Singlet Exciton Fission in Nanostructured Organic Solar Cells

Feb 28, 2011 - r 2011 American Chemical Society. 1495 dx.doi.org/10.1021/nl104202j |Nano Lett. 2011, 11, 1495-1498. LETTER pubs.acs.org/NanoLett...
0 downloads 0 Views 1MB Size
LETTER pubs.acs.org/NanoLett

Singlet Exciton Fission in Nanostructured Organic Solar Cells Priya J. Jadhav,*,† Aseema Mohanty,† Jason Sussman,‡ Jiye Lee,† and Marc A. Baldo† †

Department of Electrical Engineering and Computer Science and ‡Department of Materials Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, United States

bS Supporting Information ABSTRACT: Singlet exciton fission is an efficient multiexciton generation process in organic molecules. But two concerns must be satisfied before it can be exploited in low-cost solution-processed organic solar cells. Fission must be combined with longer wavelength absorption in a structure that can potentially surpass the single junction limit, and its efficiency must be demonstrated in nanoscale domains within blended devices. Here, we report organic solar cells comprised of tetracene, copper phthalocyanine, and the buckyball C60. Short wavelength light generates singlet excitons in tetracene. These are subsequently split into two triplet excitons and transported through the phthalocyanine. In addition, the phthalocyanine absorbs photons below the singlet exciton energy of tetracene. To test tetracene in nanostructured blends, we fabricate coevaporated bulk heterojunctions and multilayer heterojunctions of tetracene and C60. We measure a singlet fission efficiency of (71 ( 18)%, demonstrating that exciton fission can efficiently compete with exciton dissociation on the nanoscale. KEYWORDS: Organic semiconductor, photovoltaic, singlet fission, solar inglet exciton fission is observed to occur spontaneously in some organic materials where the energy of the singlet exciton, ES, is approximately twice the energy of the triplet exciton, ET. It has been extensively studied in crystalline tetracene,1,2 and it has also been reported in pentacene,3,4 anthracene at very high excitation energies,5 polydiacetylene,6 and carotenoids.7 Singlet exciton fission is of particular interest in solar cells and photodetectors because it can double their quantum efficiency. Indeed, singlet exciton fission has recently been exploited in a pentacene-based photodetector to yield an external quantum efficiency (EQE) of over 100%.8 Possibilities for sensitizers in dyesensitized solar cells have also been explored,9 and the desire for rational design of materials that exhibit singlet fission across the spectrum has prompted theoretical investigation of the process.10 Although singlet exciton fission can double the quantum efficiency of a solar cell, it also halves the potential voltage. Thus, on its own, singlet exciton fission does not necessarily benefit the performance of organic solar cells. In fact, the increase in entropy after the fission event may lower the maximum extractable power from the solar cell. To increase the efficiency of a solar cell, singlet exciton fission must be combined with a long wavelength absorber. This increases the photocurrent by absorbing photons within the energy gap between the singlet and triplet energies of the singlet exciton fission material. The net result of the combination is ideally an open circuit voltage that is defined by the absorption cutoff of the solar cell but a photocurrent efficiency that reaches 200% for the highest energy photons. This is illustrated in Figure 1a using, as an example, a combination of tetracene, copper phthalocyanine (CuPC), and the buckyball C60. The complete device implementation is shown in Figure 1b, and described below. In addition, we measure the efficiency of tetracene singlet exciton fission using the multijunction photodetector structure in Figure 1c. Finally, we

S

r 2011 American Chemical Society

confirm the presence of singlet exciton fission in blends using the bulk heterojunction solar cell shown in Figure 1d. In the planar heterojunction solar cell shown in Figure 1b, charge is generated at the CuPC/C60 donor/acceptor interface.12 Tetracene absorbs at λ < 550 nm, and it exhibits singlet exciton fission, generating two triplet excitons with energy ET = 1.2 eV. CuPC absorbs at λ < 700 nm, and its triplet energy is ET = 1.1 eV, lower than that of tetracene. Thus, triplet excitons from tetracene can diffuse through CuPC to reach the charge generation interface. The highest occupied molecular orbital (HOMO) energies of CuPC and tetracene are close, (5.2 ( 0.1) eV and (5.40 ( 0.05) eV,13,14 respectively, so hole extraction through tetracene should not present a significant barrier. Standard fabrication and characterization techniques were employed to build and measure the devices; see Experimental Methods. The thicknesses of the layers were chosen to maximize device performance. (See Figure S1 in Supporting Information.) Figure 2a shows external quantum efficiency (EQE) data for the complete planar heterojunction device and a similar device without the CuPC layer. The optical absorption obtained through optical interference modeling15 is shown in Figure 2b, and the current-voltage characteristics under an AM1.5 spectrum and 50 mW/cm2 illumination are shown in Figure 2c. Comparing the two EQEs in Figure 2a demonstrates that there is no significant decrease in the quantum yield of tetracene given the addition of CuPC, confirming that the singlet exciton fission yield is preserved in the solar cell with additional long wavelength absorption. Similarly, Figure 2c shows an increase in photocurrent from CuPC but only a slight decrease in the overall open circuit voltage. Received: December 2, 2010 Revised: January 31, 2011 Published: February 28, 2011 1495

dx.doi.org/10.1021/nl104202j | Nano Lett. 2011, 11, 1495–1498

Nano Letters

LETTER

Figure 1. Device structures. (a) Schematic of a photovoltaic exhibiting singlet fission. Tetracene and CuPC are donor materials and C60 is the acceptor. (b) Complete structure of the photovoltaic cell showing singlet and triplet exciton energies and lowest unoccupied and highest occupied molecular orbital energies. Singlets and triplets from tetracene diffuse through CuPC to the CuPC-C60 interface. BCP acts as an exciton and hole blocker.11 (c) Multijunction photodetector structure. (d) Bulk heterojunction solar cell with tetracene-C60 blend.

Figure 2. External quantum efficiency (EQE) and current-voltage data for the tetracene-CuPC-C60 planar heterojunction device. (a) EQE data under short circuit conditions. The blue line is the EQE of the complete device structure from Figure 1b, and the red line is the EQE of a similar device without the CuPC layer. (b) Absorption data calculated using optical interference modeling. The blue circles indicate the percentage change in photocurrent with the application of a 0.55 T magnetic field. |ΔI| follows the tetracene absorption at various wavelengths. (c) For the complete device, VOC = 0.48 V, JSC = 2.6 mA/cm2, and the power conversion efficiency = 1.27%. For the device without CuPC, VOC = 0.52 V, JSC = 1.0 mA/cm2, and the power conversion efficiency = 0.58%.

It is also notable that a comparison between the EQE of Figure 2a and the thin film absorption presented in Figure 2b is an unreliable estimate of the internal quantum efficiency (IQE). Integrating sphere measurements (see Figure S2 in the Supporting Information) demonstrate that the actual absorption in the device is much larger than expected, due to the influence of a rough cathode and significant optical scattering and light trapping in the cell, probably as a consequence of the polycrystalline tetracene layer. Thus, we perform two additional experiments: (i) we confirm singlet exciton fission by measuring photocurrent modulation under a magnetic field and (ii) we quantify the efficiency of singlet exciton fission by measuring the temperature dependence of photocurrent in the multijunction photodetector of Figure 1c. Previously, the rate of singlet exciton fission in tetracene has been observed to decrease under applied magnetic fields >0.05

T.1 This effect occurs because the fission involves an intermediate triplet pair state that has singlet character. A magnetic field modulates the singlet character of the pair state and hence affects the rate. The modulation was observed to saturate at about 0.3 T.1 The blue symbols in Figure 2b show our measurements of the absolute change in short circuit photocurrent under a magnetic field of 0.55 T as a function of excitation wavelength. The magnetic field effect is observed to match the absorption of tetracene, demonstrating that CuPC and C60 do not contribute to the effect. Consistent with prior results, we also find that the photocurrent modulation, |ΔI|, increases under a magnetic field of 0.1 to ∼0.3 T and then saturates (see Figure S3 in the Supporting Information). Singlet exciton fission in tetracene is thermally activated because the singlet exciton has slightly less energy, ES = 2.36 1496

dx.doi.org/10.1021/nl104202j |Nano Lett. 2011, 11, 1495–1498

Nano Letters

LETTER

Figure 3. Low-temperature EQE data and calculated singlet fission efficiency. (a) EQE data for the tetracene-C60 multijunction photodetector structure. Solid lines show measured EQE at three different temperatures. The dashed lines are simulations. (b) The relative contributions of tetracene and C60. All the EQE data in this figure have been normalized such that the contribution from C60 is constant at all temperatures. (c) Singlet fission efficiency determined from the EQE spectra. The solid line is a guide to the eye. The error bars represent EQE spectral simulations with a correlation of 0.88 or greater to the EQE data.

Figure 4. Magnetic field effect on bulk heterojunction devices with varying tetracene content. Each red dot represents one device. |ΔI| has been normalized by the tetracene content since C60 also generates current at λ = 530 nm. The line is a guide to the eye. The green cross represents the multijunction photodetector structure which has a fractional tetracene content of 0.67.

eV, than twice the triplet energy ET = 1.2 eV.2,4,16 Thus, we can measure the efficiency of singlet exciton fission in tetracene by comparing photocurrent from C60 and tetracene as a function of temperature. For example, at low temperatures, photocurrent from a device containing C60 and tetracene should be dominated by C60. We show in the Supporting Information that this technique requires only knowledge of the relative shapes of the absorption spectra of C60 and tetracene. EQE measurements as a function of temperature from 292 to 175K on a multijunction C60-tetracene photodetector structure from Figure 1c are shown in Figure 3a. Below 175 K there is a clear change in the absorption spectrum of tetracene, possibly due to a structural change in the polycrystalline tetracene thin film.17 The photodetector structure is used because (i) the thin layers of tetracene (2 nm) and C60 (1 nm) ensure that the efficiency of exciton diffusion to a charge generation interface is close to unity18 and (ii) the greater absorption within the photodetector minimizes the effects of the light trapping and cathode-induced phenomena that obstructed measurements of the singlet exciton fission efficiency in the solar cell structure of Figure 1b. The solid lines in Figure 3a show a sample of normalized short circuit EQE measurements at three different temperatures. All

the EQEs in Figure 3 have been normalized such that the C60 contribution is constant. The relative spectral contributions of C60 and tetracene are determined using optical interference modeling.15 Figure 3b breaks out the individual contributions, and the quality of the decomposition is shown using the dashed lines in Figure 3a. It is clear that the tetracene contribution drops with temperature, consistent with expectations for thermally activated singlet exciton fission. The obtained singlet exciton fission efficiency is shown in Figure 3c. The singlet exciton fission efficiency is 71 ( 18% at 292 K and 23 ( 7% at 175 K. Our low temperature measurement compares to prior measurements in single tetracene crystals of 0% at 160 K,16 and a 50% drop from 300 to 150 K4, as measured by delayed fluorescence and photoinduced absorption, respectively. Finally, to confirm the efficiency of singlet exciton fission in blended solar cells, we measure the magnetic-field-induced modulation of the photocurrent as a function of tetracene concentration in a coevaporated blend. As shown in Figure 4, we observe no significant change in the singlet exciton fission efficiency for tetracene concentrations down to 12%. Theory suggests that exciton fission can occur given two adjacent tetracene molecules,10 and these data confirm that the process can compete with exciton dissociation even in blended morphologies. The singlet fission rate constant of (6.3 ( 0.7)  108 s-1 at 0 °C in tetracene2 is much lower than typical CT state formation rates.19 This suggests that the tetracene is probably aggregating in the blends and that fission effectively competes with exciton diffusion in the tetracene aggregates. To conclude, we have demonstrated a photovoltaic cell structure that exploits singlet exciton fission leading to improved efficiency. The alternative structure of stacked multijunction organic solar cells must satisfy current matching constraints at each junction, and the complete device can be challenging to fabricate, especially using solution processing. In contrast, here we have demonstrated that singlet exciton fission is compatible with blended solar cell morphologies, suggesting that it may be employed to boost the performance of conventional solutionprocessed organic solar cells. Experimental Methods. Devices were fabricated on precleaned glass substrates coated with indium tin oxide. A layer of poly(3,4-ethylenedioxythiophene)/poly(4-styrenesulfonate) (PEDOT/PSS) was spin-coated onto the glass. All other layers 1497

dx.doi.org/10.1021/nl104202j |Nano Lett. 2011, 11, 1495–1498

Nano Letters were deposited by thermal evaporation at 3  10-6 Torr pressure. The aluminum cathode was defined by a 1 mm diameter shadow mask. Devices were packaged in a nitrogen environment between two glass slides using an epoxy seal. Optical n and k values for the materials were determined from reflection and transmission data. In the case of tetracene, which exhibits highly scattering thin films, the Kramers-Kronig rule was used to generate n and k values from absorption data collected through integrating sphere measurements.

LETTER

(16) Geacintov, N.; Pope, M.; Vogel, F. Phys. Rev. Lett. 1969, 22, 593–596. (17) Sondermann, U.; Kutoglu, A.; Bassler, H. J. Phys. Chem. 1985, 89, 1735–1741. (18) Peumans, P.; Bulovic, V.; Forrest, S. R. Appl. Phys. Lett. 2000, 76, 3855–3857. (19) Brabec, C. J.; Zerza, G.; Cerullo, G.; De Silvestri, S.; Luzzati, S.; Hummelen, J. C.; Sariciftci, S. Chem. Phys. Lett. 2001, 340, 232–236.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information regarding optimizing device performance, absorption measurement of planar solar cells, magnetic field saturation of photocurrent modulation, and measurements of singlet exciton fission efficiency. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported as part of the Center for Excitonics, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0001088 (MIT). ’ REFERENCES (1) Merrifield, R. E.; Avakian, P.; Groff, R. P. Chem. Phys. Lett. 1969, 3, 155–157. (2) Groff, R. P.; Avakian, G. P.; Merrifield, R. E. Phys. Rev. B 1970, 1, 815–817. (3) Jundt, C.; Klein, G.; Sipp, B.; Lemoigne, J.; Joucla, M.; Villaeys, A. A. Chem. Phys. Lett. 1995, 241, 84–88. (4) Thorsmolle, V. K.; Averitt, R. D.; Demsar, J.; Smith, D. L.; Tretiak, S.; Martin, R. L.; Chi, X.; Crone, B. K.; Ramirez, A. P.; Taylor, A. J. Phys. Rev. Lett. 2009, 102, No. 017401. (5) Klein, G.; Voltz, R.; Schott, M. Chem. Phys. Lett. 1972, 16, 340–344. (6) Lanzani, G.; Stagira, S.; Cerullo, G.; De Silvestri, S.; Comoretto, D.; Moggio, I.; Cuniberti, C.; Musso, G. F.; Dellepiane, G. Chem. Phys. Lett. 1999, 313, 525–532. (7) Gradinaru, C. C.; Kennis, J. T. M.; Papagiannakis, E.; van Stokkum, I. H. M.; Cogdell, R. J.; Fleming, G. R.; Niederman, R. A.; van Grondelle, R. Proc. Natl Acad. Sci. U.S.A. 2001, 98, 2364–2369. (8) Lee, J.; Jadhav, P.; Baldo, M. A. Appl. Phys. Lett. 2009, 95, No. 033301. (9) Paci, I.; Johnson, J. C.; Chen, X. D.; Rana, G.; Popovic, D.; David, D. E.; Nozik, A. J.; Ratner, M. A.; Michl, J. J. Am. Chem. Soc. 2006, 128, 16546–16553. (10) Zimmerman, P. M.; Zhang, Z. Y.; Musgrave, C. B. Nat. Chem. 2010, 2, 648–652. (11) O’Brien, D. F.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Appl. Phys. Lett. 1999, 74, 442–444. (12) Peumans, P.; Forrest, S. R. Appl. Phys. Lett. 2001, 79, 126–128. (13) Hill, I. G.; Kahn, A. J. Appl. Phys. 1999, 86, 2116–2122. (14) Kochi, M.; Harada, Y.; Hirooka, T.; Inokuchi, H. Bull. Chem. Soc. Jpn. 1970, 43, 2690–2702. (15) Peumans, P.; Yakimov, A.; Forrest, S. R. J. Appl. Phys. 2003, 93, 3693–3723. 1498

dx.doi.org/10.1021/nl104202j |Nano Lett. 2011, 11, 1495–1498