Insights into the Thermal Eliminations and Photoeliminations of B,N

Jan 3, 2017 - Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science of Ministry of Educati...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCA

Insights into the Thermal Eliminations and Photoeliminations of B,NHeterocycles: A Theoretical Study Wen-Jie Wu, Quan-Song Li,* and Ze-Sheng Li* Beijing Key Laboratory of Photoelectronic/Electrophotonic Conversion Materials, Key Laboratory of Cluster Science of Ministry of Education, Beijing Key Laboratory for Chemical Power Source and Green Catalysis, School of Chemistry and Chemical Engineering, Beijing Institute of Technology, Beijing 100081, China S Supporting Information *

ABSTRACT: Understanding the photochemistry of organoboron compounds is essential to expand optoelectronic applications. In this work, the complete active space self-consistent field (CASSCF) and its second-order perturbation (CASPT2) methods combining with density functional theory (DFT) have been employed to investigate the elimination mechanisms of compound 6,7-dihydro-54-benzo[d]pyrido[2,1-f ][1,2]azaborininr (B4) on the ground state (S0) and the first excited state (S1). B4 is one of the 1,2-B,N-heterocycles that undergo competitive thermal elimination and photoelimination depending on the substitution groups on the B atom and the chelate backbone, thus providing a highselectivity synthesis strategy for luminescent compounds. Since the energy barrier from B4 to BH3-pyrido[1,2-a]isoindole (D1) and pyrido[1,2a]isoindole (A1) on the ground state is lower than that from B4 to 54-benzo[d]pyrido[2,1-f ][1,2]azaborininr (C4), the retraction ring reaction is expected to proceed with larger probability than the H2 elimination upon heating. On the contrary, photoelimination of H2 may take place easily due to the almost barrierless pathway on the S1 state. Remarkably, we have located an energetically available conical intersection (S1/S0)X-1, which allows for ultrafast S1 → S0 decay and subsequently generation of C4. Our results not only throw light on the experimental observations of the selectivity of thermal elimination and photoelimination but also provide detailed information on the excited state as instructional implications for further synthesis and application of B,N-embedded aromatics.

1. INTRODUCTION Organoboron species are well-known for their wide application in a large number of fields, such as probes,1,2 sensors,2−9 and optoelectronic/photoresponsive materials.7−14 Moreover, the synthetic methods of organoboron light-emitting materials are very significant in photochemistry, of great interest for experimentalists and theoreticians.15−18 In particular, B,Nheterocyclic compounds have attracted much attention because of their rich photochemistry16,19,20 related with the special electronic structure of boron and nitrogen atoms. On the basis of hybrid orbital theory, the boron atom adopts sp 2 hybridization displaying trigonal planar geometry with an unoccupied p orbital perpendicular to the molecular plane. The unoccupied p orbital of the boron atom is liable to accept extra electrons such as the lone pair electrons of nitrogen atom. The replacement of C−C unit with isoelectronic B−N unit in aromatics results in a significant difference in luminescent and chemical properties.21−23 In recent years, B,N-heterocycle compounds have been widely investigated on the luminescent properties and synthetic methods.24−28 Wang and co-workers first reported very interesting competitive elimination reactions for B,N-heterocycle compounds: HBR2 elimination versus R−H elimination, depending on the R groups (Scheme 1).26 Under heating © XXXX American Chemical Society

conditions, the B−N bond of compound B1 (X = H, R = mesityl) splits and the borane group leaves, generating greenyellow fluorescent product pyrido[1,2-a]isoindole (A1) in mineral oil at 280 °C. Nevertheless, upon heating B2 (X = methyl, R = mesityl) and B3 (X = methyl, R = methyl) at 310 and 360 °C, respectively, R−H eliminations take place and generate products C2 and C3. Interestingly, when the R group of B1 is replaced by 9-borabicyclo[3.3.1]nonane, the corresponding heating product cannot be identified. Under irradiation, only the R−H photoelimination reactions from series B to C take place, leading to a generic 1,2-B,Nheterocycle species (shown in Scheme 1).25 The analogous photoelimination reaction also happens easily in a polymer molecule or a metal-chelation molecule that contains a B,Nheterocycle.25,27 Apart from 1,2-B,N-heterocycle species, 1,3B,N-heterocycle compounds also exhibit similar competitive elimination reactivity.27 These competitive eliminations of HBR 2 or R−H are of great interest, showcasing the unprecedented high-efficiency selectivity to achieve azaborine Received: September 20, 2016 Revised: December 28, 2016 Published: January 3, 2017 A

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A Scheme 1. Thermal Eliminations and Photoeliminations Reactions of Series Ba

a

For the reaction to the right side, Bn (n = 1−4) lead to Cn (n = 1−4). For the reaction to the left side, B1 and B4 produce A1, and B2 and B3 yield A2.

calculations were performed to confirm the nature of stationary points (minima or transition states) at the same level of theory as geometry optimizations. The reaction path calculations were started from the transition-state structure following the vector with a negative eigenvalue in forward and reverse directions, using the intrinsic reaction coordinate (IRC) approach.48 The CASSCF method49,50 along with the standard 6-31G* basis set was employed to calculate the minima, transition states, and CIs on the S1 potential energy surface. The selection of active space is a crucial step of CASSCF calculations. A proper CASSCF wave function must have sufficient flexibility to model the large changes in electronic structure that can take place during the chemical reactions. Considering the feature of the system and the computational efficiency, 12 electrons in 12 orbitals (focused on the reaction domain and as many as possible) were chosen as the active space for the CASSCF calculations, so-called CAS(12, 12) hereafter. The molecular orbitals included in the (12, 12) active space at Franck− Condon (FC) geometries of A1, B4, 54-benzo[d]pyrido[2,1f ][1,2]azaborininr (C4), and BH3-pyrido[1,2-a]isoindole (D1) are displayed in Figure S1 of Supporting Information, which includes ten π/π* orbitals and two σ/σ* orbitals for B4 and 12 π/π* orbitals for A1, C4, and D1. At B4 minimum, the two σ/ σ* orbitals of one C−H bond neighboring to boron atom are included in the active space because the C−H bond breaks in either BH3 or H2 elimination. As can be seen in Figure S1 and Scheme 1, along the BH3 elimination pathway to A1, the two σ/σ* orbitals turn into the π/π* orbitals of the newly formed CC bond in the five-membered ring. Along the H 2 elimination pathway to C4, the two σ/σ* orbitals change into the π/π* orbitals of the BC bond. Moreover, the stateaveraged-CASSCF (SA-CASSCF)51 approach was applied to guarantee that the geometries of crossing space of S0 and S1 states have uniform spin multiplicity. Furthermore, we have carefully checked the active orbitals at different geometries to make them as consistent as possible. All the calculated stationary configurations were confirmed by numerical vibrational frequency calculations at the same theory level as the optimizations. The transition states and CIs on the S1 excited state were also confirmed by the IRC method. In cases of CI points, the initial relaxation directions (IRD) were determined by hypersphere geometry optimizations before IRC searching.52 All the Cartesian coordinates of optimized geometries were listed in the Supporting Information. All the geometry optimizations in this work were performed with the Gaussian 03 program package.53 To account for the dynamic electron correlation effects, the CASPT254,55 energies with the ANO-S basis set56 (contracted to B,C,N[3s2p1d]/H[2s1p]) were recalculated using MOL-

compounds without multistep reactions or transition metal catalysts.24−26 On the theoretical side, density functional theory (DFT) calculations have been employed to investigate the thermal reactions of B,N-heterocycle species. For compound B1, the calculated activation energy for HBMes2 elimination is about 100 kJ/mol less than the Mes-H elimination barrier on the ground state.26 Similarly, it is revealed that the indirect HBMes2 elimination is favored by about 50 kJ/mol over the indirect Mes-H elimination in terms of activation barrier in 1,3-BNheterocycles.27 However, previous theoretical studies mainly focus on the thermal eliminations on the ground state. The detailed behaviors of photoelimination reactivity, including the information about the excited-state potential energy surface and the conical intersection (CI) between the excited state and ground state, have not been known, which are beyond the capacity of conventional DFT calculations.29 CIs are crossings between different potential energy surfaces of the same multiplicity which act as funnels that enable efficient radiationless decay.30,31 They are known as the reasons why some molecules lack of fluorescence. The knowledge of CI has been successfully applied to illustrate the radiationless decay of benzene,32 fulvene,33 azobenzene,34 and nucleobases.35 Furthermore, CIs have been proven playing decisive roles in photoinduced Wolff rearrangement,36,37 aggregation-induced emission,38,39 and photocatalytic oxidation of methanol to formaldehyde on TiO2(110).40 Herein, the complete active space self-consistent field (CASSCF) and its second-order perturbation (CASPT2) methods in combination with DFT approach were employed to study the thermal elimination and photoelimination mechanisms of a typical B,N-heterocycle compound 6,7dihydro-54-benzo[d]pyrido[2,1-f ][1,2]azaborininr (B4) (X = H, R = H). Note that the elimination reactions of B4 have not been reported explicitly up to now, thus this study has guiding significance to some extent. We found B4 is likely to undergo retraction ring reaction when heated but prefers to undergo H2 elimination upon irradiation. The results reported here are expected to provide some new insights into the photochemistry of B,N-heterocycle species.

2. COMPUTATIONAL DETAILS In this work, the thermal chemical reactions of 1,2-B,Nheterocycle compound B4 were investigated by using the DFT method.41−43 The geometries of minima and transition states on the S0 potential energy surface were optimized at the B3LYP/6-311G* level. In the optimizations, dispersion corrections were considered using the empirical formula by Grimme et al. (i.e., DFT-D3).44−47 Analytical frequency B

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A CAS 8.057 at the CASSCF optimized structures. In the CASPT2 calculations, five roots were computed with equal weights in the CASSCF reference calculation, using a real level shift parameter58 of 0.1 and no IPEA modification.59 The facts that the CASSCF and CASPT2 energy profiles are smooth and there are no large discrepancies between them validate the suitability of our computation approach. The absolute and relative energies of all critical points are provided in Table S1 of the Supporting Information. The solvent effects were considered with the polarizable conductor model (PCM)60−62 using toluene as solvent for both thermal elimination and photoelimination.

little lower excitation energy (2.91 eV, 427 nm), a similar oscillator strength ( f = 0.011), and consistent electronic character of π6 → π7, where π6 and π7 resemble the HOMO and LUMO of TD-DFT results. Both TD-DFT and CASPT2 calculations reveal that the S2 state of A1 is a bright state ( f TD‑DFT = 0.174, f CASPT2 = 0.192) with excitation energy 3.71 eV (334 nm) by TD-DFT or 3.42 eV (362 nm) by CASPT2. This excitation involves the HOMO (π6) orbital and the LUMO+1 (π8) orbital. The CASPT2 excitation energy of 362 nm matches well with the experimental peak at about 370 nm.26 For higher excited states S3 and S4, the electronic characters by TD-DFT and CASPT2 are also similar in general. For compound B4, TD-DFT and CASPT2 methods provide analogous electronic characters for S1, S2, and S3, which are dominated by the electronic transition configurations of HOMO → LUMO (π4 → π6), HOMO → LUMO+1 (π4 → π7), and HOMO−1 → LUMO (π5 → π6), respectively. Regarding the S4 state, the TD-DFT calculation indicates that it involves the HOMO−2 orbital located at the boron atom, whereas the CASPT2 result shows that the related orbital is the π3 orbital situated on the pyridyl ring. Both methods show that the S1 state of B4 is a bright state with an excitation energy of 3.56 eV (348 nm) by TD-DFT or 4.12 eV (301 nm) by CASPT2. The main peaks of the UV/vis absorption spectra of compounds B1, B2, and B3 (Scheme 1) in toluene observed in experiments are characterized by two main peaks around 300 and 340 nm, where the lowest absorption peaks locate at 344, 343, and 320 nm, whereas the most intense absorption peaks locate at 309, 309, and 297 nm, respectively.25 It can be inferred that the substituent groups R and X have little influence on the absorption spectra for series B. For compound B4, the spectral data are not available in experiments, but we deduce that its absorption spectrum is similar to those of B1, B2, and B3, which is in reasonable line with our theoretical predications. For compound C4, similar electronic transitions for the S1 and S2 are obtained by TD-DFT and CASPT2 methods. CASPT2 method discloses that both S1 and S2 states are bright states with quite large oscillator strengths lying at 434 and 405 nm, whereas the TD-DFT method suggests that the S2 state is a bright state at 393 nm with an oscillator strength of 0.2. In addition, TD-DFT calculations give a relatively smaller S1−S2 energy difference (0.07 eV) than that (0.2 eV) by CASPT2. The CASPT2 results are in good agreement with the experimental absorption spectra of compounds C1, C2, and C3 (Scheme 1) in toluene, where three intense absorption bands appear at about 470, 450, and 425 nm, and the first two strong absorption peaks have similar intensities.25 3.2. Thermal Reaction Pathways. It is reported that the thermal reactivities of 1,2-B,N-heterocycle series B have two different and competitive thermal elimination pathways: HBR2 elimination and R-H elimination.26 For compound B4, the two competitive eliminations correspond to H2 elimination and BH3 elimination. We investigated these two reactions on the ground state at the B3LYP/6-311G* level. In the direction of BH3 elimination, a transition-state TS1-S0 (Figure 1e) that connects B4-S0 and D1-S0 (Figure 1d) has been found. At TS1S0, the B−N bond has broken, while the H1 atom migrates from C1 atom to B atom with the C1−H1 distance of 1.67 Å and the H1−B distance of 1.30 Å, accompanied by the approaching of C1 and N presenting a separation of 2.13 Å. At D1-S0, the retraction ring reaction has completed, where both C1−C2 and C1−N bond lengths are 1.48 Å and the C1−B distance is 1.70 Å. The thermal reaction pathways from TS1-S0

3. RESULTS AND DISCUSSION 3.1. S0 Equilibrium Structures and Vertical Excitations. The optimized ground-state structures of A1-S0, B4-S0, and C4S0 are shown in Figure 1a−c, respectively. The dihedral angle

Figure 1. Optimized ground-state structures of (a) A1-S0, (b) B4-S0, (c) C4-S0, and (d) D1-S0 and transition state (e) from B4 to D1 (TS1S0) and (f) from B4 to C4 (TS2-S0). Some critical bond lengths are given. Atomic symbols and labels are exemplified at B4-S0. Yellow, purple, pink, and blue atoms represent C, N, B, and H atoms, respectively.

with respect to the twist of phenyl and pyridyl groups in B4-S0 is −25.6°, whereas those in A1-S0 and C4-S0 are 0°. In A1-S0, the C1−C2 bond length is 1.40 Å, which is significantly shorter than that in B4-S0 (1.50 Å), which is because of the strong conjugation interaction between the 2pz orbital of C1 atom and the π orbitals of phenyl and pyridyl. Similarly, the 2pz orbitals of C1 and B atoms conjugate with the π orbitals of phenyl and pyridyl in C4-S0, leading to the shortening of C1−C2, C1−B, and B−N bonds from 1.50, 1.62, and 1.63 Å in B4-S0 to 1.41, 1.46, and 1.48 Å in C4-S0. The vertical excitations of A1, B4, and C4 from the minimum of S0 state to the lowest four electronic excited singlet states were calculated by using TD-DFT with B3LYP functional and CASPT2/CASSCF methods. The obtained vertical excitation energies, oscillator strengths, and dominant configurations are presented in Table 1; meanwhile the related orbitals are displayed in Table S2. For compound A1, TD-DFT calculations show that the S1 state is a dark state with small oscillator strength (f = 0.022) and excitation energy of 3.18 eV (389 nm). It corresponds to electronic excitation from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO). The CASPT2 calculations give a C

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

Table 1. Transition Configurations (TC),a Vertical Excitation Energies (ΔE, eV), Excitation Wavelengths (λ, nm), and Oscillator Strengths ( f) of FC Structures of A1, B4, and C4 Calculated at the TD-B3LYP/6-311G* and CASPT2/CASSCF Levels TD-B3LYP/6-311G* compound

state

A1

S0 S1 S2 S3 S4

B4

C4

S0 S1 S2 S3 S4 S0 S1 S2 S3 S4

a

TC

f

CASPT2/CASSCF ΔE (eV) (λ (nm)) 0.00 3.18 3.71 4.81 4.84

H → L (87%) H → L+1 (72%) H−1 → L+1 (50%) H−1 → L (37%) H → L+2 (27%)

0.022 0.174 0.137 0.126

H → L (97%) H → L+1 (70%) H−1 → L (57%) H−2 → L (40%)

0.038 0.007 0.080 0.155

H → L (87%)

0.021

0.00 3.56 4.11 4.28 4.45 0.00 3.09

H → L+1 (72%) H−1 → L (67%) H → L+2 (42%) H−2 → L (19%)

0.200 0.030 0.101

3.16 (393) 4.28 (290) 4.69 (264)

TC

ΔE (eV) (λ (nm))

f

0.00 2.91 3.42 4.26 4.52

(389) (334) (258) (256)

π6 π6 π5 π5 π6

→ → → → →

π7 (58%) π8 (62%) π8 (25%) π8 (15%) π10 (13%)

0.011 0.192 0.029 0.017

(348) (302) (290) (278)

π4 π4 π5 π3

→ → → →

π6 π7 π6 π6

(63%) (31%) (15%) (25%)

0.440 0.016 0.076 0.006

(401)

π6 π6 π6 π5 π4

→ → → → →

π8 π7 π7 π7 π7

(38%) (27%) (40%) (35%) (25%)

0.145

0.00 4.12 4.21 5.18 5.30 0.00 2.86

0.100 0.007 0.050

3.06 (405) 4.21 (294) 4.51 (275)

(427) (362) (291) (274)

(301) (294) (239) (234) (434)

H represents HOMO and L represents LUMO. The related orbitals are displayed in Table S2.

of H2 thermal elimination. The structure of TS2-S0 was shown in Figure 1f, in which the dihedral angle of ∠C2C7C8N is −24°. The distance between H1 atom and H2 atom is 2.72 Å in compound B4-S0, and it decreases to 1.04 Å in TS2-S0. Meanwhile, the H1 and H2 atoms leave away from the C1 and B atoms, respectively, and the bond length of C1−H1 elongates from 1.09 to1.50 Å whereas the bond length of B−H2 increases from 1.21 to1.44 Å during the transformation from B4-S0 to TS2-S0. The bond lengths of C1−C2, C1−B, and B−N shorten by 0.10, 0.04, and 0.09 Å from B4-S0 to TS2-S0. Meanwhile, the point group of the molecule changes from C1 to Cs for the reaction from B4-S0 to C4-S0, and the product C4-S0 is in planar geometry. The IRC calculations confirm that TS2-S0 is the transition state connecting B4-S0 and C4-S0 (Figure S4). The energy barrier to eliminate H2 from B4-S0 is 2.87 eV calculated at the B3LYP/6-311G* level. The energy of C4S0+H2 lies higher than that of B4-S0 by 0.67 eV, so the process from C4-S0 to B4-S0 is an endothermic reaction. From Figure 2, it can be seen that the energy barrier from B4-S0 to D1-S0 is 0.45 eV lower than that of the H2 elimination reaction, so the retraction ring reaction from B4-S0 to D1-S0 takes place more easily. Note that B4-S0 is much lower in energy than D1-S0 and C4-S0+H2, indicating that B4-S0 is very stable and both thermal reactions are reversible. This is different with other cases of B family, highlighting the importance of substitution groups.26 3.3. Photoelimination Pathways. The photoelimination reaction of B4 was investigated by the CASPT2/CASSCF method. All the optimized structures on S1 state were summarized in Figure 3. Upon electronic excitation, the structural changes in compound B4 are scanty: the C−C bond lengths in phenyl increase by 0.04, 0.03, 0.03, 0.05, 0.05, and 0.05 Å for C2−C3, C3−C4, C4−C5, C5−C6, C6−C7, and C2−C7 bonds from B4-S0 to B4-S1; The C−C bond lengths in pyridyl increase by 0.01 Å for C10−C11 and C11−C12 bonds, respectively; the B−N bond length increases from 1.63 to 1.66 Å. For the structure of A1, the bond lengths of C1−C2 and

to D1-S0 and from TS1-S0 to B4-S0 were confirmed by the IRC method (Figure S2). Furthermore, we have tried to optimize the transition state for the C1−B dissociation process from D1S0 but failed. To get some information about the reaction from D1-S0 to A1-S0+BH3, we have carried out constrained optimizations where the B−C1 distance was fixed at different values from 1.8 to 4.0 Å with an interval of 0.1 Å. The energy profile as a function of the C1−B distance was shown in Figure S3, which illustrates that the energy is rising with the C1−B bond elongation. The ground-state BH3 elimination reaction pathways for B4-S0 in toluene solvent were shown in Figure 2.

Figure 2. Thermal reaction energy profile for B4-S0 in toluene solution calculated at the B3LYP/6-311G* level.

The energies of TS1-S0 and D1-S0 are 2.42 and 1.27 eV with respect to B4-S0. Hence, the activation energy from B4-S0 to D1-S0 is 2.42 eV. The calculations show that D1-S0 is thermodynamically stable because the energy sum of A1-S0 and BH3 (2.49 eV) is higher than that of D1-S0. In addition, the total energy of A1-S0 and BH3 is a little higher than the energy of TS1-S0 by 0.07 eV. It means that the possibility of obtaining A1-S0 from B4-S0 by heating is slim. The second thermal elimination pathway for B4-S0 is H2 elimination. We found a transition-state TS2-S0 for the reaction D

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

C4-S1. The reaction pathways from TS2-S1 to B4-S1 and from TS2-S1 to C4-S1 were confirmed by the IRC method (Figure S7). The bond lengths of C1−C2, C1−B, and B−N are 1.50, 1.62, and 1.66 Å in B4-S1, whereas those in TS2-S1 are 1.44, 1.58, and 1.50 Å, which shorten by 0.06, 0.04, and 0.16 Å, respectively. Moreover, the activation energy from B4-S1 to TS2-S1 is just 0.02 eV, so it is very easy to perform the reaction from B4-S1 to C4-S1 on the condition of photoexcitation. Fourth, along with the cleavages of H1−C1 and H2−B bonds and the formation of the H1−H2 bond, phenyl and pyridyl turn into one plane, forming compound C4-S 1 displaying Cs symmetry. The relative energy of C4-S1+H2 is 3.33 eV, which is 0.44 eV lower than the TS2-S1. Finally, the excited-state C4S1 returns to the ground-state C4-S0 in the form of fluorescence. All in all, the H2 elimination reaction is very easy to take place on the condition of irradiation at about 300 nm, which is similar to the experimental results replacing the R group with mesityl and replacing X group with hydrogen atom.25 On the condition of heating, it prefers to conduct the reaction from B4-S0 to D1-S0, and on the condition of irradiation, the elimination reaction from B4-S0 to C4-S0 will take place first. Hence, this competitive thermal elimination/ photoelimination reaction has excellent selectivity, which plays a vital role in the synthesis of fluorescent material. To get more insight into the photochemistry of compound B4, we have investigated the CIs between two adjacent states S0 and S1, which may play key roles in the radiationless decay. Three CIs ((S1/S0)x-1, (S1/S0)x-2, and (S1/S0)x-3) have been located. Figure 5 shows the structures and the two key orbitals

Figure 3. Optimized molecular structures on the S1 state. Critical bond lengths are given. Yellow, purple, pink, and blue atoms represent C, N, B, and H atoms, respectively.

C1−N on S1 state are 1.41 and 1.40 Å, which are a little longer than those on the ground state (1.40 and 1.37 Å). For the direction of BH3 elimination, we have located a transition-state TS1-S1 that connects compounds B4-S1 and D1-S1 (see the IRC profile in Figure S5), followed by the conversion of D1-S1 to A-S1. At the structure of TS1-S1, the H1−B bond length is 1.28 Å, which is shorter than that of TS1-S0 (1.30 Å). The vertical excitation energy of B4-S0 calculated by CASPT2 method is 4.12 eV (301 nm), which is 2.62 eV lower than the activation energy of TS1-S1 (6.74 eV). Hence, the BH3 elimination pathways on the S1 state can be ruled out because the barrier is too high to overcome upon the photoexcitation of 300 nm. Concerning the photoelimination reaction of H2, the elimination mechanism can be summarized as the following steps (Figure 4). First, compound B4 is excited to the S1 state

Figure 4. Photoelimination reaction energy profile on the first excited state for B4 in toluene solution calculated at the CASPT2//CASSCF/ 6-31G* level.

Figure 5. Structures of TS2-S1, (S1/S0)x-1, (S1/S0)x-2, (S1/S0)x-3, and the two main orbitals involved in the excitations with occupation numbers (a/b, a for S1 state, b for S0 state) in brackets.

in the FC region, accompanying by the intramolecular electron transfer from occupied orbital of phenyl to the unoccupied orbital of pyridyl. Second, the excited-state compound B4 rapidly reaches the most stable structure B4-S1 with the little change of structure. The energy profile along the linear interpolation internal coordinates from B4-S0 to B4-S1 was listed in Figure S6. We can see that B4-S1 is 0.37 eV lower than the vertical excitation energy, and there is no barrier between the FC structure and B4-S1 on S1 surface. Third, the elongations of C1−H1 and B−H2 bond lengths accompanied by the decrease of the distance between H1 and H2 yield the transition-state TS2-S1 (Figure 3f), which connects B4-S1 and

involved in the electron excitation (orbital 1 and orbital 2) of TS2-S1, (S1/S0)x-1, (S1/S0)x-2, and (S1/S0)x-3. The structures of (S1/S0)x-2 and (S1/S0)x-3 are characterized by the out-ofplane bending of N or one C in the pyridyl or phenyl ring, which are similar to the half-boat shaped intersection in benzene.63 These structural changes with respect to B4-S0 can be rationalized with the help of the orbitals involved in the excitation, where the two orbitals in question are localized in the pyridyl ring and phenyl ring for (S1/S0)x-2 and (S1/S0)x-3, E

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

stationary points on S0 state were calculated by the IRC method. The results show that in one direction, the (S1/S0)x-1 decays to B4-S0 (Figure S9), and in the other direction, (S1/ S0)x-1 decays to products H2 and C4-S0 (Figure S10). Thus, radiationless decay through (S1/S0)x-1 also is a possible reaction process. In summary, on the FC point, there are two possible reaction paths. The first pathway is that the excited-state molecule goes through (S1/S0)x-1, and either returns to B4-S0 or arrives at C4S0. The second pathway is getting through TS2-S1 to reach C4S1 and then decaying to C4-S0 by the way of fluorescence. 3.4. Discussion. The mechanistic picture of thermal elimination/photoelimination reactions of B4 was summarized in Figure 6. On the ground state, the energy barrier from B4-S0 to A1-S0 is lower than that from B4-S0 to C4-S0, so the retraction ring reaction is expected to proceed with larger probability than the H2 elimination upon heating. The high energy of A1-S0+BH3 is due to the well-known instability of BH3, which does not exist in the natural world but easily forms B2H6. If BH3 is continuously removed from the reaction medium, A1-S0 might be obtained by heating B4-S0. This strategy has been successfully employed to produce A1 from B1 experimentally.26 Alternatively, our calculations propose species D1-S0 as the key intermediate in the BH3 elimination process from B4-S0 to A1-S0, which acts as a bridge or a depot between the reactant and the product. NMR spectroscopy data have proved the existence of D1 analogues in the adduct reactions of A1 with highly Lewis acidic boranes including HB(C6F5)2 and B(C6F5)3.26 On the S1 state, the energy barrier for the BH3 elimination reaction from B4-S1 to A1-S1 is up to 3 eV, which is too high to traverse upon normal UV/vis excitation. On the contrary, the H2 elimination is likely to take place either bypassing transitionstate TS2-S1 or going through a conical intersection point (S1/ S0)x-1 between the S1 state and the S0 state. The pathway via TS2-S1 is almost barrierless, resulting in the product C4 on the S1 state. According to our calculations, B4 is not fluorescent in the visible region because the energy gap between S1 and S0 at the B4-S1 geometry is 3.65 eV (340 nm) and more importantly it can easily convert to C4 on the S1 state, which gives off bright green fluorescence (539 nm, f = 0.026). This is consistent with the experimental observations of color changing from the colorless precursors in the B family to the bright green or yellow-green fluorescent products of C compounds in solution and in the solid state.25,26 As for the other reaction pathway via (S1/S0)x-1, the possibility is minor owing to the sizable barrier of 0.3 eV compared with the almost barrierless reaction channel via TS2-S1 mentioned above. This explains the apparent paradox that delayed fluorescence is observed experimentally in conjunction with energetically accessible conical intersections. Meanwhile, this does not weaken the important roles of conical intersections in photochemistry, which have been emphasized by recent experimental and theoretical work.67−69 It has been established experimentally that the thermal eliminations and photoeliminations of BN-heterocycles are competitive depending on the substitution groups on the B atom and the chelate backbone.26 Recent work reveals that both electronic and steric effects have significant and distinct impact on the photoreactivity of BN-heterocycles, where electron-donating bulky aryl rings on the B atom enhance the photoelimination efficiency while electron-withdrawing groups (bulky or not) on the B atom exclusively lead to photoisomerization to a dark colored BN-1,3,5-cycloocatriene or

respectively (Figure 5). The calculated energies of (S1/S0)x-1, (S1/S0)x-2, and (S1/S0)x-3 are 3.56, 4.31, and 4.50 eV, respectively. Note that the energies of (S1/S0)x-2 and (S1/ S0)x-3 are much higher than that of the vertical excitation energy at the FC structure of B4 (4.12 eV). These two CIs cannot be reached by the excitation of 300 nm, so we focused on the property and reactivity of (S1/S0)x-1 hereafter. The structure of (S1/S0)x-1 resembles that of TS2-S1, and there are only slight differences in bond lengths and dihedral angles. In detail, the C1−C2, C1−B, and B−N bonds are 1.49, 1.62, and 1.48 Å in (S1/S0)x-1, and those in TS2-S1 are 1.44, 1.58, and 1.50 Å, respectively. The dihedral angle of ∠C2C7C8N is 14° in (S1/S0)x-1, and it is −19° in TS2-S1. Meanwhile, it can be seen that the two main orbitals involved in the excitation of (S1/S0)x-1 are similar to those of TS2-S1 (Figure 5). In orbital 1, electron population locates on B, C1, C3, C5, and C7 atoms, and in orbital 2, the electron population mainly locates on pyridyl ring. The energy of (S1/S0)x-1 is 3.56 eV, which is 0.19 eV lower than the energy of B4-S1. In short, the structures and electronic populations of (S1/S0)x-1 and TS2-S1 are similar to each other. Hence, these two structures locate closely on the first excited-state potential energy surface (Figure 6).

Figure 6. Potential energy surfaces related with the thermal elimination/photoelimination reactions of B4 in toluene solution together with the relative energies (in parentheses, in eV) of the key points.

The CIs are characterized by two vectors (hypersphere geometry) that realize the radiationless decay: the gradient difference (GD) and derivative coupling (DC) vectors,64−66 which constitute the crossing seam and determine the directions when molecules pass through the confluences. The topography of the seam has two types of shape: sloped and peaked CIs. Passing through a sloped CI leads to a structurally similar stationary point, while passing through a peaked CI results in different stationary points.64 Figure 6 illustrates the path from the minimum of B4-S1 to (S1/S0)x-1 (imaginary line in pink), together with the decay paths from (S1/S0)x-1 to different stationary points (imaginary line in blue). The pathway from B4-S1 to (S1/S0)x-1 characterized by linear interpolations internal coordinates (Figure S8) gives an estimated barrier of 0.31 eV relative to B4-S1. This estimated barrier is the upper bound to the optimized barrier. Moreover, the excess vibrational energy of B4 on S1 state after excitation (difference between the vertical excitation energy and the energy of B4-S1) is 0.37 eV, higher than the estimated barrier of 0.31 eV. Thus, the system has enough energy to conquer the barrier after excitation, and the decay to (S1/S0)x-1 is energetically possible. The decay paths from (S1/S0)x-1 to F

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A further to BN-1,3,6-cycloocatriene with high stereoselectivity.70 Our present work focused on the reaction mechanisms of B4, which is the simplest BN-heterocycle in Scheme 1 and has not been explicitly explored experimentally on its thermal elimination/photoelimination, thus providing meaningful instructions for future experimental work. To get a general mechanistic picture on all BN-heterocycles, the subtle differences on the S1 potential energy surface responsible for the distinct excite-state relaxations should be considered. Recent electronic structure calculations and nonadiabatic dynamics simulations reveal the nearly stereospecific unidirectional excited-state relaxation in the photoisomerization of arylazopyrazole.71 Therefore, ongoing theoretical work on BNheterocycles in our group pays attention to not only the electronic and steric effects but also the dynamics effects.

ACKNOWLEDGMENTS

■

REFERENCES

(1) Zhang, G.; Palmer, G. M.; Dewhirst, M. W.; Fraser, C. L. A DualEmissive-Materials Design Concept Enables Tumour Hypoxia Imaging. Nat. Mater. 2009, 8, 747−751. (2) Saito, S.; Matsuo, K.; Yamaguchi, S. Polycyclic π-Electron System with Boron at Its Center. J. Am. Chem. Soc. 2012, 134, 9130−9133. (3) Wade, C. R.; Broomsgrove, A. E.; Aldridge, S.; Gabbai, F. P. Fluoride Ion Complexation and Sensing Using Organoboron Compounds. Chem. Rev. 2010, 110, 3958−3984. (4) Shirota, Y.; Kageyama, H. Charge Carrier Transporting Molecular Materials and Their Applications in Devices. Chem. Rev. 2007, 107, 953−1010. (5) Jakle, F. Advances in the Synthesis of Organoborane Polymers for Optical, Electronic, and Sensory Applications. Chem. Rev. 2010, 110, 3985−4022. (6) Dou, C.; Saito, S.; Matsuo, K.; Hisaki, I.; Yamaguchi, S. A BoronContaining PAH as a Substructure of Boron-Doped Graphene. Angew. Chem., Int. Ed. 2012, 51, 12206−12210. (7) Entwistle, C. D.; Marder, T. B. Boron Chemistry Lights the Way: Optical Properties of Molecular and Polymeric Systems. Angew. Chem., Int. Ed. 2002, 41, 2927−2931. (8) Entwistle, C. D.; Marder, T. B. Applications of Three-Coordinate Organoboron Compounds and Polymers in Optoelectronics. Chem. Mater. 2004, 16, 4574−4585. (9) Chen, P.; Lalancette, R. A.; Jakle, F. π-Expanded Borazine: An Ambipolar Conjugated B-π-N Macrocycle. Angew. Chem., Int. Ed. 2012, 51, 7994−7998. (10) Wakamiya, A.; Taniguchi, T.; Yamaguchi, S. Intramolecular B-N Coordination as a Scaffold for Electron-Transporting Materials: Synthesis and Properties of Boryl-Substituted Thienylthiazoles. Angew. Chem., Int. Ed. 2006, 45, 3170−3173. (11) Lorbach, A.; Hubner, A.; Wagner, M. Aryl(Hydro)Boranes: Versatile Building Blocks for Boron-Doped π-Electron Materials. Dalton. Trans. 2012, 41, 6048−6063. (12) Rao, Y.-L.; Amarne, H.; Zhao, S.-B.; McCormick, T. M.; Martic, S.; Sun, Y.; Wang, R.-Y.; Wang, S. Reversible Intramolecular C-C Bond Formation/Breaking and Color Switching Mediated by a N,C-Chelate in (2-Ph-Py)BMes2 and (5-BMes2-2-Ph-Py)BMes2. J. Am. Chem. Soc. 2008, 130, 12898−12900. (13) Nagura, K.; Saito, S.; Frohlich, R.; Glorius, F.; Yamaguchi, S. NHeterocyclic Carbene Boranes as Electron-Donating and ElectronAccepting Components of π-Conjugated Systems. Angew. Chem., Int. Ed. 2012, 51, 7762−7766. (14) Rao, Y.-L.; Amarne, H.; Wang, S. Photochromic FourCoordinate N,C-Chelate Boron Compounds. Coord. Chem. Rev. 2012, 256, 759−770. (15) Sun, Y.; Ross, N.; Zhao, S.-B.; Huszarik, K.; Jia, W.-L.; Wang, R.Y.; Macartney, D.; Wang, S. Enhancing Electron Accepting Ability of Triarylboron via π-Conjugation with 2,2′-Bipy and Metal Chelation: 5,5′-Bis(BMes2)-2,2′-Bipy and Its Metal Complexes. J. Am. Chem. Soc. 2007, 129, 7510−7511. (16) Wang, J.-Y.; Pei, J. BN-Embedded Aromatics for Optoelectronic Applications. Chin. Chem. Lett. 2016, 27, 1139−1146. (17) Shi, Y.-G.; Yang, D.-T.; Mellerup, S. K.; Wang, N.; Peng, T.; Wang, S. 1,1-Hydroboration of Fused Azole-Isoindole Analogues as an Approach for Construction of B,N-Heterocycles and Azole-Fused B,NNaphthalenes. Org. Lett. 2016, 18, 1626−1629. (18) Yuan, K.; Suzuki, N.; Mellerup, S. K.; Wang, X.; Yamaguchi, S.; Wang, S. Pyridyl Directed Catalyst-Free Trans-Hydroboration of Internal Alkynes. Org. Lett. 2016, 18, 720−723. (19) Agou, T.; Kobayashi, J.; Kawashima, T. Syntheses, Structure, and Optical Properties of Ladder-Type Fused Azaborines. Org. Lett. 2006, 8, 2241−2244.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpca.6b09495. Molecular orbitals included in the (12, 12) active space, energies and Cartesian coordinates of the critical points, and energy profiles of reaction processes (PDF)

■

■

This work is supported by the National Natural Science Foundation of China (21303007, 21473010).

4. CONCLUSIONS The thermal elimination/photoelimination reactions of B4 were investigated by studying the reaction paths on the ground and the first excited state. On the ground state, the energy barrier from B4-S0 to D1-S0 is 2.42 eV, which is lower than the energy barrier from B4-S0 to C4-S0 (2.87 eV). Hence, on the condition of heating, the retraction ring reaction prefers to take place. The photoelimination reaction starts with spectroscopic state S1 when the FC point of B4 absorbs light of 301 nm. On the first excited state, the reaction energy barrier (6.74 eV) from B4-S1 to D1-S1 is too high to surmount, whereas the energy barrier from B4-S1 to C4-S1 (0.02 eV) is negligible. Therefore, under the condition of photoexcitation, the reaction from B4-S1 to C4-S1 is likely to occur. Moreover, three CIs between S0 and S1 states of compound B4 have been found. Getting through one of three CIs, (S1/S0)x-1, the decay of excited-state B4 leads to either B4-S0 or C4-S0. The reactions from B4-S0 to D1-S0 and from B4-S0 to C4-S0 are two competitive progresses upon heating and irradiation. These two reactions have excellent selectivity: on the heating condition, the retraction ring reaction can take place, and on the irradiant condition, the H2 elimination reaction is likely to occur. The present work not only provides a clear mechanism for the thermal elimination/photoelimination reactions of B4 but also contributes to the fundamental understanding of the photochemistry of B,N-heterocycles.

■

Article

AUTHOR INFORMATION

Corresponding Authors

*Q. S. Li. E-mail: [email protected]. Phone: +86 10 68911354. *Z. S. Li. E-mail: [email protected]. Phone: +86 10 68918670. ORCID

Ze-Sheng Li: 0000-0002-6993-8414 Notes

The authors declare no competing financial interest. G

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

(41) Becke, A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648−5652. (42) Becke, A. D. A New Mixing of Hartree−Fock and Local Density-Functional Theories. J. Chem. Phys. 1993, 98, 1372−1377. (43) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle-Salvetti Correlation-Energy Formula into a Functional of the Electron Density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. (44) Grimme, S. Accurate Description of Van Der Waals Complexes by Density Functional Theory Including Empirical Corrections. J. Comput. Chem. 2004, 25, 1463−1473. (45) Grimme, S. Semiempirical GGA-Type Density Functional Constructed with a Long-Range Dispersion Correction. J. Comput. Chem. 2006, 27, 1787−1799. (46) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132, 154104. (47) Goerigk, L.; Grimme, S. A Thorough Benchmark of Density Functional Methods for General Main Group Thermochemistry, Kinetics, and Noncovalent Interactions. Phys. Chem. Chem. Phys. 2011, 13, 6670−6688. (48) Gonzalez, C.; Schlegel, H. B. Reaction Path Following in MassWeighted Internal Coordinates. J. Phys. Chem. 1990, 94, 5523−5527. (49) Roos, B. O.; Taylor, P. R.; Siegbahn, P. E. M. A Complete Active Space SCF Method (CASSCF) Using a Density Matrix Formulated Super-CI Approach. Chem. Phys. 1980, 48, 157−173. (50) Ruedenberg, K.; Schmidt, M. W.; Gilbert, M. M.; Elbert, S. T. Are Atoms Intrinsic to Molecular Electronic Wavefunctions? I. The Fors Model. Chem. Phys. 1982, 71, 41−49. (51) Werner, H.-J.; Knowles, P. J. A Second Order Multiconfiguration SCF Procedure with Optimum Convergence. J. Chem. Phys. 1985, 82, 5053−5063. (52) Celani, P.; Robb, M. A.; Garavelli, M.; Bernardi, F.; Olivucci, M. Geometry Optimisation on a Hypersphere. Application to Finding Reaction Paths from a Conical Intersection. Chem. Phys. Lett. 1995, 243, 1−8. (53) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; et al. Gaussian 03, Revision E.01; Gaussian, Inc.: Wallingford, CT, 2004. (54) Andersson, K.; Malmqvist, P.-Å.; Roos, B. O.; Sadlej, A. J.; Wolinski, K. Second-Order Perturbation Theory with a CASSCF Reference Function. J. Phys. Chem. 1990, 94, 5483−5488. (55) Andersson, K.; Malmqvist, P.-Å.; Roos, B. O. Second-Order Perturbation Theory with a Complete Active Space Self-Consistent Field Reference Function. J. Chem. Phys. 1992, 96, 1218−1226. (56) Pierloot, K.; Dumez, B.; Widmark, P.-O.; Roos, B. O. Density Matrix Averaged Atomic Natural Orbital (ANO) Basis Sets for Correlated Molecular Wave Functions. Theor. Chim. Acta 1995, 90, 87−114. (57) Aquilante, F.; De Vico, L.; Ferre, N.; Ghigo, G.; Malmqvist, P.Å.; Neogrady, P.; Pedersen, T. B.; Pitonak, M.; Reiher, M.; Roos, B. O.; et al. Molcas 7: The Next Generation. J. Comput. Chem. 2010, 31, 224−247. (58) Forsberg, N.; Malmqvist, P.-Å. Multiconfiguration Perturbation Theory with Imaginary Level Shift. Chem. Phys. Lett. 1997, 274, 196− 204. (59) Ghigo, G.; Roos, B. O.; Malmqvist, P.-Å. A Modified Definition of the Zeroth-Order Hamiltonian in Multiconfigurational Perturbation Theory (CASPT2). Chem. Phys. Lett. 2004, 396, 142−149. (60) Barone, V.; Cossi, M. Quantum Calculation of Molecular Energies and Energy Gradients in Solution by a Conductor Solvent Model. J. Phys. Chem. A 1998, 102, 1995−2001. (61) Cossi, M.; Barone, V.; Robb, M. A. A Direct Procedure for the Evaluation of Solvent Effects in MC-SCF Calculations. J. Chem. Phys. 1999, 111, 5295−5302. (62) Cossi, M.; Rega, N.; Scalmani, G.; Barone, V. Polarizable Dielectric Model of Solvation with Inclusion of Charge Penetration Effects. J. Chem. Phys. 2001, 114, 5691−5701.

(20) Wang, S.; Yang, D.-T.; Lu, J.; Shimogawa, H.; Gong, S.; Wang, X.; Mellerup, S. K.; Wakamiya, A.; Chang, Y.-L.; Yang, C.; et al. In Situ Solid-State Generation of (BN)2-Pyrenes and Electroluminescent Devices. Angew. Chem., Int. Ed. 2015, 54, 15074−15078. (21) Neue, B.; Araneda, J. F.; Piers, W. E.; Parvez, M. BNDibenzo[a,o]picenes: Analogues of an Unknown Polycyclic Aromatic Hydrocarbon. Angew. Chem., Int. Ed. 2013, 52, 9966−9969. (22) Wang, X.-Y.; Lin, H.-R.; Lei, T.; Yang, D.-C.; Zhuang, F.-D.; Wang, J.-Y.; Yuan, S.-C.; Pei, J. Azaborine Compounds for Organic Field-Effect Transistors: Efficient Synthesis, Remarkable Stability, and BN Dipole Interactions. Angew. Chem., Int. Ed. 2013, 52, 3117−3120. (23) Bosdet, M. J. D.; Piers, W. E. B-N as a C-C Substitute in Aromatic Systems. Can. J. Chem. 2009, 87, 8−29. (24) Ko, S.-B.; Lu, J.-S.; Wang, S. Chelation-Assisted Photoelimination of B,N-Heterocycles. Org. Lett. 2014, 16, 616−619. (25) Lu, J.-S.; Ko, S.-B.; Walters, N. R.; Kang, Y.; Sauriol, F.; Wang, S. Formation of Azaborines by Photoelimination of B,N-Heterocyclic Compounds. Angew. Chem., Int. Ed. 2013, 52, 4544−4548. (26) Yang, D.-T.; Mellerup, S. K.; Wang, X.; Lu, J.-S.; Wang, S. Reversible 1,1-Hydroboration: Boryl Insertion into a C-N Bond and Competitive Elimination of HBR2 or R-H. Angew. Chem., Int. Ed. 2015, 54, 5498−5501. (27) McDonald, S. M.; Mellerup, S. K.; Peng, J.; Yang, D.; Li, Q.-S.; Wang, S. Thermal and Photolytic Transformation of NHC-B,NHeterocycles: Controlled Generation of Blue Fluorescent 1,3Azaborinine Derivatives and 1H-imidazo[1,2-a]indoles by External Stimuli. Chem. - Eur. J. 2015, 21, 13961−13970. (28) Campbell, P. G.; Marwitz, A. J.; Liu, S. Y. Recent Advances in Azaborine Chemistry. Angew. Chem., Int. Ed. 2012, 51, 6074−6092. (29) Adamo, C.; Jacquemin, D. The Calculations of Excited-State Properties with Time-Dependent Density Functional Theory. Chem. Soc. Rev. 2013, 42, 845−856. (30) Bernardi, F.; Olivucci, M.; Robb, M. A. Potential Energy Surface Crossings in Organic Photochemistry. Chem. Soc. Rev. 1996, 25, 321− 328. (31) Yarkony, D. R. Conical Intersections: The New Conventional Wisdom. J. Phys. Chem. A 2001, 105, 6277−6293. (32) Blancafort, L.; Robb, M. A. A Valence Bond Description of the Prefulvene Extended Conical Intersection Seam of Benzene. J. Chem. Theory Comput. 2012, 8, 4922−4930. (33) Ruiz-Barragan, S.; Blancafort, L. Photophysics of Fulvene under the Non-Resonant Stark Effect. Shaping the Conical Intersection Seam. Faraday Discuss. 2013, 163, 497−512. (34) Conti, I.; Garavelli, M.; Orlandi, G. The Different Photoisomerization Efficiency of Azobenzene in the Lowest nπ* and ππ* Singlets: The Role of a Phantom State. J. Am. Chem. Soc. 2008, 130, 5216−5230. (35) Improta, R.; Santoro, F.; Blancafort, L. Quantum Mechanical Studies on the Photophysics and the Photochemistry of Nucleic Acids and Nucleobases. Chem. Rev. 2016, 116, 3540−3593. (36) Li, Q.-S.; Migani, A.; Blancafort, L. Wave Packet Dynamics at an Extended Seam of Conical Intersection: Mechanism of the LightInduced Wolff Rearrangement. J. Phys. Chem. Lett. 2012, 3, 1056− 1061. (37) Cui, G.; Thiel, W. Photoinduced Ultrafast Wolff Rearrangement: A Non-Adiabatic Dynamics Perspective. Angew. Chem., Int. Ed. 2013, 52, 433−436. (38) Li, Q.-S.; Blancafort, L. A Conical Intersection Model to Explain Aggregation Induced Emission in Diphenyl Dibenzofulvene. Chem. Commun. 2013, 49, 5966−5968. (39) Peng, X.-L.; Ruiz-Barragan, S.; Li, Z.-S.; Li, Q.-S.; Blancafort, L. Restricted Access to a Conical Intersection to Explain Aggregation Induced Emission in Dimethyl Tetraphenylsilole. J. Mater. Chem. C 2016, 4, 2802−2810. (40) Migani, A.; Blancafort, L. Excitonic Interfacial Proton-Coupled Electron Transfer Mechanism in the Photocatalytic Oxidation of Methanol to Formaldehyde on TiO2(110). J. Am. Chem. Soc. 2016, 138, 16165−16173. H

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A (63) Palmer, I. J.; Ragazos, I. N.; Bernardi, F.; Olivucci, M.; Robb, M. A. An MC-SCF Study of the S1 and S2 Photochemical Reactions of Benzene. J. Am. Chem. Soc. 1993, 115, 673−682. (64) Blancafort, L. Photochemistry and Photophysics at Extended Seams of Conical Intersection. ChemPhysChem 2014, 15, 3166−3181. (65) Atchity, G. J.; Xantheas, S. S.; Ruedenberg, K. Potential Energy Surfaces near Intersections. J. Chem. Phys. 1991, 95, 1862−1876. (66) Teller, E. The Crossing of Potential Surfaces. J. Phys. Chem. 1937, 41, 109−116. (67) Polli, D.; Altoe, P.; Weingart, O.; Spillane, K. M.; Manzoni, C.; Brida, D.; Tomasello, G.; Orlandi, G.; Kukura, P.; Mathies, R. A.; et al. Conical Intersection Dynamics of the Primary Photoisomerization Event in Vision. Nature 2010, 467, 440−443. (68) Li, Q.-S.; Giussani, A.; Segarra-Martí, J.; Nenov, A.; Rivalta, I.; Voityuk, A. A.; Mukamel, S.; Roca-Sanjuán, D.; Garavelli, M.; Blancafort, L. Multiple Decay Mechanisms and 2D-UV Spectroscopic Fingerprints of Singlet Excited Solvated Adenine-Uracil Monophosphate. Chem. - Eur. J. 2016, 22, 7497−7507. (69) Li, H.-J.; Migani, A.; Blancafort, L.; Li, Q.-S.; Li, Z.-S. Early Events in the Photochemistry of 5-Diazo Meldrum’s Acid: Formation of a Product Manifold in C-N Bound and Pre-Dissociated Intersection Seam Regions. Phys. Chem. Chem. Phys. 2016, 18, 30785−30793. (70) Yang, D.-T.; Mellerup, S. K.; Peng, J.-B.; Wang, X.; Li, Q.-S.; Wang, S. Substituent Directed Phototransformations of BN-Heterocycles: Elimination vs Isomerization via Selective B−C Bond Cleavage. J. Am. Chem. Soc. 2016, 138, 11513−11516. (71) Wang, Y.-T.; Liu, X.-Y.; Cui, G.; Fang, W.-H.; Thiel, W. Photoisomerization of Arylazopyrazole Photoswitches: Stereospecific Excited-State Relaxation. Angew. Chem., Int. Ed. 2016, 55, 14009− 14013.

I

DOI: 10.1021/acs.jpca.6b09495 J. Phys. Chem. A XXXX, XXX, XXX−XXX