to the distance being too large. Interaction between the polymer and porphyrin depends upon the distance between them. The optical measurements can confirm the performance of materials described in the schematic diagram. The Figure 2f and 2g display the optical properties of nanoparticles’ film in single-component and two-component system, respectively. Both films exhibit the absorption profile correspondent to the porphyrin and the copolymer. In the case of the single-component nanoparticle, the absorption is similar to the sum of pristine materials, representing that the materials are in different nanoparticles. Considering the film of nanoparticle with two component, the absorption shows an interaction between the materials due to their proximity. For the single component, there is emission from both materials, Zn(5BrTTP) and PSiF-DBT nanoparticles (Figure 2f and Figure SI3). In twocomponent case, there is a high probability of exciton transfer without losses by radiative recombination, indicating a higher energy-transfer rate than in films with individual nanoparticles (Figure 2g). Excited-state-lifetime measurements (Figure 2h and Table SI1) confirm the shorter time to nanoparticle with closest materials (0.44 ns and 0.32 ns to individual and two components, respectively). Moreover, the exciton lifetime in nanoparticle systems becomes shorter than

13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

that in the morphology with an organic solvent (0.32 ns and 0.46 ns, respectively). This indicates that the interaction between Zn(5BrTTP) and PSiFDBT becomes stronger in the nanoparticle system, allowing exciton hopping to be easily performed, another advantage of the nanoparticle film. By studying the photoluminescence shape, this model can be used to correlate the morphological behavior of materials processed by an organic solvent and nanoparticles29. The photoluminescence lineshape of nanoparticles matches with emission from films prepared by organic solvents (Figure SI4), suggesting that, as observed during nanoestructuration, higher interfacial contact between porphyrin and copolymer is achieved after treatment at 100 °C than in films without treatment.

Bi-layer Photovoltaic Device with Two Donors To elucidate the energy transfer between Zn(5BrTTP) and PSiF-DBT in organic photovoltaic devices, bi-layer devices with C70 were chosen. The copolymer reported here has been used as an active layer in organic photovoltaic devices14,30–33. The porphyrins are usually employed as sensitizer for dye-sensitized solar cells34,35, but recently, they have been applied to organic solar cells as electron donors36,37 or bulk-heterojunction additives38–40. For this reason, and with the aim of using the complementary absorption profile as well as the FRET mechanism, we employed the Zn(5BrTTP):PSiF-DBT layer combined with an electron-acceptor fullerene derivative, C70, as the active layer in the OPV.

Here, we will discuss the results for the blend containing the

14 ACS Paragon Plus Environment

Page 14 of 34

Page 15 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

majority proportion of PSiF-DTB, i.e., in the proportion 1:3 w/w. This suggests that the polymer is majorly responsible for device operation. Figure 3a shows the current-density-voltage (J-V)-characteristic curves measured under one sun illumination (simulated AM 1.5G irradiation at 100 mW cm-2). The parameters of the devices that combined with the means value of 10 devices are summarized in Table 1. Devices using pristine materials as their active layer without thermal treatment have PCEs of 1.62% and 0.71% for PSiFDBT and Zn(5BrTTP), respectively. With a short-circuit current density (Jsc) of 5.05 mA cm-2 , an open-circuit voltage (Voc) of 0.60 V, and a fill factor (FF) of 52%, the PSiF-DBT showed a better performance than Zn(5BrTTP) (4.57 mA cm-2, 0.40 V, and 38%, respectively). For devices in which the donor layer was the Zn(5BrTTP):PSiF-DBT (1/3 w/w), an enhancement in the JSC (6.48 mA cm-2) and FF (55%) values were observed, compared with the pristine devices. This resulted in an increase of 27% in JSC without losses in VOC. An improvement greater than 37% in PCE which reached 2.24%, was achieved. This enhancement in JSC corresponds to an increase in external quantum efficiency (EQE) (Figure 3b). The EQE spectra in the 400—500-nm region increases in donors in the blended device compared with the pristine one. This result can be explained by the complementary absorption of materials. The efficiency is higher in the mixture around 300—400 nm and 500—700 nm than in the pristine material, implying that charge recombination has been reduced in this region. Lower charge recombination suggests an improvement in FF. Thus, porphyrin inclusion can also improve exciton diffusion to the interface with fullerene. As reported, the efficiency of a bi-layer device can be enhanced through a change in the electron-donor/acceptor interface30,41. To investigate

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

this phenomenon, film topography was characterized by AFM measurement (Figure 3c). It is possible to verify that the root mean square of the mixture is higher than that of pristine material. The addition of porphyrin changed the film surface roughness (around 60%), enabling an increase in the contact area between electron-donor/acceptor layers. This improvement leads to a higher efficiency in exciton dissociation, as observed in the JSC and EQE results. After thermal treatment at 100 °C of the donor-blend film, a better performance was observed, even with the reduction in the PSiF-DBT absorption-spectral region. The JSC was improved up to 9.32 mA cm-2, 43% higher than that obtained from as-cast devices. The FF was reduced to 47% and VOC was kept the same, causing an efficiency improvement to 2.72% (23% higher than a blend without treatment). The performance of donor-blend film treated at 100 °C was also higher than pristine materials in the same condition (Figure SI5 and Table SI2). Keeping in mind that the porphyrin is the minor component of the blend, one might expect that an exciton generated therein could not diffuse to the interface with fullerene, resulting in losses by recombination. However, due to FRET between these two materials, the excitation energy can be transferred from the porphyrin to the copolymer. Therefore, the resonant excitons can diffuse through the polymeric chain and be dissociated at the donor-blend/fullerene interface. In this case, the energy is not lost, but induces the generation of more excitons in the PSiF-DBT phase, which eventually dissociate into free charge carriers. Thus, the reduction in the absorption range corresponding to the copolymer does not affect device performance because of resonant excitons. Moreover, devices treated at 100 °C show an intensified efficiency in the 400—500-nm range, compared to the

16 ACS Paragon Plus Environment

Page 16 of 34

Page 17 of 34 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

maximum peak (~390 nm). This is the range in which porphyrin is the main contribution to the photocurrent. Improvement in this wavelength may be due to an increase the FRET rate, since those excitons not contributing to photocurrent can be transferred when the porphyrin is excited. Together with the additional exciton formation in the PSiF-DBT, morphological analysis showed that in the thermal-treated films the increase of the thin-film surface roughness has been moderate (around 20%). So, the improvement of JSC with thermal treatment cannot be ascribed solely to a larger contact area between the donor and acceptor induced by the slightly higher roughness of donor film. This result indicates that the higher FRET between the Zn(5BrTTP) and the PSiF-DBT may contribute to the increase on JSC after treatment at 100 °C. In addition to the exciton dissociation, the optimization of energy transfer in two-donor system is necessary to improve the photovoltaic performance. The reasons behind the change in the photovoltaic device were also investigated by hole-transport mobility. Considering that the devices have approximately the same thickness, a non-linear J-V curve approximately following a J ∝ V² at high voltage is observed. Using the Mott-Gurney law (6 = 977 8(9 − 9:; ) /8