A Claisen rearrangement route to novel electron-rich perylene dyes

A combined tandem Friedel-Crafts annulation/Claisen rearrangement strategy was developed for the preparation of tetrasubstituted perylenes. Diallyloxyoctahydroperylene was prepared from commercially available 1,2,3,4-tetrahydronaphth-1,5-diol and oxidized with chloranil to the perylene analogue. Claisen rearrangement followed to the novel 2,8-diallylperylene-3,9-diol, which was acylated in situ to the corresponding dioctanoyl ester. These results show that allyloxyperylene undergoes Claisen rearrangement efficiently, while partially hydrogenated analogue generates multiple products under identical conditions.


Introduction
Emerging dye applications in photovoltaics and photonics are prompting vigorous effort to increase the accessibility of novel perylene analogues. 1Perylenes bearing electron-donating substitutents are attractive as organic dyes for solar cells due to predicted stability of the radical cations generated upon photoelectron emission. 2Most novel perylenes are prepared via functionalization of commercially available perylene diimides, 3 which in turn are obtained upon oxidative coupling of the corresponding naphthalene imides.Stepwise coupling of halonaphthalenes followed by oxidative annulation has also been reported but is less common. 4ur laboratory recently disclosed a novel route to 3,9-dialkoxyperylenes via Tandem Friedel-Crafts Annulation (TFCA) of tetralin analogues (Scheme 1). 5 Herein, we report an extension of this approach for the preparation of novel, dialkoxyperylenes that bear allyl groups via Claisen rearrangement.The target products bear electron-donating allyl groups, which also provide sites for attachment in novel photonic materials. 6cheme 1. Tandem Friedel-Crafts annulation.
Prior studies demonstrated that 1,2,3,4-tetrahydronaphth-1,5-diol (1a: R = H) and corresponding ethers (1b: R = CH3 and 1c: R = (CH2)7CH3) are suitable substrates for TFCA, each generating the corresponding octahydroperylene (2) in modest to good yields. 5Allyl ether 1d (R = allyl), 7 also produced 2d upon TFCA.Aryl esters of 1 (for example, R = COCH3 or CO(CH2)6CH3), however, were resistant to TFCA, presumably due to the reduction of both aryl nucleophilicity and carbocation intermediate stability, limiting the scope of this transformation.We envisaged a modified route to additional dyes in which key intermediate 2d (R = allyl) is transformed to 5 via sequential Claisen rearrangement and oxidation steps (Scheme 2).Novel product 5a and its ether (R = alkyl) derivatives are predicted to be easily elaborated to perylenes that are electronically activated toward photoemission. 8Conversion from known 2d to 5b was planned via Claisen rearrangement followed by oxidation (Route A) or the reverse sequence (Route B).The former requires the preparation of acylated octahydroperylene intermediate 4b en route to 5b to avoid phenolic oxidation to quinones. 5,8In the present work, we examined feasibility of both synthetic avenues to 5. Scheme 2. Proposed routes to 2,8-diallyl-3,9-diacyloxyperylene.
Efforts to implement Route A commenced upon examination of the thermal Claisen rearrangement of 2d to 4a under conditions established in this laboratory for the rearrangement of allyloxybenzene.Thus, solvent-free heating of solid 2d (190 °C, 10 h) gave a black solid that exhibited loss of 1 H NMR signal at 4.8 ppm (allyloxy methylene of 2d) and concurrent emergence of a complex multiplet between 3.0-3.5 ppm, both suggesting Claisen rearrangement.However, multiple doublets in the 7.8-8.0ppm region of 1 H NMR were consistent with oxidation of the polycyclic core to the corresponding anthracenyl or perylenyl products.The spectrum was further complicated by the presence of multiple vinyl signals indicating multiple products.Because each of these products consists of similar Rf value in TLC, no further studies were conducted on this pathway to avoid difficult purification steps.
To investigate the feasibility of Route B, chloranil oxidation of 2d (4 equiv, toluene, room temperature) was examined.This transformation and others producing analogues with fully aromatized perylene cores were protected from ambient light to avoid photodegradation.An aliquot of the incomplete reaction was strongly fluorescent and exhibited new signals at 7.8 to 8.0 ppm in 1 H NMR, suggesting the partial oxidation of 2d to the anthracenyl derivative.Continued exposure of the sample to chloranil (60 h total) induced clean conversion to 3d (Scheme 2), as indicated by the expected aromatic resonances given in 1 H NMR spectrum.The product, which precipitated in the reaction milieu, was filtered, washed with acetonitrile/MeOH (1:1), and dried under vacuum to give a greenish yellow solid in good yield (72%).Product 3d was poorly soluble in available NMR solvents (CDCl 3 , acetone-d 6 , acetonitrile-d 3 , DMSO-d 6 , THF-d8, benzene-d6), precluding the collection of 13 C NMR data. 5hermal Claisen rearrangement of 3d under solvent-free conditions (150 °C; 4 h) gave a darkened product that was shown to be mainly unreacted substrate by 1 H NMR. At higher temperature (180 °C) the product was extensively decomposed.Since the Claisen rearrangement of structurally similar 1-allyloxynaphthalene proceeds at 150 °C (2 h, 98% conversion), the failure of the rearrangement of 3d at this temperature likely results from inhibition caused by the solid physical state of the perylene core. 9Thermal rearrangement of 3d in 1,2-dichlorobenzene at 150 °C gave degraded product as determined by 1 H NMR. Heating 3d in a minimum amount of polyethylene glycol dimethyl ether (2 h, 180 °C) gave targeted product 5a contaminated with unidentified byproducts.Several attempts made to produce clean 5a were unsuccessful, possibly due to facile product oxidation.To facilitate isolation and purification of a derivative of 5, the Claisen rearrangement was followed by acylation in situ (ClCO(CH2)6CH3, pyridine, 0 °C to rt), which gave 5b in good yield after chromatography (56% yield).Product 5b exhibited the expected spectroscopic signals including UV/vis profile with the typical perylene absorbance pattern. 8hese findings establish a pathway for novel, electron-rich diallyldihydroxyperylene and its diester. 10The products support potential sites for ligation or chemical modification en route to novel photoactive materials, which are under investigation in our laboratory and those of collaborators.

Conclusions
Tetrasubstituted, electron-rich perylene analogues 5a and 5b were prepared in 4 and 5 steps, respectively from commercially available 1,2,3,4-tetrahydronaphth-1,5-diol in good overall efficiency.The present sequence generates electron rich perylene cores that support olefinic and alkoxy groups, which may be employed for subsequent tethering to materials for photonic applications.

Experimental Section
General.Melting points were measured on a Mel-Temp apparatus.Proton NMR and 13 C NMR spectra were recorded at 500 MHz and 125 MHz (Varian INOVA), respectively, in CDCl3 at 25 °C unless otherwise noted.Chemical shift values are reported in ppm with TMS as reference at 0.00 ppm for proton and CDCl3 carbon residue at 77.0 ppm for 13 C spectra.Spin-spin coupling constants (J) are given in Hz.Mass (Finnigan LCQ DUO) and FT-IR (Bruker Vector 22) spectra were recorded using APCI and ATR technology, respectively, unless otherwise noted.Kieselgel 60F254 silica gel TLC plates were used with ethyl acetate and hexanes as solvents for monitoring reaction progress.All necessary chemicals were purchased in 98% or better purities and used without purification.