Quenching of Triplet Benzophenone by Benzene and Diphenyl Ether

Sep 14, 2010 - Margaret J. Smith and Götz Bucher*. WestCHEM, Department of .... Noble Research in The Journal of Organic Chemistry. The following pap...
1 downloads 0 Views 664KB Size
10712

J. Phys. Chem. A 2010, 114, 10712–10716

Quenching of Triplet Benzophenone by Benzene and Diphenyl Ether: A DFT Study Margaret J. Smith and Go¨tz Bucher* WestCHEM, Department of Chemistry, UniVersity of Glasgow, Joseph-Black-Building, UniVersity AVenue, Glasgow G12 8QQ, United Kingdom ReceiVed: June 28, 2010; ReVised Manuscript ReceiVed: August 24, 2010

The reaction of triplet benzophenone with benzene and diphenyl ether has been studied by density functional theory. Quenching of the triplet ketone is predicted to occur by addition of the carbonyl oxygen to the arene chromophores. The reaction is accompanied by a significant degree of charge transfer. In case of the reaction of triplet benzophenone with diphenyl ether (DPE), addition is predicted to occur preferentially at the ortho position of the DPE molecule. Addition to the ipso-position of DPE, which provides a pathway for formation of the phenoxy radical, is predicted to occur as a minor reaction pathway. Introduction The lifetime of the (n,π*) lowest triplet excited state of benzophenone 1 is a function of the nature of the solvent, with the longest lifetimes being observed in perhalogenated alkanes such as carbon tetrachloride or perfluoromethylcyclohexane (τ ) 720 µs).1 If benzene is used as solvent, then the triplet lifetime of 1 is considerably shorter (τ ) 12 µs).2 The evidence for the mechanism of quenching of triplet 1 by arenes is somewhat controversial. Product studies of the photolysis of very pure samples of 1 in very pure, degassed benzene indicate that hydrogen abstraction from the solvent takes place, with benzopinacol 2, biphenyl 3 and 4-biphenyldiphenylcarbinol 4 being formed as main products (Scheme 1). The quantum yields, however, are exceedingly low (Φ e 0.0025).3 The very low quantum yield observed for the hydrogen atom abstraction reaction indicates that it likely is not the reason for the short lifetime of triplet benzophenone in benzene. Another mechanism that has been suggested to account for the experimental observations is addition of the triplet ketone to the π-system of the benzene molecule, resulting in formation of a biradical 5 (Scheme 2).4 This general mechanism has recently been shown by us to be responsible for the very short triplet lifetime of β-phenylpropiophenone 6.5 The reaction, called β-phenyl quenching (BPQ), proceeds via addition of the triplet carbonyl oxygen atom to the ipso- or ortho-carbon of the β-phenyl ring, followed by ISC of biradicals 7 or 8 and relaxation to the ground-state ketone (Scheme 3). It is noted that 7 and 8 are intramolecular triplet exciplexes, as they have no minima on the singlet hypersurface. Whereas no products due to the addition pathway could be isolated in the photolysis of 1 in the presence of benzene, or in BPQ of 6, photolysis of 1 in the presence of diphenyl ether 9 yielded products indicative of initial addition of the triplet ketone to the electron-rich diphenyl ether molecule.6 Irradiation of 1 in neat diphenyl ether yielded benzophenonediphenylketal 10, which is formed by recombination of the phenoxy-diphenylmethyl radical 11 and a phenoxy radical 12 formed in the fragmentation of biradical 10 (Scheme 4). Small and Scaiano later studied the reaction by laser flash photolysis (LFP).4 Laser excitation (λ ) 337 nm) of 1 in

isooctane containing diphenyl ether (DPE) gave a rate constant kq ) (6.2 ( 1.0) × 105 M-1 s-1 for the quenching of 31 by DPE. In neat DPE, the transient spectrum observed 1 µs after the laser pulse showed a maximum around λ ) 440 nm, decaying with τ ) 13 µs according to first-order kinetics. As the transient differed in its behavior from both triplet 1 and from the phenoxy radical 13, it was assigned to 11, or to another biradical resulting from addition of the triplet ketone to the ortho-, meta-, or para-positions of DPE. Later fluorescence studies also indicated the formation of exciplexes in the reaction of triplet cyclohexanone or acetone with alkylbenzenes.7,8 SCHEME 1: Products Formed in the Photoreduction of Benzophenone Triplet in Benzene

SCHEME 2: Addition Mechanism for Quenching of Triplet Benzophenone by Benzene

* To whom correspondence should be addressed. E-mail: goebu@ chem.gla.ac.uk.

10.1021/jp105962r  2010 American Chemical Society Published on Web 09/14/2010

Quenching of Triplet Benzophenone

J. Phys. Chem. A, Vol. 114, No. 39, 2010 10713

SCHEME 3: Deactivation Mechanism of Triplet β-Phenylpropiophenone (β-Phenyl Quenching)

SCHEME 4: Formation of Ketal 10 in the Photolysis of 1 in Diphenyl Ether 9

The current study employs up-to-date DFT methods to investigate the reactions of triplet benzophenone with two aromatic substrates, benzene and diphenylether, in order to gain a thorough understanding of the reaction pathways. Computational Methods Two DFT methods were utilized to characterize stationary points along the reaction coordinate of the triplet benzophenone + benzene reaction. In addition to the standard B3LYP hybrid functional,9 we employed the M05-2X method,10 which has been shown to provide very good results, even if used on larger systems.11 In all calculations, the Dunning cc-pVTZ basis set was used.12 All stationary points were characterized by performing a vibrational analysis. UV/vis

spectra were calculated at the TD-(U)B3LYP/6-31+G* level of theory.13 IRC calculations were performed at the UB3LYP/ 6-31G(d) level of theory,14 followed by UB3LYP/cc-pVTZ single point energies for each point. Singlet biradicals were optimized using broken-symmetry wave functions (Gaussian keywords guess ) (mix,always)). The UV/vis spectra were obtained from the calculated transitions by Lorentzian broadening of the bands (Chemcraft software).15 The influence of the solvent was taken into account by employing a polarizable continuum model (scrf ) pcm).16 All DFT calculations were performed employing the Gaussian09 suite of programs.17 Partial charges on the benzophenone chromophore were calculated by summing the Mulliken charges of the atoms making up the benzophenone moiety.

10714

J. Phys. Chem. A, Vol. 114, No. 39, 2010

Smith and Bucher

Figure 1. Reaction energies (∆U and ∆U‡, relative to 31 + benzene ) 0.0 kcal mol-1) calculated for the reaction of 31 and benzene. Normal font: (U)B3LYP/cc-pVTZ + ZPE. Italics: (U)M05-2X/cc-pVTZ + ZPE. n.m.: no minimum.

Figure 2. Plot of the energy (in atomic units) vs the O (carbonyl)-C (benzene) distance, as calculated for the system 31 + benzene at the UB3LYP/cc-pVTZ//UB3LYP/6-31G(d) level of theory. Black circles: triplet energy. Light circles: singlet energy at triplet geometry. The transition state of the reactions is indicated by the data points at R ≈ 1.95 Å having two larger gaps on each side.

Figure 3. Plot of the spin density at the carbonyl oxygen atom of 31, vs the O-C distance as calculated for the system 31 + benzene at the UB3LYP/cc-pVTZ//UB3LYP/6-31G(d) level of theory. The transition state of the reactions is indicated by the data points at R ≈ 1.95 Å having two larger gaps on each side.

Results and Discussion Triplet Benzophenone + Benzene. The transition state for the reaction of triplet 1 with benzene and the resulting biradical were optimized. The results indicate that the reaction should be quite facile, easily occurring on a sub-µs time scale. We note that the M05-2X functional, which has been reported to give results far better than B3LYP,11,18 gives a lower value for the activation enthalpy. The results are shown in Figure 1. The triplet energy of benzophenone is calculated as ∆UT ) 62.1 kcal mol-1 at the B3LYP level of theory, and 66.4 kcal mol-1 using M05-2X. Again, the M05-2X result is in much better agreement with the experimental value (68.7 kcal mol-1).19 We analyzed the reaction between 31 and benzene further by calculating (UB3LYP/6-31G(d)) the intrinsic reaction coordinate, starting with the transition structure. Of the resulting series of geometries along the reaction coordinate thus obtained, we calculated single point energies using a cc-pVTZ basis set for both the triplet state and the singlet state at the geometry of the triplet state. As the reaction coordinate corresponds to the change in the distance between the carbonyl oxygen and the benzene carbon atom being attacked, this RC-O was taken as x-axis in the IRC plots. Figure 2 shows that the energy gap between the triplet state and the singlet state at the geometry of the triplet state is only small very close to the geometry of the biradical 5. ISC to the singlet spin manifold therefore is expected to be efficient only late along the reaction coordinate.

Figure 4. Plot of the negative charge on the benzophenone moiety vs C-O distance in the reaction of 31 with benzene as calculated for the system 31 + benzene at the UB3LYP/cc-pVTZ//UB3LYP/6-31G(d) level of theory. Black circles: negative charge on the benzophenone moiety. The transition state of the reactions is indicated by the data points at R ≈ 1.95 Å having two larger gaps on each side.

Figure 3 reveals the sigmoidal shape typical for the spin evolution in the addition of a (n,π*) triplet to a benzene ring, with maximum spin density at O for the unperturbed triplet ketone. Figure 4 shows that the reaction goes along with significant charge transfer from the benzene molecule to the benzophenone triplet. The degree of charge transfer amounts to approximately one-third of an elementary charge, and its maximum is reached

Quenching of Triplet Benzophenone

J. Phys. Chem. A, Vol. 114, No. 39, 2010 10715

Figure 5. Stationary points along the reaction coordinate 31 + benzene f 35 f 15 f 11 + benzene. From left: TS for formation of 35, 35, 15, TS for cleavage of 15. The numbers give the distance between C1 of the benzene moiety and the carbonyl oxygen (in Å).

Figure 6. Reaction energies (∆U and ∆U‡, relative to 31 + 9 ) 0.0 kcal mol-1) calculated for the reaction of 31 and diphenylether. Normal font: (U)B3LYP/cc-pVTZ + ZPE. Italics: (U)M05-2X/cc-pVTZ + ZPE.

well beyond the transition state. Overall, the parameters studied indicate that the reaction of triplet benzophenone with benzene is very similar to BPQ of derivatives of 6 having an (n,π*) lowest triplet excited state.5 As in BPQ of 36, the singlet biradical 1 5 formed by ISC of 35 is no minimum on the potential energy hypersurface, if the B3LYP method is employed. Using the M05-2X/cc-pVTZ method, however, we could localize a very shallow minimum for 15 that is slightly lower in energy than the triplet biradical 35. Singlet biradical 15 is so weakly bound that the transition state for cleavage into ground-state benzophenone and benzene is practically equal in energy, and the length of the C-O bond to be broken is R ) 1.50 Å in 15 and only 1.55 Å in the transition state. Figure 5 shows the optimized geometries of the relevant stationary points in the quenching of 31 by benzene. The reaction of triplet benzophenone with diphenyl ether was also investigated by computational methods. Again, both UB3LYP/cc-pVTZ and UM05-2X/cc-pVTZ were employed. All stationary points were fully optimized at these levels of theory and characterized by vibrational analyses. Figure 6 shows the results. While the two DFT methods used agree on the overall exothermicity of the formation of 12 + 13 from 31 + 9, they strongly disagree on the energies of the triplet biradicals involved. The reason for the significantly higher energies of formation of 11, and 14-16, when B3LYP is used, likely lies in its disregard for weak interactions between the π-systems involved. In contrast, M05-2X is capable of describing these interactions far better.10 Taken together, they add up to a significant stabilization of the triplet biradicals.

According to the M05-2X calculations, addition of triplet benzophenone does not take place with a clear regiochemical preference. Addition to the ortho-position of 9 is predicted to be slightly favored, but all other reaction modes, including metaaddition, are predicted to be viable as well. A decision about which intermediate is formed in the reaction 31 + 9 thus cannot be met without reasonable doubt, based on the calculated energies alone. We have therefore calculated the UV/vis spectra of all four triplet biradicals via time-dependent DFT, taking the effect of the solvent diphenyl ether into account by using a polarizable continuum model (PCM-TD-B3LYP/6-311++G(d,p)// B3LYP/cc-pVTZ). We chose the B3LYP hybrid functional, as it had recently been shown to provide superior performance in calculating triplet-triplet absorptions.20 The calculated spectra are shown in Figure 7. The experimental spectrum, measured 1 µs after LFP (λexc ) 337 nm) of 1 in neat diphenyl ether, shows an absorption maximum at λ ) 440 nm, tailing down to ca. λ ) 500 nm.4 Based on the absorption maxima, the triplet biradicals formed by ipso- and ortho-addition (311 and 314) appear to be the best candidates. Both are predicted to show intense bands with λmax ) 444 nm (311) or 442 nm (314), whereas the triplet biradicals formed by meta- or para-addition (315 or 316) are calculated to exhibit very prominent bands at λmax ) 459 nm (315) or 470 nm (316). The tailing of the experimental band could indicate that the species observed has another weaker absorption above λ ) 440 nm, which would be in agreement with the calculated spectra for both 311 and in particular 314. It therefore appears likely that the triplet biradical observed experimentally corresponds to the product of ortho-addition, 314, which is also

10716

J. Phys. Chem. A, Vol. 114, No. 39, 2010

Smith and Bucher benzene and diphenyl ether, was studied in detail by density functional theory. The results indicate that quenching operates by addition of the carbonyl oxygen atom to the arene chromophores. The reaction with diphenyl ether is predicted to be more facile than benzene quenching, with addition taking place preferentially at the ortho-position of DPE. The reaction is accompanied by significant charge transfer from the arene quencher to the triplet benzophenone moiety. Acknowledgment. This work was performed as part of the Glasgow Centre for Physical Organic Chemistry. The authors thank EPSRC for funding. Supporting Information Available: Cartesian coordinates and energies of stationary points optimized. This information is available free of charge via the Internet at http://pubs.acs.org.

Figure 7. Calculated spectra (PCM-TD-B3LYP/6-311++G(d,p)// B3LYP/cc-pVTZ) of triplet biradicals 11, 14-16, in the same spectral range as the experimental spectrum of Small and Scaiano.4 Black: 311, red: 314, green: 315, yellow: 316.

predicted (M05-2X/cc-pVTZ) to be formed with the smallest activation energy. Clearly, the other triplet biradicals 311, 315, and 316 will also be formed to a lesser degree. Only the less favorable formation of 311 provides a pathway for productive photochemistry, yielding radicals 12 and 13. The relatively low quantum yield for the formation of 10 (Φ ) 0.03)6 is in full agreement with this picture. The findings presented provide an efficient and general mechanism for the quenching of triplet ketones by arenes. In some cases, like the reaction of triplet benzophenone with diphenyl ether, there is good experimental evidence for the intermediacy of triplet biradicals with a lifetime in the microsecond range.4 In other cases, like β-phenyl quenching of triplet β-phenyl propiophenone, biradical intermediates cannot be detected.21,22 As the singlet biradicals 1BR formed by ISC of the triplet biradicals are not minima or barely so (k2 is very fast), it appears plausible that it is the ratio of rate constants k1/kISC that determines whether a triplet biradical 3BR formed by addition of a triplet ketone 3K to an arene AR is observed or not (eq 1). 3

k1

kISC

k2

K + AR 98 3BR 98 1BR 98 1K + AR

(1)

fast

If k1 > kISC, then the triplet biradical 3BR should be observable. If kISC > k1, then 3BR will be formed in quasistationary concentration only, and will not be detected. While the ratio k1/kISC can in principle be calculated using ab initio methods, such a computational investigation will require multireference treatment of the singlet biradicals and of the conical intersection linking the triplet and singlet potential energy hypersurfaces. It is beyond the scope of the current contribution. Conclusions The quenching of the quintessential “workhorse” triplet ketone, triplet benzophenone, by two aromatic quenchers,

References and Notes (1) Saltiel, J.; Curtis, H. C.; Metts, L.; Miley, J. W.; Winterle, J.; Wrighton, M. J. Am. Chem. Soc. 1970, 92, 410. (2) Bell, J. A.; Linschitz, H. J. Am. Chem. Soc. 1963, 85, 528. (3) Dedinas, J. J. Phys. Chem. 1971, 75, 181. (4) Small, R. D.; Scaiano, J. C. J. Phys. Chem. 1978, 82, 2064. (5) Bucher, G. J. Phys. Chem. A 2008, 112, 5411. (6) Nowada, K.; Hisaoka, M.; Sakuragi, H.; Tokumaru, K.; Yoshida, M. Tetrahedron Lett. 1978, 2, 137. (7) Wilson, T.; Halpern, A. M. J. Am. Chem. Soc. 1981, 103, 2412. (8) Jacques, P.; Allonas, X.; Von Raumer, M.; Suppan, P.; Haselbach, E. J. Photochem. Photobiol. A: Chem 1997, 111, 41. (9) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (10) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput. 2006, 2, 364. (11) Schreiner, P. R. Angew. Chem., Int. Ed. 2007, 46, 4217. (12) Kendall, R. A.; Dunning, T. H., Jr. J. Chem. Phys. 1992, 96, 6796. (13) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454– 464. (14) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724. (15) Zhurko, G. A.; Zhurko, D. A. Chemcraft, Vers. 1.6, 2010. (16) (a) Miertusˇ, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117. (b) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. ReV. 2005, 105, 2999. (17) Gaussian 09, ReVision A.02, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, ¨ .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; S.; Daniels, A. D.; Farkas, O Fox, D. J. Gaussian, Inc., Wallingford CT, 2009. (18) Bucher, G. Eur. J. Org. Chem. 2009, 4340. (19) Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed.; Marcel Dekker: New York, NY, 1993. (20) Silva-Junior, M. R.; Scheiber, M.; Sauer, S. A. R.; Thiel, W. J. Chem. Phys. 2008, 129, 104103. (21) Netto-Ferreira, J. C.; Leigh, W. J.; Scaiano, J. C. J. Am. Chem. Soc. 1985, 107, 2617. (22) Leigh, W. J.; Banisch, J.-A.; Workentin, M. S. Chem. Commun. 1993, 988.

JP105962R