Transient spectroscopic characterization of the ring ...

2 downloads 0 Views 944KB Size Report
Joseph W. Perry a. ** and W. J. Brittain b. *. Tetrahydrochromeno is a structural variant of spiropyran that undergoes a reversible ring-opening to generate a ...
Research article Received: 12 May 2015,

Revised: 6 October 2015,

Accepted: 6 November 2015,

Published online in Wiley Online Library: 29 December 2015

(wileyonlinelibrary.com) DOI: 10.1002/poc.3522

Transient spectroscopic characterization of the ring-opening reaction of tetrahydrochromeno [2,3-dimethyl]indole Ariel S. Marshalla, Robert A. Rogersb, Joseph W. Perrya** and W. J. Brittainb* Tetrahydrochromeno is a structural variant of spiropyran that undergoes a reversible ring-opening to generate a colored nitrophenolate intermediate. Earlier work confirmed this intermediate through trimethylsilyl cyanide trapping under continuous irradiation. We have performed transient absorption spectroscopy to further characterize the mechanism of the ringopening reaction. Excitation at 355 nm produced a transient species with an absorption maximum at 445 nm, which we assign to the nitrophenolate unit of the ring-opened product. The transient absorption decays after ~970 ns with small optical density changes corresponding to a 0.15 quantum yield. Exposure to oxygen did not exhibit a significant deleterious effect on the photoisomerization of the chromeno dye. Time-dependent density functional theory corroborated spectroscopic assignments of the starting chromeno and the putative ring-opened intermediate. The excited state behavior of this system parallels the structurally similar oxazine system reported by Raymo and coworkers. The one significant difference is the longer lifetime of the photochemically generated intermediate from chromeno. Copyright © 2015 John Wiley & Sons, Ltd. Keywords: photochromism; ring-opening; spiropyran; oxazine; transient spectroscopy; density functional theory

INTRODUCTION

J. Phys. Org. Chem. 2016, 29 221–226

* Correspondence to: W. J. Brittain, Department of Chemistry and Biochemistry, Texas State University, San Marcos, TX 78666, USA. E-mail: [email protected] ** Correspondence to: Joseph W. Perry, School of Chemistry and Biochemistry, Center for Organic Photonics and Electronics, Georgia Institute of Technology, 901 Atlantic Drive NW, Atlanta, GA 30332-0400, USA. E-mail: [email protected] a A. S. Marshall, J. W. Perry School of Chemistry and Biochemistry, Center for Organic Photonics and Electronics, Georgia Institute of Technology, 901 Atlantic Drive NW, Atlanta, GA, 30332-0400, USA b R. A. Rogers, W. J. Brittain Department of Chemistry and Biochemistry, Texas State University, San Marcos, TX, 78666, USA

Copyright © 2015 John Wiley & Sons, Ltd.

221

The potential for applications of light-induced molecular switching has been demonstrated in a wide range of examples including photoresponsive materials,[1–5] optical switches,[6–8] memory storage,[9] and colormetric analytics.[10,11] In particular, the use of photochromic compounds[12] in photonic devices has garnered much attention, as each isomer of the photochromic compound can represent a “0” or “1” in the binary number system.[12–14] An ideal photoswitch should possess two spectrally well-separated states, and exhibit high thermal stability, fast response times regardless of medium, high sensitivity, and low fatigue, which determines the number of switching cycles the molecules can endure. Such specificity of desired optical and material traits has resulted in considerable efforts toward the understanding of key parameters responsible for efficacious isomerization. Ultrafast pump-probe spectroscopy has allowed for studies of the photophysical properties of well-known photochromic organic molecules, such as spiropyrans (SPs), spirooxazines, azobenzenes, and diarylethenes.[15,16] For example, the dynamics of SP reveal that excitation in the UV spectral range results in spirocyclic [C–O] bond cleavage followed by cis–trans isomerization, transforming SP into the open, colored merocyanine (MC) form (Scheme 1).[17] While the C–O bond cleavage occurs on the picosecond timescale, the isomerization to the MC form occurs over several milliseconds.[17] Moreover, the thermal reversion of MC back to SP is limited by the slow cis–trans isomerization, making the photoisomerization of SP susceptible to degradation with repeated photoisomerization cycles.[18] Because of the unfavorable dynamics of SP, new chromic systems that do not require cis–trans isomerization have been explored. Notable work by Raymo, Sortino, and coworkers[19] have shown that [1,3]-oxazines (OX) do not require cis–trans isomerization as part of their photocycle and exhibit fast

picosecond switching times with no degradation after thousands of cycles (Scheme 1). Raymo et al. demonstrated that the OX substituents influenced the efficiency of the photochemical process and isomerization kinetics.[20] We have previously described the synthesis of a system analogous to OX, tetrahydrochromeno[2,3-b]indole (CR) (Scheme 1).[10] The impetus for the present work was to further study the details of the ring-opening process in this class of chromophores that involve C–O bond cleavage at hemiaminal carbon. SP, OX, and CR all possess a hemiaminal carbons and generate a 4nitrophenolate chromophore upon ring-opening; the structures differ in the indole ring location for the bridging group and presence/absence of C = C bond between indole and phenolate. Like OX, the CR system does not require cis–trans isomerization as part of its photocycle. We have reported the photoactivity of CR using steady-state spectroscopic techniques, wherein the ringopened photoisomer reacted with trimethylsilyl cyanide (trapping was selective for the putative intermediate and did not occur with CR).[10] Our results suggest that CR is a viable chromic system with

A. S. MARSHALL ET AL.

a tunable wavelength range of 430–2000 nm was used as an excitation source. The white-light probe beam was provided by a 250 W tungsten-halogen lamp (300 W radiometric power supply, model Oriel 69931, Newport). The details of the nanosecond-TA pumpprobe pump experiment were described elsewhere.[21] Computational methods Computational methods were guided by the work of Raymo.[22] Ground state calculations using DFT[23] were performed with the 6-311++G(d,p) basis set and the restricted MPW1PW91[24] and B3YLP[25] functionals as implemented in Gaussian 09.[26] Geometry optimizations and frequency calculations were carried out Scheme 1. Ring-opening of 1′,3′,3′-trimethyl-6-nitrospiro[chromene-2,2′indoline] (SP), with the polarizable continuum model for acetonitrile 5a,6,6-trimethyl-2-nitro-6,12-dihydro-5aH-benzo[5,6][1,3]oxazino[3,2-a]indole (OX1), using the integral equation formalism[27] variant. Moand 5a,6,10b-trimethyl-2-nitro-5a,6,10b,11-tetrahydrochromeno[2,3-b]indole (CR1) lecular orbitals were computed at the same level of theory, including integral equation formalism polarizable continuum model solvation model. The starting geometry similar reactivity to OX. However, further study of the ultrafast dyfor CR1 was taken from the published X-ray structure.[10] The geomnamics of CR is necessary to gain a better understanding of the etry of CR1-ROP was determined by a series of optimizations as the mechanism and kinetics of this molecule. length of the [C–O] was increased in 0.1 Å increments. At each step, Here, we report the first transient absorption (TA) studies the [C–O] bond was constrained and the remaining coordinates of CR photochromic compounds on sub-nanosecond timeoptimized at the same level of theory, B3YLP/6-31++G(d,p). The opscales. The focus of this study is the dimethyl substituted detimized energy at each step was plotted, relative to that of the initial rivative, 5a,6,10b-trimethyl-2-nitro-5a,6,10b,11-tetrahydrochrom step against the [C–O] distance to build a profile of the ring-opening eno[2,3-b]indole (CR1). Time-dependent density-functional process (Figure S1, Supporting Information). The geometry at [C–O] theory (TD-DFT) was used to characterize the excited-state = 2.5 Å corresponds to a minima following initial ring-opening and absorption (ESA) bands and assign them to either the ringwas used for subsequent optimizations and excited state calculations. opened (CR1-ROP) or ring-closed (CR1) isomers. The computaTo ensure that our starting geometry for CR1-ROP was correct, we tional component of this report is limited to excited state also performed a potential energy scan about the bridging bond dynamics and does not include data on the thermal ringbetween the indole ring and the nitrophenolate units. closing process. Excited state calculations were performed by TD[28]-DFT calculations using the same basis set and functionals of ground state EXPERIMENTAL calculations. The energies of the first three excited states were carried out with the unrestricted MPW1PW91 and B3YLP funcUnless otherwise noted, all reagents were obtained from commertionals using the same solvation model and basis set as ground cial sources and used without purification. Steady-state absorption state calculations. For both CR1 and CR1-ROP, no imaginary measurements were performed using an Ocean Optics component frequencies were found, confirming that optimized geometries system (Ocean Optics, Inc., 830 Douglas Ave, Dunedin, FL 34698, are minima on their respective potential energy surfaces. TM USA) with a HR2000+ detector, qpod sample holder, DH 2,000 source and SPECTRA SUITE software for data collection. The synthesis RESULTS of 5a,6,10b-trimethyl-2-nitro-5a,6,10b,11-tetrahydrochromeno[2,3[10] b]indole (CR1) has been previously described. Transient absorption spectroscopy Transient absorption spectroscopy

222

Femtosecond-TA spectra and kinetic traces were measured with a commercially available broadband pump-probe spectrometer (HELIOS, Ultrafast Systems LLC, 1748 Independence Blvd, Sarasota, FL 34234, USA) using a femtosecond Ti : Sapphire regenerative amplifier laser source (Solstice, Spectra-Physics (Santa Clara, CA, USA), 800 nm, 3.7 W average power, 100 fs pulse width, and 1 KHz repetition rate) and a computer-controlled optical parametric amplifier (TOPAS, Spectra-Physics (Santa Clara, CA, USA), wavelength range: 266–2290 nm, pulse width : ~75 fs HW1/e) pumped by the amplified laser. The details of the measurement have been described earlier.[21] Samples were prepared in acetonitrile at concentration of 0.1 mM, and were excited at 355 nm with a pulse energy of ~2800 nJ. For nanosecond TA, a Q-switched Nd : YAG (Pro250, Quanta-Ray, ~5 ns pulse duration, 10 Hz repetition rate) pumped optical parametric oscillator laser (MOPO 730, Quanta-Ray), with

wileyonlinelibrary.com/journal/poc

In order to understand the ultrafast mechanisms underlying the photoisomerization of chromeno dyes, TA spectroscopy on CR1 was performed. Representative femtosecond-TA spectra are shown in Fig. 1(a) where broad, ESA band in the visible region with a maximum wavelength of 445 nm, and two resolved shoulders at 485 nm and 525 nm are observed. The generation of CR1-ROP appears to be instantaneous, suggesting fast dissociation of the C–O bond cleavage upon excitation. Indeed, the femtosecond-TA kinetic results of CR1 displayed in Fig. 1(b) show that the rise time of the ring-opened species is instrument response function-limited. Substantial reduction in the ESA amplitude is observed within the first 25 ps; however, thereafter the ESA remains relatively constant in shape and magnitude. The ESA kinetic traces at 445 nm are reproduced by a monoexponential function with fast time decay of ~2 ps and an offset denoting an unrecovered or “long-lived transient”, relative to the time window of the femtosecond-pulsed pump-probe experiment. We attribute

Copyright © 2015 John Wiley & Sons, Ltd.

J. Phys. Org. Chem. 2016, 29 221–226

TETRAHYDROCHROMENO RING-OPENING

Figure 1. Femtosecond-transient absorption (TA) spectra and kinetic traces of the CR1 following excitation at 355 nm. (a) Femtosecond-TA spectra of femtosecond at various probe delay times. (b) Representative femtosecond-TA kinetic trace of CR1 at 445 nm (red circle) overlaid with exponential fit (black line). Inset shows the early decay trend of each wavelength plot

J. Phys. Org. Chem. 2016, 29 221–226

Figure 2. Nanosecond transient absorption decay kinetics of CR1-ROP monitored at 450 nm. The sample was excited at 355 nm with a pump energy of 5 mJ/pulse

very little change in the absorption band after excitation at lower pulse energies; however, higher energy excitations induce changes in the spectra of CR1 for both degassed and aerated solutions. We conclude from this study that there is little reaction between oxygen and transient species. This is consistent with only traces of degradation for CR1 solutions after repeated photocycles at high energies.

Excited state isomerization The steady state absorption spectrum of CR1 (Fig. 3(a)) in acetonitrile shows a band centered at 311 nm that resembles the spectrum of 4-nitroanisole under same conditions.[19] We have assigned this absorption to the nitrophenolate portion of CR1. The geometry of the CR1 ground state (S0) was optimized followed by single-point TD-DFT calculations at same level of theory to determine the three first excited states (Table 1). In agreement with the experimental spectrum, the B3YLP and MPW1PW91 functionals estimate an electronic absorption at 320 and 327 nm (Fig. 3(a)) with an oscillator strength of ca.

Copyright © 2015 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/poc

223

the shorter decay component to initial structural relaxation of the S1 state of CR1-ROP to the ground state of the photoisomer. The long-lived transient, however, is associated with the thermal reversion of CR1-ROP to the original chromeno dye. In previous work, we demonstrated that the reaction of CR1 with tetrabutylammonium cyanide (TBACN) produces the cyanide adduct of CR1-ROP, which is characterized by a strong absorption at 440 nm.[10] Figure 3(b) displays the position of the absorption maximum for the adduct relative to the 1 ps spectrum from femtosecond-TA; the full absorption spectrum of this adduct is analogous to tetrabutylammonium pnitrophenolate which has a molar absorptivity of 25,000 L M-1 cm-1 at 430 nm. Therefore, we assign the ESA band maximum to be the absorbance of CR1-ROP. However, the transient signal received from CR1 after excitation with sizable pulse energies is relatively weak. We estimate a 0.15 quantum yield for the long-lived transient species which was estimated as the quotient of the infinite component, and the sum of all amplitudes obtained from the exponential fit of the ground-state recovery. Thus, the equilibrium between the ring-opened and ring-closed isomers of CR1 does not exist when excluded from irradiative conditions. Nanosecond-TA was utilized to further identify and determine the lifetime of the long-lived transient seen in the femtosecondTA. Nanosecond transient kinetic traces displayed in Fig. 2 show a monoexponential decay with a 970 ns half-life of the intermediate. The nanosecond-TA measurement was conducted under both degassed and aerated conditions and revealed no quenching of the excited state, indicating that the transient signal is not due to triplet–triplet absorption of CR1. It then follows that the long-lived ESA observed in the femtosecond and nanosecond time regime is the absorption of photoisomer, CR1-ROP. The monoexponential kinetic decay is indicative of a simple first order process during conversion from the ringopened isomer of CR1-ROP to the ring-closed CR1 isomer and suggests that no additional intermediate plays a significant role in the overall interconversion process. Using nanosecond-pulsed excitation under degassed and aerated conditions, measurements were also conducted to determine the susceptibility of the CR1 system to photodegradation, wherein the steady-state absorption spectrum of the chromeno dye was monitored after exposure to a 6 ns pulse at 355 nm at various excitation energies. The steady-state absorption spectra of CR1 (Figure S2, Supporting Information) reveal that there is

A. S. MARSHALL ET AL.

Figure 3. (a) Experimental absorption spectrum of CR1 (0.06 M, MeCN) and electronic transitions calculated with MPW1PW91 (1) and B3YLP (2) functionals. (b) Experimental absorption spectrum of CR1-ROP recorded 1 ps after pulsed laser excitation (355 nm), electronic transitions calculated with MPW1PW91 (1) and B3YLP (2) functionals, and (3) absorption maximum of CR1 adduct with TBACN (0.06 M, MeCN; λmax = 437 nm, ε = 4.1 mM -1 1 Lcm )

Table 1. Excitation energy (ΔE), wavelength (λ), oscillator strength (f), and main orbital pair with its contribution for electronic transitions from the ground state of CR1 to the first three singlet excited states calculated ΔE (eV) MPW1PW91 Functional S1 2.81 S2 3.87 S3 3.98 B3YLP Functional S1 2.75 S2 3.78 S3 3.87

λ (nm)

f

Main orbital pair

Contribution (%)

442 320 312

0.0853 0.3033 0.0118

[HOMO] → [LUMO] [HOMO-1] → [LUMO] [HOMO-5] → [LUMO]

99 93 78

450 327 320

0.0821 0.3129 0.0205

[HOMO] → [LUMO] [HOMO-1] → [LUMO] [HOMO-5] → [LUMO]

99 92 73

0.3. This absorption involves an electronic transition from [HOMO-1] to [LUMO]. Visualization of the corresponding isosurfaces (Fig. 4(a)) reveals that the molecular orbitals are primarily centered on the 4-nitrophenolate chromophore. This electronic transition results in population of the second singlet excited state (S2) of CR1. S1 is an intramolecular charge transfer state that is not observed, and TD-DFT predicts a negligible oscillator strength for this forbidden process that corresponds to a [HOMO] to [LUMO] transition.

Excitation of CR1 using femtosecond-TA spectroscopy results in appearance of band with a maximum centered at 445 nm. This TA is similar to the ground-state absorption of the 4nitrophenolate chromophore. The TD-DFT method used for CR1 was repeated to determine the first three excited states of CR1-ROP (Table 2). In agreement with experimental spectrum, the B3YLP and MPW1PW91 models estimated an electronic absorption at 453 and 454 nm, respectively (Fig. 3(b)), with a small oscillator strength of 0.01–0.001. This absorption involves an electronic transition from [HOMO] to [LUMO]. Visualization (Fig. 4(b)) indicates that the HOMO is localized on the 4nitrophenolate chromophore and the unoccupied orbital molecular orbital extends over most of the molecule.

DISCUSSION

224

Figure 4. Isosurfaces calculated with MPW1PW91 functional: (a) CR1 [HOMO – 1], [HOMO], and [LUMO]; (b) CR1-ROP [HOMO] and [LUMO]

wileyonlinelibrary.com/journal/poc

Copyright © 2015 John Wiley & Sons, Ltd.

The TA and TD-DFT results for CR1 mirror many features of the analogous oxazine system reported by Raymo and coworkers. Raymo[22] used computations to elucidate the elementary steps and confirm ESA assignments. He also proposed a mechanistic model for excited state dynamics that involves (1) initial excitation of OX1 to the S2 state, (2) internal conversion and intersystem crossing of S2 to S1, (3) internal conversion to S1 to T1, and (4) ring-opening from T1 state.

J. Phys. Org. Chem. 2016, 29 221–226

TETRAHYDROCHROMENO RING-OPENING Table 2. Excitation energy (ΔE), wavelength (λ), oscillator strength (f), and main orbital pair with its contribution for electronic transitions from the ground state of CR1-ROP to the first three singlet excited states calculated ΔE (eV) MPW1PW91 Functional 2.73 S1 S2 3.40 S3 3.61 B3YLP Functional S1 2.73 S2 3.36 S3 3.55

λ (nm)

f

454 364 344

0.0125 0.402 0.0219

[HOMO] → [LUMO] [HOMO] → [LUMO + 1] [HOMO-1] → [LUMO]

92 87 55

453 369 349

0.0015 0.4445 0.0191

[HOMO] → [LUMO] [HOMO] → [LUMO + 1] [HOMO-1] → [LUMO + 1]

93 90 52

Ring-closure occurs after intersystem crossing of T1 to S0. The involvement of the triplet state was deduced from energy profiles of the putative excited states in which the T1 state possessed the lowest energy pathway to ring-opening. The involvement of the triplet state has been established in related chromophores. SP molecules undergo significant degradation during photoisomerization, which is attributed to a long-lived (milliseconds) triplet state participating in the cis–trans isomerization of MC.[16] Specifically, the presence of the triplet state encourages the production of singlet oxygen, which is presumably responsible for the oxidative degradation of MC.[18] While Raymo implicates triplet participation in the excited state dynamics, Raymo, Sortino, and coworkers[19,20] observed no significant degradation of OX2 after 3,000 excitation cycles in the presence of molecular oxygen. The lack of degradation for OX after thousands of photocycles suggests that the triplet state is sufficiently short-lived to prevent oxidative degradation. We did not assess involvement of a triplet state in our computational study but the lack of excited state quenching in aerated solutions of CR1 argues that if triplet states are involved, their lifetime is short. The primary difference between the present study and Raymo is the lifetime of the ringopened species. The use of femtosecond-TA provided absorption data in the picosecond regime of the excitation process. We speculate that the initial, rapid absorption decay within the first 25 ps (Fig. 1(a)) corresponds to initial relaxation of S2, and the longer time component of the nanosecond-TA studies corresponds to the ring-closure of CR1-ROP. The lifetime of the ring-opened isomer from CR1 is 40× higher than OX1 lifetime. The origin of this longer lifetime is speculative and may be due to the difference between C and N (capable of pyramidal inversion) as the anchor point for the 4-nitrophenolate chromophore. In OX1, ring-closure can occur on either face of the indole ring, while ring-closure of CR1ROP is restricted to a single face.

REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

[19] [20] [21]

[24] [25] [26]

V. I. Minkin, Chem. Rev. 2004, 104, 2751–2776. V. I. Minkin, Russ. Chem. Rev. 2013, 82, 1–26. M. Ueda, H.-B. Kim, K. Ichimura, Mater. Lett. 1994, 20, 245–247. J. Whelan, D. Abdallah, J. Wojtyk, E. Buncel, J. Mater. Chem. 2010, 20, 5727–5735. R. Klajn, Chem. Soc. Rev. 2014, 43, 148–184. A. Broo, Int. J. Quantum Chem. 2000, 77, 454–467. M. Saphiannikova, V. Toshchevikov, J. Ilnytskyi, Nonlinear Opt. Quantum Opt. Concepts Mod. Opt. 2010, 41, 27–57. M. Q. Zhu, L. Zhu, J. J. Han, W. Wu, J. K. Hurst, A. D. Q. Li, J. Am. Chem. Soc. 2006, 128, 4303–4309. G. Berkovic, V. Krongauz, V. Weiss, Chem. Rev. 2000, 100, 1741–1753. N. Douglas, C. J. Neef, R. A. Rogers, J. A. Stanley, J. Armitage, B. Martin, T. W. Hudnall, W. J. Brittain, J. Phys. Org. Chem. 2013, 26, 688–695. Y. Shiraishi, K. Adachi, M. Itoh, T. Hirai, Org. Lett. 2009, 11, 3482–3485. X. Guo, D. Zhang, T. Wang, D. Zhu, Chem. Comm. 2003, 914–915. F. M. Raymo, S. Giordani, J. Am. Chem. Soc. 2002, 124, 2004–2007. H. Tian, B. Qin, R. Yao, X. Zhao, S. Yang, Adv. Mater. 2003, 15, 2104–2107. S. V. Paramonov, V. Lokshin, O. A. Fedora, J. Photochem. Photobiol. C. 2011, 12, 209–236. J. Zhang, Q. Zou, H. Tian, Adv. Mater. 2013, 25, 378–399. (a) N. P. Ernsting, T. Arther-Engelend, J. Phys. Chem. 1991, 95, 5502–5509; (b) H. Görner, Phys. Chem. Chem. Phys. 2001, 3, 416–423; (c) H. Görner, Chem. Phys. Lett. 1998, 282, 381–390. (a) K. Ichimura, in: Photochromism: Molecules and Systems(Edss: H. Dürr , H. Bouas-Laurent ), Eds: Amsterdam, Elsevier, 1990, 903–918; (b) K. Ichimura, T. Seki, Y. Kawamishi, Y. Suzuki, M. Sakuagi, T. Tamaki, in: Photo-refractive Materials for Ultrathin Density Optical Memory(Eds: M. Irie ), Ed: Elsevier, Amsterdam, 1994, 55–83; (c) K. Ichimura, in: Photochromic Polymers(Edss: J. Crano , R. Gugliemetti ), 2 Eds: Plenum Press, New York, 1999, 9–63. M. Tomasulo, S. Sortino, A. J. P. White, F. M. Raymo, J. Org. Chem. 2005, 70, 8180–8189. E. Deniz, M. Tomasulo, S. Sortino, F. M. Raymo, J. Phys. Chem. C. 2009, 113, 8491–8497. J. M. Hales, M. Cozzuol, T. E. O. Screen, H. L. Anderson, J. W. Perry, Opt. Express 2009, 17, 18478–18488. F. M. Raymo, J. Phys. Chem. A. 2012, 116, 11888–11895. R. G. Parr, W. Yang, Density-functional Theory of Atoms and MoleculesOxford University Press, Oxford, U. K., 1989. (a) A. D. Becke, Phys. Rev. A. 1988, 38, 3098–3100; (b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B. 1988, 37, 785–789. J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson, M. R. Pederson, D. J. Singh, C. Fiolhais, Phys. Rev. B. 1992, 46, 6671–6687. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr. , J. E. Peralta, F. Ogliaro,

Copyright © 2015 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/poc

225

We thank F. Raymo for the extensive and helpful discussions and Dr. San-Hui for the data processing. WJB thanks the National Science Foundation (PREM Center for Interfaces, DMR1205670) and the ACS Petroleum Research Fund (51997UR7) for financial support of this research. JWP acknowledges support from the DARPA ZOE program (Grant No. W31P4Q-09-10012). The authors also acknowledge the support of the National Science Foundation for X-ray and NMR instrumentation

Contribution (%)

(CR1IF:MU-0946998 and MRI-0821254) and the Welch Foundation (AI-0045) for support.

[22] [23]

Acknowledgements

J. Phys. Org. Chem. 2016, 29 221–226

Main orbital pair

A. S. MARSHALL ET AL.

M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09, Revision A.1Gaussian, Inc, Wallingford, CT, 2009.

[27] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–3093. [28] F. Furche, R. Ahlrichs, J. Chem. Phys. 2002, 117, 7433–7447.

SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher's web site.

226 wileyonlinelibrary.com/journal/poc

Copyright © 2015 John Wiley & Sons, Ltd.

J. Phys. Org. Chem. 2016, 29 221–226

Copyright of Journal of Physical Organic Chemistry is the property of John Wiley & Sons, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.