Stabilization of the merocyanine form of photochromic compounds in fluoro alcohols is due to a hydrogen bond

摘要

N H N H H O N H N HH O O H O H 1' ( a) ( b) N H N H N H O N H N HH O 2' ( a) ( b) Stabilization of the merocyanine form of photochromic compounds in fluoro alcohols is due to a hydrogen bond Takayuki Suzuki, Fu-Tyan Lin, Satyam Priyadashy and Stephen G. Weber* Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania 5260, USA. E-mail: sweber+@pitt.edu Received (in Columbia, MO, USA) 10th August 1998, Accepted 3rd November 1998 Fluoroalcohols 1,1,1,3,3,3-hexafluoropropan-2-ol (HFP), 2,2,2-trifluoroethanol (TFE) and 2-fluoroethanol (FE), acting as Lewis acids, stabilize the p-conjugated, colored merocyanine forms of spiropyran and spirooxazine photochromic compounds as metal ions do.We1 and others2 have been interested in the interactions of metal ions with photochromic compounds such as spiropyrans, spironaphthoxazines, and chromenes for potential applications in optical switching, memory and sensors. In these photochromic compounds, light is used to cleave a single Cndash;O bond in the pyran or oxazine ring (so-called closed form) which results in the creation of a relatively more polar species (socalled open or merocyanine form).Metals influence this process by associating with this now electron-rich oxygen atom in the open form. It has been reported that 1,1,1,3,3,3-hexafluoropropan- 2-ol (HFP) stabilizes the merocyanine form of polymer- bound nitrospiropyran through a general effect of the solventrsquo;s lsquo;polarityrsquo;.3 We wondered if HFP and other fluoro alcohols4 known as good H-bond donors could stabilize the open form of spirooxazines in particular, and other similar photochromics in general, and if so, do they work as the metal ions do, through a specific Lewis acid/Lewis base interaction? Compound 1, 1,3-dihydro-5-methoxy-1,3,3-trimethylspiro- 2H-indole-2,3A-3Hnaphtho2,1-b1,4oxazine is purple at room temperature at equilibrium without photolysis in HFP-d (2 atom 1H on the hydroxy group). 1H NMR integration shows that about 50 of 1 exists in the open form(s) 1A. 1H ROESY spectra for these solutions have cross peaks between the solvent hydroxy proton and the N-methyl protons in 1. There is also a cross peak between the solvent hydroxy proton and the methoxy methyl in the merocyanine form of the molecule 1A. There is evidence for both so-called TTC5 and TTT isomers of 1A; cross peaks are indicated as arrows in Fig. 1. Compound 2 (1,3,3-trimethyl 5,6-dimethylspiro2H-indole- 2,3A-3Hpyrido3,2-f1,4benzoxazine6) also opens to the merocyanine form(s) 2A in fluoro alcohols.Compound 2A comprises 70 mol in HFP, 15 mol in TFE, 9 mol in FE, and 2 mol in EtOH in the dark. A TFE-d3 (5 atom 1H) solution containing 2 is also purple at equilibrium at room temperature, although it is only about 15 open form. A signal is observed between the hydroxy proton of TFE and the 2Aproton of 2A (ring proton ortho to the oxygen) in the 1H NOESY spectrum.Furthermore, the resonance of the hydroxy proton of TFE is deshielded (d 5.4) in comparison to pure TFE (d 5.2) as shown in Fig. 2. When the solution is irradiated with UV (300ndash;400 nm) light, the resonance becomes more deshielded (Fig. 2). We infer that TFE interacts with 2A via an H-bond to the oxygen with rapid exchange on the NMR time scale (500 MHz). Additionally, we note a signal between the N-methyl protons at the indole group and the imine proton for 2A in the 1H NOESY spectrum, while there are no signals between the imine proton and the geminal methyl protons, nor are there signals between the imine proton and the 2A-proton; thus, for 2A, TTC is the only isomer in TFE.In a solution of 2 in HFP, which is 70 ringopened, there are several cross peaks (1H ROESY) indicating the presence of both the TTC and TTT isomers. The cross peaks are illustrated schematically in Fig. 3. There is also a cross peak between the solvent hydroxy proton and the open-form Nmethyl protons, once again indicating a possible H-bond to the former oxazine ring oxygen.If, as implied in the 1H NMR studies, the stability of the open forms is due to H-bonding, there should be additional manifestations of the H-bond on the properties and spectroscopy of the compounds. Fig. 1 (a) TTC and (b) TTT merocyanines 1A. Cross peaks in the 1H ROESY spectrum used to assign the structure are shown as curved arrows. The cross peaks of the bold H atoms are particularly diagnostic.Fig. 2 1H NMR spectra (500 MHz) of the hydroxy proton of (a) TFE equilibrated in the dark, and (b) the same solution measured after irradiation with UV light. The TFE-d3 solution contains 2/2A (15 mM) and TFE (0.5 mM). Fig. 3 (a) TTC and (b) TTT merocyanines 2A. Cross peaks in the 1H ROESY spectrum used to assign the structure are shown as curved arrows. Chem. Commun., 1998, 2685ndash;2686 2685By perturbing a solution of 2/2A in TFE with the appropriate wavelength range of light, excess 2 or 2A can be formed.Both time courses can be analyzed as first order reactions within the temperature range of 15ndash;40 deg;C (Fig. 4). Both activation energies are equal (25 plusmn; 1 kcal mol21 2?2A, 25 plusmn; 1 kcal mol21 2A?2), so the resulting difference is the enthalpy difference between 2 and 2A which is 0 plusmn; 1.4 kcal mol21. Typical values for similar molecules are ca. 4 kcal mol21 in most solvents, from the non-polar benzene and toluene to the polar EtOH and MeCN.7 The difference of 4 kcal mol21 is consistent with the enthalpy of an H-bond in general8 and with those between TFE and various acceptors in noncompetetive solvents9 in particular.If there is a specific H-bond formed as hypothesized, then there should be a shift in the lmax of the long wavelength band of the merocyanine.6 In common solvents, the long wavelength band of the merocyanine is weakly solvatochromic (e.g. lmax in toluene, MeCN and MeOH for 2A is 602 nm, EtOH, 605 nm, ethylene glycol and 3-methyloxazolidin-2-one, 610 nm).However, in the fluoro alcohols there is considerable solvatochromism (lmax values: HFP, 534 nm; TFE, 582 nm; FE, 590 nm). The values of ET(30) as a benchmark for the strength of hydrogen-bond donation and polarity of the fluoro alcohols are also in the order HFP TFE FE ET(30): 63.3, 59.8, 55.5 respectively.4f Furthermore, there is a good correlation between the energy of the long wavelength band of the merocyanine and the logarithm of the equilibrium constants (estimated as 2A/2 from the NMR data).Thus, changes in the energy of the ground state, through solvent H-bonding to the oxygen, influence the spectroscopy and the equilibrium consistently. We have performed ab initio calculations13,14 on all the solvent molecules and the solute 2A. We also performed preliminary calculations on the H-bonded complexes at the semiempirical quantum level using PM3 parameterization.Full geometry optimizations of the complexes with the solvents HFP, TFE and FE were performed. The calculated H-bond distance between the O of the 2A carbonyl and the H of the hydroxy hydrogen of the solvent are 1.7729, 1.818 and 1.820 Aring; for HFP, TFE and FE, respectively. The calculations support the observed order in the effectiveness of the fluoro alcohols in stabilizing the merocyanine form. Theory also shows very little change in the Ondash;H bond distances in the fluoro alcohols (difference: complex-free solvent (Aring;) HFP: 0.014, TFE: 0.0163, FE: 0.0058). Finally, we note that the behavior in the Lewis acid solvents differs dramatically from the behavior in the weak Brouml;nsted acid, glacial acetic acid (AA).A sample of 2/2A in AA, after a considerable time, yields a band at 420 nm. This solution is not photoactive. The band at 420 nm arises from protonated 2A with TTT geometry Fig. 3(b).1b When a quantity of Et3N equivalent to the AA is added to the solution, the band due to the protonated form disappears. There is no significant absorbance in the visible region.Illumination with UV light yields a transient absorption peak at 610 nm, which is characteristic of 2A in polar solvents. The Lewis acid ZnII causes the ring opening reaction 2?2A. The complex is stable, as is the solvate of HFP. The visible wavelength is 538 nm, between that of HFP and TFE solutions. Thus, the original hypothesis that Lewis acids of different sorts similarly influence the structure and properties of the photochromic species is confirmed.We are grateful to Dr Barry van Gemert and Mr David Knowles at PPG, Inc., Chemicals Division, for providing 1 and 2. This work was supported by the Office of Naval Research and the National Science Foundation (CHE-9710213). Notes and references 1 (a) M. T. Stauffer, D. B. Knowles, C. Brennan, L. Funderburk, F.-T. Lin and S. G. Weber, Chem. Commun., 1997, 287; (b) M.J. Preigh, F. Lin, K. Z. Ismail and S. G. Weber, J. Chem. Soc., Chem. Commun., 1995, 2091. 2 J. D. Winkler, C. M. Bowen and V. Michelet, J. Am. Chem. Soc., 1998, 120, 3237 and references cited therein. 3 (a) F. Ciardelli, D. Fabbri, O. Pieroni and A. Fissi, J. Am. Chem. Soc., 1989, 111, 3470; (b) A. Fissi, O. Pieroni, F. Ciardelli, D. Fabbri, G. Ruggeri and K. Umezawa, Biopolymers, 1993, 33, 1505. 4 (a) K. F. Purcell and S. T. Wilson, J. Mol. Spectrosc., 1967, 24, 468; (b) A.D. Sherry and K. F. Purcell, J. Phys. Chem, 1970, 74, 3535; (c) K. F. Purcell, J. A. Stikeleather and S. D. Brunk, J. Am.Chem. Soc., 1969, 91, 4019; (d) Y. Marcus, Chem. Soc. Rev., 1993, 22, 409; (e) M. J. Kamlet, J.-L. M. Abboud, M. H. Abraham and R. W. Taft, J. Org. Chem., 1983, 48, 2877; (f) C. Reichardt, Chem. Rev., 1994, 94, 2319. 5 This shorthand nomenclature reflects the transoid or cisoid orientation about the partial double bonds between the former spiro carbon and the naphthalene ring. 6 The synthesis generates about 50 each of the 1,3,3-trimethyl 5,6-dimethyl compound and 1,3,3-trimethyl 4,5-dimethyl compound. Such mixtures are not distinguishable as mixtures except by NMR spectroscopy. For example, first order processes fit a single exponential, and UVndash;VIS spectra do not show extra bands or shoulders compared to analogous compounds with no methyl substituents on the aromatic portion of the indole ring (positions 4, 5, 6 and 7). 7 J. B. Jr. Flannery, J. Am. Chem. Soc., 1968, 90, 5660; N. Y. C. Chu, Can. J. Chem., 1983, 61, 300; S. Keum, M. Hur, P. M. Kazmaier and E. Buncel, Can. J. Chem., 1991, 69, 1940. 8 M. D. Joesten and L. J. Schaad, Hydrogen Bonding, Marcel Dekker, New York, 1974; S. N. Vinogradov and R. H. Linnell, Hydrogen Bonding, Van Nostrand Reinhold, New York, 1971; G. A. Jeffrey and Y. Yeon, Acta Crystallogr., 1986, B42, 410. 9 L. Eberson, M. P. Hartshorn, O. Persson and F. Radner, Chem. Commun., 1996, 2105. 10 A full geometry optimization was carried out at the RHF level using the 6-31G* basis set with the GAUSSIAN94 package: GAUSSIAN 94, Revision D.4, M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head- Gordon, C. Gonzalez, and J. A. Pople, Gaussian, Inc., Pittsburgh PA, 1995. 11 Y. J. Chang and E. W. Castner, Jr. J. Phys. Chem., 1996, 100, 2684; W. J. Hehre, L. Radom, P. v. R. Schelyer and J. A. Pople, Ab Initio Molecular Orbital Theory, Wiley, New York, 1986. Communication 8/06316H Fig. 4 Kinetics of the opening and closing reactions of 2/2A in TFE (0.7 mM) at 22 deg;C. Activation energies are in kcal mol21. 2686 Chem. Commun., 1998, 2685ndash;2686