Energetics of Baird aromaticity supported by inversion of photoexcited chiral [4n]annulene derivatives

For the concept of aromaticity, energetic quantification is crucial. However, this has been elusive for excited-state (Baird) aromaticity. Here we report our serendipitous discovery of two nonplanar thiophene-fused chiral [4n]annulenes Th4 COT Saddle and Th6 CDH Screw, which by computational analysis turned out to be a pair of molecules suitable for energetic quantification of Baird aromaticity. Their enantiomers were separable chromatographically but racemized thermally, enabling investigation of the ring inversion kinetics. In contrast to Th6 CDH Screw, which inverts through a nonplanar transition state, the inversion of Th4 COT Saddle, progressing through a planar transition state, was remarkably accelerated upon photoexcitation. As predicted by Baird’s theory, the planar conformation of Th4 COT Saddle is stabilized in the photoexcited state, thereby enabling lower activation enthalpy than that in the ground state. The lowering of the activation enthalpy, i.e., the energetic impact of excited-state aromaticity, was quantified experimentally to be as high as 21–22 kcal mol–1.


β-linked thiophene dimer (Th2)
To a THF solution (200 mL) of 4-bromo-2-methylthiophene (10.0 g, 56.5 mmol) was added a hexane solution of tert-butyllithium (37.3 mL, 60.8 mmol) at -78 °C dropwise under Ar, and the mixture was stirred at -78 °C for 1 h. B(OMe) 3 (9.48 g, 91.3 mmol) was slowly added to this reaction mixture, and the resultant mixture was stirred at -78 °C for 1 h followed by 25 °C for 12 h. 4-Bromo-2-methylthiophene (10.0 g, 56.5 mmol), Pd(PPh 3 ) 4 (1.76 g, 1.52 mmol), and 2 M aqueous Na 2 CO 3 (65 mL) were successively added to this reaction mixture, and the resultant mixture was refluxed for 12 h under Ar. Then, the reaction mixture was poured into aqueous NH 4 Cl, and the separated aqueous layer was extracted 3 times with EtOAc. The combined organic extract was washed with brine, dried over Na 2 SO 4 , and evaporated to dryness under reduced pressure. The residue was subjected to column chromatography on silica gel using hexane/CH 2 Cl 2 (10/1 v/v) as eluent to allow isolation of Th2 as a white solid (8.71 g, 44.8 mmol) in 79% yield. 1  calcd. for C 10 H 10 S 2 194.022).

Th4 COT Saddle and Th6 CDH Screw
To a THF solution (120 mL) of Th2 (1.94 g, 10 mmol) was added a hexane solution of butyllithium (9.1 mL, 24 mmol) at -78 °C dropwise under Ar, and the mixture was stirred at -78 °C for 1h and slowly warmed up to reach 25 °C in 1 h. The reaction mixture was added to a flask with anhydrous CuBr 2 (7.95 g, 35.6 mmol), and the mixture was stirred for 12 h under Ar. Then, the reaction mixture was poured into a 30% aqueous ammonia solution, and the separated aqueous layer was extracted 5 times with EtOAc. The combined organic extract was washed with brine, dried over Na 2 SO 4 , and evaporated to dryness under reduced pressure. The residue was subjected to column chromatography on silica gel using hexane/EtOAc (10/1 v/v), followed by recycling size exclusion chromatography (SEC). The fifth fraction of SEC was collected and recrystallized from CH 2 Cl 2 /hexane to give a racemic mixture of Th4 COT Saddle as a white solid (319 mg, 0.83 mmol) in 17% yield. The third fraction of SEC was collected and recrystallized from THF/hexane to give a racemic mixture of Th6 CDH Screw as a yellow solid (119 mg, 0.21 mmol) in 6.2% yield.

Supplementary Figure 3 | 1 H NMR spectrum of Th4 COT Saddle in CDCl 3 at 25 °C.
Asterisked signals at δ 7.26 and 1.54 ppm are due to partially non-deuterated residues of CDCl 3 and water, respectively.

Supplementary Figure 4 | 13 C NMR spectrum of Th4 COT Saddle in CDCl 3 at 25 °C.
Asterisked signal at δ 77.2 ppm is due to CDCl 3 .

Supplementary Figure 5 | 1 H NMR spectrum of Th6 CDH Screw in CDCl 3 at 25 °C.
Asterisked signals at δ 7.26 and 1.54 ppm are due to partially non-deuterated residues of CDCl 3 and water, respectively.

Supplementary Figure 6 | 13 C NMR spectrum of Th6 CDH Screw in CDCl 3 at 25 °C.
Asterisked signal at δ 77.2 ppm is due to CDCl 3 .

X-ray Crystallography of Th4 COT Saddle and Th6 CDH Screw
Crystals of Th4 COT Saddle and Th6 CDH Screw suitable for X-ray analysis were obtained by slow diffusion of ethanol into a CHCl 3 solution of Th4 COT Saddle , slow diffusion of hexane into a THF solution of Th6 CDH Screw , respectively. CCDC 1526430 and CCDC 1526429 contain the supplementary data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.ac.uk/data_request-cif.  (13) 75.213 (14) 69.559 (12) 882.4 (

Correction Methods for Kinetic Studies under Photoirradiation
All the kinetic studies for the chiral inversion processes under photoexcitation conditions were conducted at a temperature where the decay profiles of the CD intensity were negligibly small without photoirradiation. Since the CD spectra cannot be measured during photoirradiation, we alternately performed irradiation (250 s, red vertical lines in Supplementary Fig. 7a) and CD spectroscopic measurement (50 s, black dots are the CD intensity at a certain wavelength). Under the irradiation conditions, we can extrapolate the time-dependent CD intensity decay profiles drawn as grey solid lines in Supplementary Fig. 7a.
When performing kinetic studies using the Eyring equation, we corrected the experimental time to be the total irradiation time and used the plot for kinetic analysis ( Supplementary Fig. 7b). Ti:sapphire regenerative amplifier system (Integra-C, Quantronix) operating at 1 kHz repetition rate and an optical detection system. The generated OPA pulses, which were used as pump pulses, had a pulse width of ~ 100 fs and an average power of 100 mW in the range 280-2700 nm. White light continuum (WLC) probe pulses were generated using a sapphire window (

Determination of the Lifetimes of the Excited Species
The lifetime of Th4 COT Saddle in the S 1 state was determined to be 5.5 ps by fs-TAS ( Supplementary Fig. 16  The S 1 transition state was optimized at the CASSCF level (where it is the third root due to the lack of dynamic correlation) using the saddle method with the keyword RP-Coordinates 24 .
The approximate Hessian generated from this procedure displays two negative frequencies, one corresponding to the inversion coordinate and one associated with methyl rotation. Because no dispersion forces are accounted for in CASSCF, the barrier for methyl rotation is usually very small; therefore, negative frequencies corresponding to these rotations often occur. As this negative frequency is not important to the inversion process, it can be disregarded for the purposes of transition state structure validation.    25 . This good agreement between the theory and experiment in the case of the parent COT and only slight overestimation in the case of Th4 COT Saddle makes us confident in the computational data for these compounds.  25 12.7 ± 0.5

S 1 Transition State for Th4 COT Saddle at the CASSCF(8in8) Level
To verify the validity of the calculated TD-DFT S 1

Ring Inversion Kinetic Studies
The observed decay profiles of Th4 COT Saddle and Th6 CDH Screw both satisfied first-order kinetics, where the rate constants of racemization (k rac ) and ring inversion (k inv ) were obtained from Supplementary Equation (1): The half-life (τ 1/2 ) of the CD intensity was obtained from Supplementary Equation (2): The k inv values obtained at various temperatures were plotted and analysed using the Eyring equation, Supplementary Equation (3): where T, ΔΗ ‡ , R, k B , h and ΔS ‡ are the absolute temperature, activation enthalpy, gas constant, Boltzmann constant, Planck's constant and activation entropy, respectively. By plotting ln(k inv /T) versus 1/T, we obtained ΔH ‡ and ΔS ‡ from the slope and y-intercept, respectively.
The Eyring plots for the ring inversion of Th4 COT Saddle and Th6 CDH Screw in the dark are shown in Supplementary Fig. 17.

Calculation of k inv under Photoirradiation
The calculation of inversion kinetics under photoirradiation requires special attention.
First, one must consider the measurement time. Since the CD intensity decays only under photoexcitation conditions, the measurement time was corrected according to the procedure shown in Supplementary Fig. 11. The second factor to consider is the rate constant (k inv ).
Because photoirradiation cannot excite all the molecules in the system simultaneously, the observed rate constant (k obs ) is not an intrinsic value of the molecule (k inv ) but an apparent value.
Thus, k obs satisfies the following Supplementary Equation (4): Since the experimental set-ups for photoirradiation were identical within a series of experiment nm) is shown in Supplementary Fig. 18.
In the case of the photosensitization conditions, the triplet excited state of the substrate ( 3 M*) is generated through multiple electronic transition events, including a photoexcitation of the sensitizer and an intersystem crossing, followed by a bimolecular energy transfer from the sensitizer to the substrate M. These processes are known to be temperature dependent 27 Supplementary Fig. 19c). To determine whether the temperature effect is small within the range we investigated, we compared the activation energies obtained using the data points at a lower temperature range (0, 5 and 10 °C), then at a higher temperature range (10, 15 and 20 °C) and then using the full temperature range (0, 5, 10, 15 and 20 °C). All the values were within the range of 2.4 ± 0.14 kcal mol -1 indicating the small effect of temperature on the activation enthalpy.

Exclusion of the Possibility of Photothermal Activation of Ring Inversion
An alternative hypothesis for the photo-acceleration of the ring inversion of Th4 COT Saddle is thermal activation of molecular motion by absorbing photon energy. If photothermal activation plays a dominant role, the effect of photoirradiation on Th6 CDH Screw should be much higher than that on Th4 COT Saddle because the absorptivity of Th6 CDH Screw at λ = 365 nm (ε = 9.3

ACID Plots and NICS Values of Th4 COT Saddle in the S 0 and T 1 States
The ACID plots indicate that the cyclooctatetraene (COT) ring is non-aromatic in the S 0 saddle-shaped minimum energy structure (Supplementary Fig. 33) but antiaromatic in the transition state ( Supplementary Fig. 34). The thiophene units also lose part of their aromaticity.
For the T 1 state, the ACID plots suggest that the saddle-shaped minimum energy structure has aromaticity in both the thiophene ring and the COT ring ( Supplementary Fig. 35). This current is delocalized as a 16π-electron current. In the transition state structure, the aromaticity of the COT unit is more pronounced and that of the thiophene units decreases ( Supplementary Fig.   36).
The NICS scans of the NICS zz values with distance from the COT ring centre are consistent with the ACID plots. In the S 0 state, the COT ring of Th4 COT Saddle is non-aromatic at the minimum energy structure ( Supplementary Fig. 37) and antiaromatic at the transition state structure (Fig. 5c). In the T 1 state, it can be concluded that both the minimum energy and transition state structure have aromatic tendency (Supplementary Fig. 37 and Fig. 5c).
The NICS scans in the S 1 state are given for the S 1 minimum energy structure

Energetics of the Planar Conformation of [12]annulene
To confirm that Th6 CDH Screw cannot take a planar conformation, energy calculation of [12]annulene in its C 2 -symmetric conformation (similar to the minimum energy structure of Th6 CDH Screw ) and D 12h -symmetric conformation (hypothetical planar conformation) was performed. The planar conformation is enormously higher in energy (ΔE = 106.3, ΔH = 114.8, ΔG = 110.9 kcal mol -1 ), which indicates the absence of the planar conformation in Th6 CDH Screw conformers ( Supplementary Fig. 46).

Triplet Energies of Fluorenone, Th4 COT Saddle and Th6 CDH Screw
For efficient photosensitization, the triplet state energy (adiabatic) of the photosensitizer should be slightly higher than the triplet state energy (adiabatic) of the substrates. As shown in Supplementary Table 6, fluorenone, the sensitizer used in this study, has a ca. 5-10 kcal mol -1 higher triplet state energy than those of Th4 COT Saddle and Th6 CDH Screw , which is reasonable for smooth photosensitization.

Detailed Analysis of the Ring Inversion Transition State of Th4 COT Saddle
For the structural transition of COT, a possibility of pseudorotation 31 ( Supplementary Fig.   49), other than going through planar transition state, was significantly discussed around the 1990s for the ring inversion and bond shifting. The process of ring inversion from one tub conformer to its mirror image proceeds through an intermediate tub conformer via two pseudorotation transition states. For the parent COT, it is now widely accepted that the structural transition instead occurs through a planar transition state in the ground state and not through pseudorotation 32,33 . However, in some COT derivatives with bulky substituents, pseudorotation pathways for structural transition from one tub conformation to the other one have been reported 34,35 . Also for the excited state of parent COT, the lowest-energy conical intersection of COT reachable from the S 1 state (12 kcal mol -1 above the planar D 8h minimum that could also relax to either tub conformer) has a structure similar to a pseudorotation TS 36 .