Fully Conjugated Porphyrin Tapes with Electronic Absorption Bands That Reach into Infrared

See allHide authors and affiliations

Science  06 Jul 2001:
Vol. 293, Issue 5527, pp. 79-82
DOI: 10.1126/science.1059552


Scandium(III)-catalyzed oxidation ofmeso-meso–linked zinc(II)-porphyrin arrays (up to dodecamers) with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) led to efficient formation of triply meso-meso–, β-β–, and β-β–linked zinc(II)-oligoiporphyrins with 62 to 91% yields. These fused tape-shaped porphyrin arrays display extremely red-shifted absorption bands that reflect extensively π-conjugated electronic systems and a low excitation gap. The lowest electronic absorption bands become increasingly intensified and red-shifted upon the increase in the number of porphyrins and eventually reach a peak electronic excitation for the dodecamer at ∼3500 wavenumber. The one-electron oxidation potentials also decreased progressively upon the increase in the number of porphyrins. These properties in long and rigid molecular shapes suggest their potential use as molecular wires.

Discrete molecules with a very long π-system are of interest as organic conducting materials, near-infrared (near-IR) dyes, nonlinear optical materials, and molecular wires (1–3). Numerous attempts that have been made to extend the π-electronic systems have, however, encountered serious problems, such as synthetic inaccessibility, chemical instability, poor solubility, and conjugation saturation behavior that arises through the effective conjugated length (ECL) effect. The ECL defines the extent of π-conjugated systems in which the electronic delocalization is limited and at which point the optical, electrochemical, and other physical properties reach a saturation level that is common with the analogous polymer (1). A straightforward strategy for maximizing π-overlap may be to hold the π-systems coplanar within a tapelike framework by fusing the units edge-to-edge, to make a covalently linked long, flat array, but this goal is synthetically quite demanding. Fused π-conjugated systems are promising also in circumventing the above ECL limit, as seen for the [n]acene series (n = 1 to 7) (1, 4), but extension to the higher conjugated systems suffers from problems of poor solubility caused by the resulting planar structures. Within a confined pigment number, charged dyes such as oxonols and cyanines can escape the ECL effect because of the absence of the bond alternation arising from effective resonance (5). Again, extension to the higher homologs is difficult to achieve and reveals the ECL effect (6).

Porphyrins are intriguing building units from which to construct large π-conjugated molecules. Two types of conjugated porphyrins have been developed, meso-ethyne–bridged andmeso-butadiyne–bridged porphyrin arrays (7–9) and fused porphyrin arrays (10–13), both of which show unusual properties that result from strong π-conjugation. Here we report the synthesis of highly conjugated porphyrin arrays, in which the electronic π-conjugation over the arrays is far stronger than the π-conjugation of these precedents, as seen from extremely low-energy IR electronic excitations.

Recently, we reported the synthesis ofmeso-meso–linked porphyrin arrays of up to 128-oligomers by AgI salt–promoted coupling reaction (14). This extremely long, discrete, rodlike organic molecule has a molecular length of about 108 nm. These arrays adopt a nearly orthogonal conformation that tends to minimize the electronic interaction between the neighboring porphyrins (14,15). The aryl-end–cappedmeso-meso–linked CuII-diporphyrin 1can be converted to triply linked fused diporphyrin 2by the oxidative double-ring closure (ODRC) reaction upon treatment with (p-BrC6H4)3NSbCl6in C6F6 (Scheme 1) (13). The planar structure of 2 has been revealed by x-ray analysis, and full conjugation over the two porphyrins has been demonstrated by its substantially broadened and red-shifted absorption spectrum.

Figure 1

Ultraviolet-visible-infrared absorption spectra of porphyrin 5 and meso-meso–linked porphyrin arrays 6 to 12 (top) and triply linked fused porphyrin arrays 13 to 19 (bottom) taken in CHCl3 at room temperature. The absorption spectra of5 to 12 and 13 to 19 were normalized at 23,700 to 24,000 cm−1 and 23,800 to 24,600 cm−1, respectively. The background absorbance at ∼6000, ∼4000, and ∼3500 cm−1 may arise from the overtones of C-H vibration of the solvent.

Figure 2

Plot of absorption peak of bands III versus the numbers of porphyrins in the triply linked fused arrays (N). The line shown is represented by the equation λmax = 174.5N + 793 (nm) with coefficient of variationr 2 = 0.995.

Figure 3

IR spectra of molecules 13to 19 taken in a KBr pellet at room temperature with a JASCO-FT-IR 420 spectrometer at room temperature.

Figure 4

Plot of the first oxidation potentials [E ox (mV) versus AgClO4/Ag taken in CHCl3] versus N −1. The first oxidation potentials (E ox) were obtained from cyclic voltammetry as follows: 13, 212;14, 14; 15, −89; 16, −147;17, −180; and 18, −241 mV. The oxidation potential of 19 could not be determined owing to its poor solubility. The line shown is represented by the equationE ox = 1197N −1 − 386 (mV) with correlation coefficientr 2 = 0.999.

Scheme 1

1, M = Cu, R1 = R2 = Ar1; 3, M = Zn, R1 = R2 = Ar1;6, M = Zn, R1 = Ar2, R2 = Ar3; 2, M = Cu, R1 = R2 = Ar1;4, M = Zn, R1 = R2 = Ar1; 13, M = Zn, R1 = Ar2, R2 = Ar3. Number of porphyrins (N): 7 (1), 8 (2),9 (3), 10 (4), 11 (6), 12(10); 14 (1), 15 (2), 16 (3),17 (4), 18 (6), and 19 (10).

We now describe a highly efficient synthetic method that allows the ODRC reaction of higher meso-meso–linked Zn(II)-porphyrins in good yields. The ODRC reaction was conducted simply by refluxing a toluene solution of meso-meso–linked Zn(II)-diporphyrin3 in the presence of five equivalents of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) and scandium trifluoromethanesulfonate [Sc(OTf)3] for 0.5 hour, which afforded the triply linked fused diporphyrin 4in 86% yield as a sole product. DDQ or Sc(OTf)3alone did not effect any change of 3. Under similar conditions, the Zn(II)-porphyrin monomer 5 was also effectively coupled to give triply linked diporphyrin 4 in an almost quantitative yield. The undesirable chlorination at the peripheral β positions, a serious problem in the ODRC reaction with (p-BrC6H4)3NSbCl6(13), was completely suppressed. More important, the ODRC reaction was successfully applied to the synthesis of higher fused porphyrin arrays as follows: 13 (dimer, 75%) from6; 14 (trimer, 91%) from 7;15 (tetramer, 78%) from 8; 16(pentamer, 77%) from 9; 17 (hexamer, 85%) from 10; 18 (octamer, 60%) from11; and 19 (dodecamer, 73%) from 12(16). Scheme 2 shows the ODRC reaction of12 to 19, in which as many as 22 β-β bonds are formed in a one-pot reaction. The largest fused array, 19, provides spacious coplanar π-electronic area with about 100 Å by 7 Å.

Scheme 2

The electronic absorption spectra of the porphyrin 5, the meso-meso–linked porphyrin arrays (6 to12), and triply linked porphyrin arrays (13 to19) taken in CHCl3 are shown in Fig. 1. The absorption spectra of themeso-meso–linked porphyrin arrays show split Soret bands (B-bands), which can be qualitatively accounted for in terms of the exciton coupling that is originating from the Coulombic interactions between the transition dipole moments (14). These results indicate that electronic π-conjugation is disrupted in the array, probably as a result of the almost perpendicular conformation of the neighboring porphyrins.

In sharp contrast, the triply linked fused porphyrin arrays display drastically red-shifted absorption spectra that reflect extensive π-conjugation over the array. The absorption spectra of the triply linked porphyrin arrays exhibit three distinct bands in CHCl3 solutions as in the meso-meso–linked porphyrin arrays but at entirely different positions (designated as bands I, II, and III in Fig. 1). Although bands I are observed at 23,800 to 24,600 cm−1—nearly the same positions as that of the monomeric Zn(II)-porphyrin 5—bands II and III are markedly red-shifted upon the increase in the number of porphyrins (17). A degree of saturation behavior is noted for the red-shift of bands II, whereas bands III exhibit a progressive red-shift of their peak positions upon the increase in the number of porphyrins as follows: 13 (dimer), 9400; 14(trimer), 7500; 15 (tetramer), 6600; 16(pentamer), 6000; 17 (hexamer), 5400; 18(octamer), 4500; and 19 (dodecamer), 3500 cm−1. Thus, bands III of 17, 18, and19 are located in the IR region with their tails extending beyond 3000 cm−1. A plot of the absorption peak of band III in wavelength units (λmax) versus the number of porphyrins (N) revealed fairly linear behavior up to19 without an indication of the ECL effect (Fig. 2). These results suggest a further possible decrease in the excitation energy of band III upon elongation of the array.

Strong electronic transitions enter the IR frequency region with their tails reaching ∼1500 cm−1 for 17,18, and 19 in solid states (Fig. 3). The relative amplitudes of the vibrational IR bands appear to decrease compared with the amplitude of the electronic absorption band (especially for the C-H stretching bands around 2900 cm−1), probably because of the increasing relative intensity of the electronic transition upon the increase in the number of porphyrins. Detailed examination of the low-energy electronic absorption should provide important information on the vibronic coupling of organic molecules.

The first one-electron oxidation potential of the fused diporphyrin4 is 0.11 V versus AgClO4/Ag in CHCl3 solution (13), which is considerably lower than the values for the parent porphyrin monomer 5 (0.52 V) and meso-meso–linked diporphyrin 3(0.47 V). Thus, the expansion of π-electronic system tends to raise the highest occupied molecular orbital (HOMO) orbital. This trend is increasingly enhanced upon the increase in the number of porphyrins as shown in Fig. 4, where a plot of the one-electron oxidation potentials versus N −1exhibits good linearity. The intercept (−0.38 V) of the plot may be assigned for the one-electron oxidation potential of the “infinite” fused porphyrin polymer.

In spite of their remarkable properties, including extremely low HOMO-LUMO (highest occupied/lowest unoccupied molecular orbital) gaps, low one-electron oxidation potentials, and rigid and giant molecular size, the triply linked arrays are stable in air and easily manipulated, and should prove useful for studies on the vibronic coupling of organic molecules. They may also find use as conducting molecular wires and in molecular-scale electronic devices. In addition, the unusual photophysical and electrochemical properties of these fused porphyrin arrays suggest potential avenues for further investigation, including the development of near-IR and IR sensors and dyes, and materials for nonlinear optics and spin ordering.


Stay Connected to Science

Navigate This Article