Synthesis of Porphyrin-C₆₀ Dyads with Potential Use in Solar Cells

Miguel Gervaldo, Fernando Fungo, Luis A. Otero, Leonides Sereno, Juana J. Silber and Edgardo N. Durantini*

Departamento de Química y Física, Universidad Nacional de Río Cuarto, Agencia Postal Nro 3, 5800 Río Cuarto, Argentina.
Fax: +58-358-4676233. E-mail adreess: edurantini@exa.unrc.edu.ar (E. N. Durantini).

Received: 20 August 2001 / Uploaded 21 August 2001

Keywords: Porphyrin; Fullerene; Solar energy conversion.

Introduction

One approach to mimicry of photosynthetic energy conversion is the construction of synthetic supramolecular system containing chromophores, electron donors and electron acceptors linked by covalent bonds [1,2]. Many artificial assemblies incorporate porphyrin derivatives as light receptors in solar energy conversion, which takes place in the spectral sensitization of wide band gap semiconductors [3-5]. The energy difference between the conduction band edge of a n-type semiconductor and the oxidation potential of the excited adsorbed dye, provides a driving force for photo-induced charge injection [6]. On the other hand, fullerene C₆₀ derivatives are capable of multiple electron reductions, and as such are good electron acceptors in systems mimicking photosynthetic electron transfer [7,8] Taking in to account the properties of porphyrins and C₆₀, a solid-state double-layer photoelectrochemical cell was developed [9]. In this system, the photoinduced electron transfer from porphyrin to C₆₀ is the primary process of photocurrent generation at the porphyrin-C₆₀ interface. Amongst the covalent functionalizations of C₆₀ [10], the linkage to porphyrins has opened the possibility of constructing artificial photosynthetic systems, in which there are light-induced transfers of electrons or energy from a porphyrin donor to a fullerene acceptor [8]. Therefore, the synthesis of well-defined asymmetric porphyrin derivatives is of great interest for the development of new supramolecular structures [1,2].

In the present work we are interested in the synthesis of porphyrin dyad 9 and 10, in which the porphyrin moiety is attached to C₆₀ by an amide bond. The proposed dyads request the synthesis of meso-substituted porphyrins bearing a phenyl group substituted with the functional group to make the link and three identical substituted phenyl groups (AB₃-porphyrin). Such peripherally asymmetric porphyrins can be prepared by a binary mixed aldehyde condensation. However, this approach is statistical in nature and usually a multiple porphyrin products are obtained [11]. The isolation requires slow and long chromatographic separation. Even so, obtaining the pure porphyrin is not always possible and the yield can be very poor [12,13] On the other hand, porphyrins bearing four different meso-substitutents (ABCD-porphyrin) can be synthesized via a stepwise synthetic approach [14,15] However, for AB₃-porphyrins the stepwise method requires many steps (i.e. about 8 starting from pyrrole and aldehydes). More direct approaches to trans-substituted porphyrins (ABAB-porphyrin) are provided by condensation of dipyrromethane with aldehyde [16]. Similar procedure was used to prepare porphyrin derivatives bearing only one different peripheral phenyl substituent from an appropriated mixture of benzaldehydes and dipyrromethane [12,13].

On the other hand, fullerene C₆₀ can be functionalized to methanofullerene carboxylic acid 8, which is a versatile compound for the preparation of amide derivatives. The amide linkage allows electronic interaction between the p systems of the two chromophores [17]. Thus, the asymmetrical dyads 9 and 10 present mainly two moieties with different electron donor-acceptor properties and they could undergo photoinduced electron transfer [7]. Preliminary studies of spectral sensitization of nanostructed wide band-gap semiconductor (SnO₂) electrodes coated with dyads 9 and 10 showed good light harvesting capacity and incident-photon-to-photocurrent efficiency. Nearly complete quenching of porphyrin fluorescence (both in solution and in adsorbed state over SnO₂ nanestructure) is observed in these dyads, indicating that the photocurrent generation mechanism probably involves the formation of photoinduced intramolecular charge separation state. The results showed that this porphyrin-C₆₀ dyads could be used as material for light energy conversion.

Results and Discussion

Synthesis of porphyrin

The condensation of 4-methoxylbenzylaldehyde with a large excess of pyrrole (1:47 aldehyde/pyrrole mol ratio), catalyzed by trifluoroacetic acid at room temperature, affords meso-(4-methoxylphenyl) dipyrromethane 1 (Scheme 1). Flash chromatography on silica gel, using cyclohexane/ethyl acetate/triethylamine (80/20/1) as eluent, yield 85% of the dipyrromethane 1. The use of mildly basic medium (triethylamine @ 1%) prevents the decomposition of the dipyrromethane on slightly acidic silica column. The dipyrromethane 1 is stable in the purified form upon storage at 0° C in nitrogen atmosphere and absence of light. High purity dipyrromethane is essential for its application in the synthesis of asymmetric meso-substituted porphyrin.

The 5-(4-acetamidophenyl)-10,15,20-tris(4-methoxy phenyl) porphyrin 2 was synthesized by the acid-catalyzed condensation of dipyrromethane 1 and a mixture of 4-methoxybenzaldehyde and 4-acetamidobenzaldehyde in chloroform at room temperature (Scheme 2.a,b). Mixed-benzaldehyde-dipyrromethane condensations were performed using about 2.2:1.4:1 molar relationship of dipyrromethane 1, 4-methylbenzaldehyde and 4-acetamidobenzaldehyde, respectively. This ratio of reactants gives the best yield of the selected porphyrin 2.

The first reaction step (Scheme 2.a) was performed using catalytic among of BF₃^.O(Et)₂ and chloroform as solvent, at room temperature. In the second step (Scheme 2.b), the reaction mixture was subject to oxidation with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ). The desired porphyrin 2 was easily separated by flash chromatography with high purity using dichloromethane/methanol gradient. The first purple band corresponds to meso-tetra substituted porphyrin, the second is the desired amido-porphyrin 2. Thus, porphyrins 2, bearing one 4-acetamidophenyl group and three identical peripheral 4-methoxyphenyl groups, was obtained with appreciable yields of 18%.

Hence, the dipyrromethane method presents two main advantages for the synthesis of AB₃-porphyrins than modified Alder’s method [12], the easier work up and the higher yields obtained. The present procedure involves only two-step one-flask reaction and it may be used for preparation of other similar porphyrin derivatives bearing only one different peripheral phenyl substituent. Amido-porphyrin 2 was hydrolyzed to amino-porphyrin 3 (75%) by heating in tetrahydrofuran (THF)/methanol/KOH medium (Scheme 2.c). Treatment of porphyrins 2 and 3 with zinc acetate in dichloromethane (Scheme 2.d) afforded the corresponding Zn(II)porphyrins 5 (98%) and 4 (96%), respectively.

C₆₀-Acid formation

The precursor 1,2-dihydro-1,2-methanofullerene[60]-61-carboxylic acid 8 (C₆₀-acid) was prepared according to the method described before in the literature (Scheme 3) [18]. Reaction of C₆₀ and (ethoxycarbonyl)methyl diazoacetate 6 yielded initially a mixture of isomers, which were equilibrated to the thermodynamic product methanofullerene ester 7, by heating in toluene. Ester 7 was hydrolyzed to C₆₀-acid 8 by treatment with boron tribromide in benzene (Scheme 3).

Synthesis of Dyad

The porphyrin-C₆₀ dyad 9 was synthesized by a dicyclohexylcarbodiimide-mediated condensation of the C₆₀-acid 8 and the amino-porphyrin 3 (Scheme 4). The reaction was performed in the presence of dicyclohexyl carbodiimide (DCC), 1-hydroxybenzotriazole (BtOH), triethylamine (TEA) and 4-(dimethylamino)pyridine (DMAP) in bromobenzene (PhBr). The reaction mixture was carried out at room temperature for 24 h, to yield after work up the desired porphyrin-C₆₀ dyad 9 with 73% yield. The Zn(II)porphyrin-C60 dyad 10 was obtained in similar procedure using Zn(II)amino-porphyrin 4 (70%).

The asymmetrical dyads 9 and 10 present mainly two moieties with different electron donor-acceptor properties. The electron donor character of the tetrapyrrolic macrocycle was enhanced by the pretense of methoxy groups in para position of the peripheral phenyl rings. This property was further enlarged in dyad 10 by forming metal complex with Zn(II). The increase in the extra electron donor character of the porphyrin was inferred by the oxidation potentials of 2 (0.86 V vs SCE) and Zn(II)porphyrin 5 (0.67 V vs SCE), which are 50 mV and 240 mV easier to oxidize than the corresponding amide-porphyrin, where the methoxy groups have been substituted by methyl groups (0.91 V vs SCE) [4]. Therefore, the structures formed by these porphyrins linked to electron acceptor C₆₀ make them good candidates for photoinduced electron transfer system [8].

Absorption Spectra

The absorption spectra of the dyads and the corresponding porphyrin in toluene are shown in Figure 1. For both dyads the absorption in the UV region is stronger than that of porphyrin due to the presence of C₆₀ moiety, whereas in the visible region the spectra of the dyads are quite similar to the porphyrin monomer. Also, porphyrins absorption show the typical Soret and Q-bands, characteristic of a free-base and the corresponding Zn(II) matalloporphyrin. The absorption maxima of Soret and Q-bands are gathered in Table 1. The spectra of both dyads are essentially a linear combination of the spectra of the corresponding monomers, with only minor differences in wavelength maxima and band shapes (Figure 2). Thus the absorption spectra are consistent with only a weak interaction between the moieties and the two chromophores retain their individual identities.

Figure 1. Absorption spectra in toluene of: (A) amido-porphyrin 2 (black) and dyad 9 (red); (B) Zn(II)amido-porphyrin 5 (black) and dyad 10 (red).
 
 

Figure 2. Absorption spectra in toluene of: (A) amido-porphyrin 2 (black), ester-C60 7 (green), dyad 9 (red) and of a linear combination of the spectra of 2 and 7 (blue); (B) Zn(II)amido-porphyrin 5 (black), ester-C60 7 (green), dyad 10 (red) and of a linear combination of the spectra of 5 and 7 (blue).

Steady-State Fluorescence Spectra

The corrected emission spectra were taken in toluene exciting the sample at 550 nm (Figure 3). The fluorescence emission maxima are shown in Table 1. The fluorescence quantum yields (f_F) of these compounds were calculated by steady state comparative method using tetraphenylporphyrin (TPP) as a reference (Table 1). Dyad 9 and 10 show only very weak emission from the porphyrin moiety, indicating strong quenching of the porphyrin excited singlet state (quenching efficiency, h_q³0.96 and 0.98, respectively) by the attached fullerene structure. As can be see in Figure 3B, dyad 10 shows weak emission with a peak maximum at 720 nm. In THF practically no fluorescence from the C₆₀ was detected when the Q-band was excited (result no show). Fluorescence spectra were also taken exciting dyad 10 at 426 nm, where the absorption in mainly due to the porphyrin moiety (Figure 3B, inset). Under this condition, fluorescence from the C₆₀ moiety was clearly evidenced in toluene, while in THF shows a very weak emission at 720 nm. This indicates that in toluene there is a relaxation pathway from the excited singlet state of the porphyrin to that of the C₆₀. In more polar solvent, another competitive electro transfer pathway from the porphyrin to the C₆₀ could be occurring to produce the ion pair state [19]. The detection of this band at 720 nm is not possible to observe in dyad 9 due to the porphyrin moiety fluorescence in the same spectral region.

Figure 3. Fluorescence emission spectra in toluene (l_exc=550 nm) of: (A) amido-porphyrin 2 (black) and dyad 9 (red); (B) Zn(II)amido-porphyrin 5 (black) and dyad 10 (red); inset (l_exc=426 nm): dyad 10 in toluene (black) and THF (red).

Photoelectrochemistry

Preliminary results about photoelectrochemical properties of dyads 9 and 10 adsorbed over nanoestructured semiconductor SnO₂ electrode demonstrated that the SnO₂ modified film exhibit photoresponse in the visible region with good light harvesting capacity (Figure 4) [20,21]. The incident-photon-to-photocurrent efficiency (IPCE) was obtained through the measure of the generated photocurrent when the electrodes are illuminated with monochromatic light, by using eqn. 1,

where isc is the short circuit photocurrent (A cm^-2), Iinc is the incident light intensity (W cm^-2), and l is the excitation wavelength (nm) [4]. The photocurrent action spectrum closely matches the absorption spectrum of dyads (Figure 4). The photo-induced charge separation efficiencies, with maximum IPCE around 10 and 3 % were found for dyads 9 and 10, respectively, at Soret band (dye absorbance 0.228 and 0.112). The charge injection yield (F_inj) from the excited dyad to the semiconductor times charge collection efficiency (h_c) of the systems at the Soret bands were calculated from IPCE and light harvesting efficiency (LHE=1-10^-A) of the dyad by using eqn. 2 [4].

The product F_inj h_c yields 0.24 and 0.13 for dyads 9 and 10, respectively. Zn(II)porphyrin-C₆₀ dyad 10 gives lower photocurrent efficiency than the unmetallized dyad 9. On the other hand, the product F_inj h_c in dyad 10 is very similar to that of the Zn(II)porphyrin 5 (0.11). Thus, a possible reason for the low photocurrent efficiency in the dyad 10 is the contribution of Zn(II)porphyrin moiety to the back electron-transfer process [4,5].

The fact that there is a close correspondence between the photocurrent action spectrum and the absorption spectrum of the electrodes, confirms that light absorption by the porphyrin-C₆₀ dyad 9 and 10 is the initial step in the generation of photo-induced charge transfer mechanism, like in the natural photosynthetic apparatus [3]. Moreover, Figure 4 shows that anodic photocurrent is generated in spectral region where C₆₀ is the mainly responsible of the light absorption, i.e. bellow 380 nm. While in control experiments, ITO/SnO₂/C₆₀ electrodes do not produce appreciable anodic photocurrent [22], which suggest that in dyad 9 and 10 the charge transfer process involves the porphyrin moiety. On the other hand, ITO/SnO₂/dyad electrodes are non-fluorescent and produce comparable or higher photoelectric effects (when are illuminated in the spectral region where porphyrin is exited) than the fluorescent SnO₂ electrodes modified by the porphyrin moiety. This indicates that a different mechanism than the direct electron injection from the excited porphyrin to SnO₂ nanoparticle is involved, probably through a photoinduced intramolecular charge separation state.

Figure 4. Photocurrent action spectrum of electrode: (A) ITO/SnO₂/dyad 9; (B) ITO/SnO₂/dyad 10.

Conclusion

Amido-porphyrin 2, bearing one 4-acetamidophenyl group and three identical peripheral 4-methoxyphenyl groups, was obtained with appreciable yields of 18% by condensation of dipyrromethane 1 with appropriate benzaldehydes. Porphyrin 2 was hydrolyzed to amine-porphyrin 3. Both 2 and 3 were treated with zinc acetate to form the corresponding metal complexes 4 and 5 respectively. The free-base 2 and Zn(II)porphyrin 4 were linked to C₆₀-acid 8 by DCC-mediated amidation reaction to yield dyads 9 and 10. These dyads contain two structural moieties with different electron donor-acceptor proprieties, a good electron acceptor C₆₀ structure and a peripherally substituted porphyrin by electron donor methoxy groups. Also, in dyad 10 the donor capacity of the porphyrin moiety was enhanced forming complex with Zn(II). Thus, these structures have good chance for observing intramolecular photoinduced electron transfer [8,19]. Thermodynamically, fluorescence quenching of the porphyrin moiety could occur by photoinduced electron transfer from porphyrin to the C₆₀ structure to yield a charge-separated species. The energies of the 0-0 transition between S₁ and the S₀ state were determined by averaging the energies of the corresponding (0,0) peaks in the fluorescence and the absorption bands. Values of 1.89 eV (toluene), 2.07 eV (toluene and 2.05 eV in THF) and 1.73 eV (toluene and THF) were estimated for the energy level of the locally excited singlet state of the porphyrin (¹P*-C₆₀ and ¹ZnP*-C₆₀) and C₆₀ (P-C₆₀* or ZnP-C₆₀*) in dyads 9 and 10, respectively. On the other hand, energies of 1.43 and 1.24 eV for the charge separation state of dyads 9 (P⁺-C₆₀^-) and 10 (ZnP⁺-C₆₀^-) were calculated considering the first oxidation potential of the porphyrin and the first reduction potential of the C₆₀ (-0.57 V) in 1,2-dichloroethane. These results suggest that depending on the solvent polarity, the energy levels of the ion pair state in porphyrin-C₆₀ linked system vary. Apparently, in polar solvent, such as THF, the charge separation occurs from both the excited singlet state of the porphyrin and the C₆₀ moiety. In toluene, charge separation also can occur from the excited singlet state of the porphyrin to the C₆₀, however in this case the energy of ion pair state can be similar that the locally excited singlet state of C₆₀ [8]. The fact that in the generation of photoelectrical effects the dyad 10 is less effective in comparison with dyad 9, despite that the driving force for charge injection is higher in the metallized one, is explained considering that the metallized porphyrin enhances the back electron-transfer process, producing low photocurrent yield. Further studies concerning the mechanism of photocurrent generation are presently in progress in our laboratory.

Acknowledgments. Authors thank Consejo Nacional de Investigaciones Científicas y Técnicas of Argentina, Agencia Nacional de Promoción Científica y Técnica, Fundación Antorchas and SECYT de la Universidad Nacional de Río Cuarto, for financial support. L.A.O., J.J.S. and E.N.D. are Scientific Members of CONICET.

References

1. D. Gust, T. A. Moore Topics in Current Chemistry 1991, 159, 103.

2. M. R. Wasielewski Chem. Rev. 1992, 92, 435.

3. L. Otero, H. Osora, W. Li, M. A. Fox J. Porphyrins Phthalocyanines 1998, 2, 123.

4. F. Fungo, L. Otero, E. N. Durantini, J. J. Silber, L. Sereno J. Phys. Chem. B 2000, 104, 7644.

5. R. B. M. Koehorst, G. K. Boschloo, T. J. Savenije, A. Goossens, T. J. Schaafsma J. Phys. Chem. 2000, 104, 2371.

6. B. O’Regan, M. Gratzel Nature 1990, 253, 737.

7. P. A. Liddel, J. P. Sumida, A. N. Macpherson, A. L. Moore, T. A. Moore, D. Gust Photochem. Photobiol. 1994, 60, 537.

8. H. Imahori, Y. Sakata Eur. J. Org. Chem. 1999, 2445.

9. K. Takahashi, K. Etoh, Y. Tsuda, T. Yamaguchi, T. Komura, S. Ito, K. J. Muruta Electroanal. Chem. 1997, 426, 85.

10. F. Diederich, L. Isaacs, D. Philp Chem. Soc. Rev. 1994, 243.

11. S. J. Lindsey, Y. C. Schreiman, H. C. Hsu, P. C. Kearney and A. M. Marguerettaz, J. Org. Chem., 1987, 52, 827.

12. E. N. Durantini, J. J Silber Synth. Commun. 1999, 29, 3353.

13. E. N. Durantini Molecules 2001, 6, 533.

14. D. M. Wallace, S. H. Leung, M. O. Senge and K. M. Smith, J. Org. Chem., 1993, 58, 7245.

15. C-H. Lee, F. Li, K. Iwamoto, J. Dadok, A. A. Bothner-By and J. S. Lindsey, Tetrahedron, 1995, 51, 11645.

16. C-H. Lee and J. S. Lindsey, Tetrahedron, 1994, 50, 11427.

17. F. Fungo, L. A. Otero, L. Sereno, J. J. Silber, E. N. Durantini, J. Mater. Chem. 2000, 10, 645.

18. L. Isaacs, F. Diederich Helv. Chim. Acta 1993, 76, 2454.

19. H. Imahori, K. Hagiwara, M. Aoki, T. Akiyama, S. Taniguchi, T. Okada, M. Shirakawa, Y. Sakata J. Am. Chem. Soc. 1996, 118, 11771.

20. P. Kamat, I. Bedja, S. Hotchandani, C. Patterson J. Phys. Chem. 1996, 100, 4900.

21. W. Li, H. Osora, L. Otero, D. C. Duncan, M. A. Fox J. Phys. Chem. A 1998, 102, 5333.

22. P. V. Kamat, M. Gevaret, K. Vinodgopal J. Phys. Chem. B 1997, 101, 4422.