Synthesis and photochemistry of two quinoline analogs of the perimidinespirohexadienone family of photochromes

https://doi.org/10.1016/j.jphotochem.2009.04.011Get rights and content

Abstract

We report the detailed synthesis and photochemistry of two analogs (specifically 3,5-di-tert-butyl-7′-methyl- and 3,5-di-tert-butyl-7′,9′-dimethyl-1′,3′-dihydrospirocyclohexa[2,5]diene-1,2′-pyrido[4,3,2-de]quinazolin-4-one) of the perimidinespirohexadienone (3,5-di-tert-butyl-1′,3′-dihydrospirocyclohexa[2,5]diene-1,2′-perimidin-4-one) family of photochromes in which the naphthalene moiety of the parent is replaced by a quinoline, and compare them to the parent compound. Molar absorptivities of both the short wavelength spirocyclic isomer (SW) and long wavelength quinonimine isomer (LW) of each were determined by a combination of proton NMR and UV–vis spectroscopy in solvents of varying polarity. Quantum yield measurements for photoisomerization of SW to LW are reported in those same solvents, with qualitative extrapolation to additional solvents. The position and rate of the thermal equilibrium reverting LW to SW is estimated for these compounds. The 9′-methyl in SW (6-methyl in LW) is found to be essential for complete reversion of LW to SW in the dark. Finally one-dimensional NOE NMR spectroscopy was used to conclusively determine the structure of LW for the quinoline analogs as the 4-(5-aminoquinolin-4-ylimino)-2,6-di-tert-butylcyclohexa-2,5-dienone resulting from opening toward the quinoline nitrogen, rather than the 4-(4-aminoquinolin-5-ylimino) structure that would result from spirocyclic ring opening away from the quinoline nitrogen which had been initially proposed by Minkin et al. for very similar compounds [V.I. Minkin, V.N. Komissarov, V.A. Kharlanov, Perimidinespirocyclohexadienones, in: J.C. Crano, R.J. Guglielmetti (Eds.), Organic Photochromic and Thermochromic Compounds, vol. 1, Plenum Press, New York, 1999, pp. 315–340, and references therein].

Introduction

We have recently been investigating analogs of the perimidinespirohexadienone photochrome family studied almost exclusively over the past 20 years by Minkin and co-workers (Scheme 1) [1], [2], [3], [4], [5]. We are ultimately interested in designing a new class of “gateable” intermolecular photooxidants based on photochromic rearrangement of a non-photooxidizing short wavelength isomer (SW, a “pro-photooxidant”) to a potentially photooxidizing long wavelength isomer (LW). The perimidinespirohexadienones are promising in this regard, as they meet several necessary criteria: photochromic coloration of SW to LW with UV or short wavelength visible light; purely thermal reversion of LW to SW (leaving LW's photochemical “channel” available for intermolecular photoinduced charge transfer); distinct and non-interfering absorbance bands in SW and LW; sufficiently slow thermal fade (LW  SW) to make use of an appreciable concentration of LW; and a large difference in ground state reduction potential (Ered0) between LW and SW (with LW more reducible). With respect to this last point, the electrochemical data available for 1b/2b already indicate that the difference in Ered0 between LW and SW is similar to or slightly greater than the difference in excitation energy between SW and LW. The flexibility to modify the naphthalene moiety of 1 without substantially impacting the photochromic rearrangement nor the reduction potential of 1 (assuming the dienone moiety is the likely electrophore in 1, as surmised by Minkin and co-workers) has led us to investigate more electron deficient replacements for the naphthalene moiety.

Minkin and co-workers have reported some investigations of the corresponding analogs where a quinoline replaces the naphthalene in the parent system (compounds 3ce), dubbing these analogs of the perimidinespirohexadienones to be quinazolinespirohexadienones (Scheme 2) [2], [5]. However, unable to devise a suitable synthesis to the necessary aminochloroquinoline precursor with the same substitution pattern, we opted to forego the 5′-methyl at R3 and devised our own synthesis to prepare compounds 3a,b (analogous to Minkin's 3c,d) from 2-methyl- and 2,4-dimethylaniline, respectively, based in large part on the work of Siim et al. [6]. In this work, we report the synthesis and detailed study of the photochromic rearrangement of 3a,b, including a conclusive study of the structure of the photochemically generated LW of these compounds. A separate report on the experimentally determined reduction potentials (in comparison to those predicted computationally based on our recently published methods [7]) and on the nature of the electrogenerated LW isomer will be forthcoming.

Section snippets

Instrumentation

UV–vis spectroscopy was performed on argon-purged 3 mL or 4 mL solutions in standard 1 cm quartz cuvettes using an Agilent 8453 diode-array spectrophotometer.

Photochemical irradiations were performed on argon-purged 3 mL or 4 mL solutions in standard 1 cm quartz cuvettes (NSG) held in a Newport Oriel 13950 cuvette holder, approximately 30 cm from the source. The source used was a 350 W Newport 6286 mercury arc lamp in a Newport 66942 research arc lamp housing powered by a Newport 69910 power supply.

Results and discussion

The syntheses reported above for 3a,b were effective, ultimately yielding gram quantities of these compounds for structural and photochemical study, in an overall yield of about 30% over nine chemical steps.

First we studied the photochromic rearrangement in several solvents of varying polarity to obtain the longest wavelength λmax of each compound in its SW and LW isomer. As Minkin and co-workers have noted for similar compounds, solutions of 3a contain a substantial amount of the LW isomer

Conclusions and future work

While we have unfortunately not found a suitable explanation for the differences in thermal reversion of 12a  3a vs. 12b  3b, we have conclusively determined the structure of the photolytically generated LW isomers of 3a,b as 12a,b (rather than 4a,b), respectively, by the observed NOE enhancements. We note that this implies, but does not conclusively demonstrate, analogous structures 12ce as the likely LW isomers generated by photolysis of 3ce. (As far as we are aware, Minkin and co-workers

Acknowledgements

This work was supported by the Camille & Henry Dreyfus Foundation Start-up Award Program, a Research Corporation Cottrell College Science Award, and start-up funding from the Hope College Department of Chemistry and Division of Natural & Applied Sciences. Additionally, several student coauthors acknowledge additional individual support from the Hope College HHMI Research Scholars (JPM) and HHMI Computational Science & Modeling Scholars (ALS) programs, the Lilly Foundation Summer Undergraduate

References (15)

  • V.A. Kharlanov

    J. Photochem. Photobiol. A

    (1999)
  • S. Um et al.

    Dyes Pigm.

    (2005)
  • V.N. Komissarov et al.

    Doklady Akad. Nauk SSSR

    (1988)
  • V.I. Minkin et al.

    Perimidinespirocyclohexadienones

  • V.A. Kharlanov

    Russ. J. Org. Chem.

    (2001)
  • V.I. Minkin

    Chem. Rev.

    (2004)
  • B.G. Siim et al.

    J. Med. Chem.

    (1997)
There are more references available in the full text version of this article.

Cited by (7)

View all citing articles on Scopus
View full text