October 7, 2008

Solar Energy-Conversion Processes in Organic Solar Cells

By Xu, Zhihua Zang, Huidong; Hu, Bin

Organic semiconducting materials have demonstrated attractive light-absorption and photocurrent-generation functions due to their delocalized pi electrons as well as intra-molecular and inter- molecular charge separation processes. On the other hand, organic semiconducting materials have easy property tuning, are mechanically flexible, and have large-area thin film formation properties. As a result, organic materials have become potential candidates in solar energy applications. This article will review critical energy- conversion processes in organic solar cells with the focus on singlet and triplet photovoltaic responses. INTRODUCTION

Conjugated polymers, traditionally called conducting polymers, have attracted much attention for their applications in organic solar cells (OSCs). Due to alternative single- and double-bond configuration, the conjugated polymers have a sigma-bond backbone with the remaining out-of-plane p^sub z-orbital forming pi-bonds with neighboring p^sub z^ orbitals. The electrons in p^sub z- orbitals, namely pi-electrons, can then delocalize over the entire molecule. Because the wavefunction overlap of p^sub z^ orbitals forms both a bonding (pi) orbital with lower energy and an antibonding (pi*) orbital with higher energy, the electronic structures essentially consist of highest occupied molecular orbitals (HOMO) and lowest unoccupied orbitals (LUMO). The HOMO and LUMO are also named as low-energy valence and high-energy conduction bands. Furthermore, the valence and conduction bands form direct band structures in conjugated polymers, allowing direct electronic transitions for pi electrons across the bandgap defined as the minimal energy difference between the valence and conduction bands.

A conjugated polymer molecule can absorb a photon when an incident photon has energy larger than the band gap, leading to photon absorption. In general, conjugated polymers have large light absorption coefficients due to facile electrical polarization of delocalized pi electrons. The photon absorption leads to an electronic transition with the outcome of generating an electron in LUMO and a hole in HOMO in an organic conjugated polymer molecule. The photon absorption-generated electron and hole can largely bond together to form an intra-molecular electron-hole pair, named as Frenkel exciton, due to Coulomb attraction. Due to the low dielectric constants in organic materials, the photon excitation- generated excitons have large binding energies typically on the order of 1.0 eV.1,2 There are three different channels for an exciton to release its energy. The first channel is light emission, which forms a mechanism for conjugated polymers to be used in solid- state lighting applications. The second channel is multi-phonon nonradiative emission.3,4 This multi-phonon emission converts excitonic energy into thermal vibration. The third channel is exciton dissociation into free electrons and holes through either the Poole-Frenkel process5 due to thermal activation or the Onsager process6 due to diffusion and Coulombic interaction.

The exciton dissociation efficiency is quite low due to the large excitonic binding energy in organic semiconducting materials. Therefore, sufficient electrical force must be applied to overcome the excitonic binding energies for dissociation in organic solar cells. It has been found that donor-acceptor interaction, formed between two different organic molecules, can effectively dissociate photon absorption-generated excitons and generates significant photocurrent. There are two parameters, energy band offsets and electrical dipole-dipole interaction, that determine the donor- acceptor interaction and consequent exciton dissociation. The energy band offsets facilitate the exciton dissociation in energy.7,8 With the energy band offsets, the exciton dissociation leads to the dissociated electron transferring to the relatively lower-energy LUMO in the acceptor molecules and the hole remaining in the HOMO in the donor molecules. The dipole-dipole interaction forms an electrical force between donor and acceptor molecules to overcome the electron-hole binding energy and thus enhances the exciton dissociation.9 The dipole-dipole interaction is determined by the electrical polarization of donor and acceptor molecules. Notably, the use of electrical polarization provides an additional pathway to adjust the donor-acceptor interaction in organic solar cells. To complete sunlight-to-electricity conversion, the dissociated electrons and holes need to travel to the cathode and anode through respective transport channels. Therefore, the efficiency of OSC is determined by the following three individual processes: light absorption, exciton dissociation, and charge collection.

The characterization of OSC is normally conducted under standard sunlight illumination from calibrated white light named as the condition of AM1.5. Three parameters are considered in the calculation of power conversion efficiency (PCE) in organic solar cells: short-circuit current (I^sub SC^), which is the photocurrent measured at short-circuit condition; open circuit voltage (V^sub OC^), which is the voltage required to balance the current to zero; and fill factor (FF), which is defined as FF=I^sub M^V^sub M^ / I^sub SC^V^sub OC^, where I^sub M^ and V^sub M^ are the current and voltage with the maximal product. The power conversion efficiency of solar cells is then given as Equation 1. (See Equations table.)

There are three critical issues in improving the PCE of organic solar cells. First, the optical absorption spectrum should be sufficiently broad to maximize the sunlight absorption. The research effort has been directed to the synthesis of narrow bandgap materials and the use of multiple sunlight absorption materials. Second, the open circuit voltage V^sub OC^ is relatively low compared to the photon energy of absorbed sunlight. It has been suggested that the energy band alignment at the donor/acceptor interface is directly related to the V^sub OC^ value.10,11 Third, the nonradiative multi-phonon emission can consume excitons before the excitonic dissociation process occurs. Finally, the dissociated electrons and holes can still bind together due to Coulomb attraction, forming a charge transfer complex at the interface between donor and acceptor molecules. Theoretically, high charge mobility and large donoracceptor dipole interaction can reduce the possibilities to form charge transfer complex from dissociated electrons and holes.12

Organic solar cells have developed with two different architectures, double-layer structures and interpenetrating molecular networks, based on donor and acceptor materials. In the double-layer design, donor and acceptor materials are prepared as two adjacent different layers between the anode and cathode. In the bulk-heterojunction structure, the donor and acceptor materials are mixed to form a single layer between the two electrodes. Since the first double-layer OSC was reported with 1% PCE in 1986 by C.W. Tang,13 continuous progress has been made in the improvement of PCE. The highest PCE efficiency witii the double-layer design has reached 4%.14,15 Recently, tandem architecture has pushed the PCE up to a value higher than 5%.16 On the other hand, bulk heteroj unction design was first introduced by Heeger's group in the early 1990s.7 Bulk-heterojunction design has led to around 5% PCE based on poly[3- hexylthiophene] (P3HT) and surface-functionalized fullerene 1 -(3- methyloxycarbonyl) propyd-phenyl [6,6] C^sub 61^ (PCBM).17,18 Furthermore, the PCE has increased to 5.5% by using low-band-gap conjugated polymer po!y[2,6-(4,4-bis-(2-ethylhexyl)-4H- cyclopenta[2,1 -b;3,4-b]-dithiophene)-alt-4,7-(2,1,3- benzothiadiazole)] (PCPDTBT).19 Recently, tandem solar cells based on P3HT:PCBM and PCPDTBT:PCBM composites have shown high PCE of 6.5%.20


Conjugated Polymers

Sunlight absorption generates singlet excitons in conjugated molecules. However, the photoexcitation generated singlet excitons can convert into triplet excitons through intersystem crossing (ISC) due to spin-orbital coupling.21 In a singlet exciton, the electron and hole have anti-parallel spin orientation. However, the electron and hole have parallel spin orientations in a triplet exciton. Because of the spin-exchange interaction, triplet excitons usually have lower energies than the corresponding singlet excitons. On the other hand, triplet excitons have stronger binding energies, larger lifetimes, and long diffusion lengths relative to singlet excitons. It is particularly noted that the intersystem crossing between singlet and triplet states is determined by three parameters: the atomic number Z, the distance between the electron and the nucleus, and the energy difference between singlet and triplet states.22,23 Adjusting three parameters can largely change the singlet-triplet ISC in organic materials. It has been reported that the triplet densities in poly[2-methoxy5-(2'-ethylhexyloxy)-!,4- phenyIenevinylene] (MEH-PPV), aluminum (III) 8-hydroxyquinoline (Alq^sub 3^), and P3HT are 1%,24 20%,25 and 70%,24 respectively. As a result, both singlet and triplet excitons can exist in conjugated molecules under sunlight absorption. It is particularly noted that singlet and triplet excitons can have significantly different involvements in the photocurrent generation due to their different lifetimes, binding energies, and spin polarizations.26 It has been a challenging issue to experimentally visualize both singlet and triplet photovoltaic processes in organic solar cells. Recently, we found that magnetic field effects of photocurrent (MFP) can be used as an effective experimental tool to reveal how singlets and triplets convert into photocurrent.27-29 The value of MFP can be determined by the relative change in photocurrent caused by applied magnetic field as given in Equation 2, where PC^sub B^ and PC^sub 0^ are the photocurrents measured with and without applied magnetic field B, respectively. In general, the MFP may consist of an increase and a decrease components, corresponding to positive and negative MFPs (Figure I a). It has been suggested that the positive MFP comes from the facts that (i) the magnetic field modifies singlet/triplet ratio at polaron-pair states and (ii) the singlet polaron pairs have higher dissociation rates relative to triplet polarons,30,31 The negative MFP can be attributed to the exciton- charge reaction.32,33 In essence, the triplet excitons can dominate the exciton-charge reaction due to sufficient physical contact with charge carriers based on their long lifetimes. Therefore, MFP measurement can reveal two photovoltaic processes in conjugated polymers: dissociation in polaron-pair states evolved from singlet excitonic states and the exciton-charge reaction occurred in triplet excitonic states in the generation of photocurrent (Figure 1b).

Clearly, MTT measurements can provide singlet and triplet photovoltaic processes in organic conjugated polymers. Figure 2 shows the MFP experimental results for MEH-PPV, Alq^sub 3^, and P3HT, respectively. For MEH-PPV with very low triplet density (1%), only positive MFP can be observed, which indicates the singlet dissociation is the dominant photovoltaic process. For P3HT and Alq^sub 3^ with higher triplet densities (20% and 70%), both positive and negative MFP can be observed. This suggests the existence of two photovoltaic channels: the dissociation in polaron- pair states and the charge reaction in excitonic states.

Bulk Heterojunction Solar Cells

It is widely accepted that donor/acceptor interaction is the necessary condition to effectively dissociate the excitons and to achieve high photovoltaic efficiencies. The most common material used as an electron acceptor in bulk-heterojunction solar cells to form donor/acceptor interaction with conjugated polymers is PCBM. It has been recently found that the MFP can show detailed photovoltaic processes in the popular P3HT:PCBM bulk-heterojunction solar cells.29 Figure 3 shows that at low concentrations (1 wt.%) within a uniform dispersion regime leads to a decrease in the amplitude of the negative magnetic field effect of photocurrent. The 5 wt.% PCBM dispersion gives a flat magnetic field dependence of photocurrent in low field from 0 mT to 150 mT. The flat magnetic field dependence of photocurrent indicates that the strong donor-acceptor interaction at high PCBM concentration removes both dissociation of polaron pairs and the exciton-charge reaction. This implies that high- concentration PCBM directly dissociates both singlet and triplet excitons to generate photocurrent in bulk-heteroj unction solar cells.

The dissociated electrons and holes yield either useful or non- useful outcomes. First, the dissociated charge carriers can transport through respective networks to generate photocurrent, leading to a useful outcome. Second, the dissociated charge carriers can recombine by forming bound electron-hole pairs-charge-transfer complexes due to attractive Coulombic interaction at the donor- acceptor material interfaces-giving a non-useful outcome. Clearly, the ratio of useful-to-non-useful outcome affects the photovoltaic efficiency in organic solar cells. As shown in Figure 4a, the highly dispersed PCBM leads to a positive magnetic field effect of photocurrent in a relatively high field range between 150 mT and 900 mT while in a low field the magnetic field effects of photocurrent remain flat from O mT up to 150 mT in the bulk heteroj unction P3HT+PCBM solar cell. This high-field positive magnetic field effect of photocurrent can be suggested as the signature of charge- transfer complex states formed from dissociated electrons and holes at the P3HT-PCBM interfaces. It can be seen in Figure 4a that the amplitude of this high-field magnetic field-dependent photocurrent depends on the PCBM weight ratio in the P3HT+PCBM solar cell. The 1:0.8 weight ratio of P3HT:PCBM yields the lowest high-field MFP and the lowest density of charge-transfer complexes. This lowest density of charge-transfer complexes corresponds to an optimized fill- factor of 0.61 and highest PV efficiency in the ITO/Pedot/P3HT+PCBM/ Al solar cell (Figure 4b). This result suggests that high-field MFP measurement can be applied to study the charge recombination processes in bulk heterojunction organic solar cells.


As discussed above, due to different binding energies, lifetimes, and diffusion lengths, the singlet and triplet excited states have different dissociation and recombination processes. This indicates that adjusting singlet/triplet ratio might be an effective way to enhance PV efficiency. It is known that the spin-orbital coupling determines the intersystem crossing and consequently the singlet and triplet ratios in organic materials. There are two ways to increase the spin-orbital coupling: chemically attaching heavy-metal atoms to organic molecules34 (i.e., the internal heavy atom effect) or physically dispersing heavy metal particles into organic materials35,36 (i.e., the external heavy atom effect). The internal heavy-atom effect requires delicate organometallic reactions to systematically change the spin-orbital coupling strength. The external heavy-atom effect can be readily obtained by dispersing heavy metal atoms into organic materials. However, the insolubility of metal particles together with the large discontinuity of dielectric constant at the material interface creates a significant difficulty in obtaining a uniform dispersion and an effective interfacial interaction in organic metallic material composites. We found that mixing two soluble organic materials, strong-spinorbital- coupling Ir(ppy)^sub 3^ molecules and weak-spin-orbital-coupling MEH- PPV, can increase the effective spin-orbital coupling in organic solar cells.28 Due to the high polarizability of heavy atoms, heavy metal complexes are also expected to contribute to exciton dissociation by enhancing the electrical polarization at the donor/ acceptor interface.

Figure 5 shows that the 5% of Ir(ppy)^sub 3^ reduces the high- field MFP and consequently improves its photovoltaic response in the MEH-PPV: PCBM (1:4) solar cell. The reduction of high-field MFP upon doping Ir(ppy)^sub 3^ molecules clearly shows that the dispersed Ir(ppy)^sub 3^ molecules minimize the formation of charge transfer complex from the recombination of dissociated electrons and holes at the interfaces between MEH-PPV chains and PCBM molecules. This experimental result suggests that the increase in spinorbital coupling and triplet density essentially decrease the recombination probabilities for dissociated electrons and holes to recombine toward the formation of the charge-transfer complex. As a consequence, the dispersion of heavy-metal complex Ir(ppy)^sub 3^ molecules can lead to significant improvement of photovoltaic efficiencies in organic solar cells.


Conjugated polymers have demonstrated attractive photovoltaic response in organic solar cells based on the strong interaction between delocalized pi electrons and incident sunlight photons and the facile formation of heterojunctions through multi-layer and bulk blend designs. Recent magnetic studies of photocurrent have found that both singlet and triplet excited states can contribute to the photocurrent generation. However, the singlet and triplet excited states undergo different photovoltaic channels, dissociation and charge reaction, in the sunlight-photocurrent conversion. Specifically, at weak donor-acceptor interaction the dissociation in polaron-pair states and the charge reaction in excitonic states are two dominate photovoltaic processes. When the donor-acceptor interaction is sufficiently strong, the photovoltaic channel only consists of the direction dissociation of both singlet and triplet excitons.

Because both singlet and triplet excitons can be effectively dissociated by strong donor-acceptor interaction, adjusting singlet and triplet excitons becomes a new pathway to further improve the photovoltaic efficiencies based on the modification of spin-orbital coupling by using heavy-metal complex structures. In addition, the use of heavy-metal complex structures can also reduce the formation of charge-transfer complex from the recombination of dissociated electrons and holes and consequently further increase the photovoltaic response. In summary, the magnetic field effects of photocurrent and the use of heavy-metal complex structures are two effective experimental approaches to revealing singlet and triplet photovoltaic processes and to improving photovoltaic efficiencies in organic solar cells. ACKNOWLEDGEMENT

This research was supported by the Air Force Office of Scientific Research (FA9550-06-10070) (Dr. Charles Lee, Program Officer), National Science Foundation Career Award (ECCS0644945), and Center for Materials Processing at the University of Tennessee.

How would you...

...describe the overall significance of this paper?

This article shows the detailed energy-conversion processes together with updated information of research progress in organic solar cells. The included experimental results can elucidate intrinsic singlet and triplet photovoltaic processes in organic materials. The discussions summarize the challenging issues and the possible solutions in the improvement of photovoltaic efficiencies in organic solar cells,

...describe this work to a materials science and engineering professional with no experience in your technical specialty?

This work uses general terminology to describe complex photovoltaic processes in organic solar cells targeted for readers from different disciplines.

...describe this work to a layperson?

This review article presents the importance of organic solar cells as well as a general but unique description of solar energy- conversion processes in organic materials.


1. M. Chandross et al., Phys. Rev. B, 50 (1994), pp. 14702- 14705.

2. D.P. Craig and S.H. Walmskey, Excitons in Molecular Crystals (New York: Benjamin, 1968).

3. D. Beljonne et al., J. Am. Chem. Soc., 118 (1996), p. 6453.

4. J. Seixas de Melo et al., J. Chem. Phys., 118 (2003), p. 1550.

5. J. Frenkel, Phys. Rev., 54 (1938), p. 647.

6. I. Onsager, Phys. Rev., 54 (1938), p. 554.

7. N.S. Sariciftd et al., Science, 258 (1992), pp. 1474-1476.

8. X. Wet et al., Phys. Rev. B, 53 (1996), p. 2187.

9. X. Sun et al., Synth. Met., 101 (1999), pp. 263-264.

10. D.C. Olson et al., Adv. Funct. Mater., 17 (2007), pp. 264- 269.

11. C.M. Ramsdale et al., J. Appl. Phys., 92 (2002), pp. 4266- 4270.

12. S.A. Choulis et al., Appl. Phys. Lett., 83 (2003), pp. 3812- 3814.

13. C.W. Tang, Appl. Phys. Lett., 48 (1986), pp. 183-185.

14. P. Peumans, S. Uchida, and S.R. Forrest, Nature, 425 (2003), p. 158.

15. J. Xue et al., Appl. Phys. Lett., 84 (2004), pp. 3013-3014.

16. J. Xue et al., Appl. Phys. Lett., 85 (2004), pp. 5757-5759.

17. G. Li et al., Nature Mater., 4 (2005), pp. 864-868.

18. W.L. Ma et al., Adv. Funct. Mater., 15 (2005), pp. 1617- 1622.

19. J. Peet et al., Nature Mater., 6 (2007), pp. 407-500.

20. J.Y. Kim et al., Science, 317 (2007), p. 222.

21. D. Beljonne et al., J. Phys. Chem., 105 (2001), pp. 3899- 3907.

22. S.P. McGlynn, J. Daigre, and F.J. Smith, J. Chem. Phys., 39 (1963), pp. 675-679.

23. H.D. Burrows et al., Chem. Phys., 285 (2002), pp. 3-11.

24. H.D. Burrows et al., J. Chem. Phys., 115 (2001), pp. 9601- 9606.

25. H.D. Burrows et at., J. Am. Chem. Soc., 125 (2003), p.15310.

26. A. Kohler and D. Beljonne. Adv. Fund Mater., 14 (2004), pp. 11-18.

27. Z.H. Xu, Y. Wu, and B. Hu, Appl. Phys. Lett., 89 (2006), no. 131116.

28. Z.H. Xu, B. Hu, and J. Howe, J. Appl. Phys., 103 (2008), no. 043909.

29. Z.H. Xu and B. Hu, Adv. Fund Mater., in press (2008).

30. El. Frankevich et al., Phys. Rev. B, 46 (1992), p. 9320.

31. J. Kalinowski et al., Chem. Phys. Lett., 380 (2003), p. 710.

32. V. Ern and R.E. Merrifield, Phys. Rev. Lett., 21 (1968), p. 609.

33. D.V Konarev et al., Eur. J. Inorg. Chem., 9 (2006). p. 1881.

34. J. Lewis et al., J. Organomet. Chem., 425 (1992), p. 165.

35. G. Kavarnos et al., J. Amer. Chem. Soc., 93 (1971), pp. 1032- 1034.

36. Z.S. Romanova, K. Deshayes, and P. Piotrowiak, J. Amer. Chem. Soc., 123 (2001), pp. 2444-2445.

Zhihua Xu, Huidong Zang, and Bin Hu are with the Department of Materials Science and Engineering, University of Tennessee, Krtoxville, JN 37919, USA. Bin Hu can be reached at [email protected]

Copyright Minerals, Metals & Materials Society Sep 2008

(c) 2008 JOM. Provided by ProQuest LLC. All rights Reserved.