Asuka
Yanagawa
,
Motoki
Tsuchiya
,
Ryo
Inoue
and
Yasuhiro
Morisaki
*
Department of Applied Chemistry for Environment, School of Biological and Environmental Sciences, Kwansei Gakuin University, 1 Gakuen Uegahara, Sanda, Hyogo 669-1330, Japan. E-mail: ymo@kwansei.ac.jp
First published on 20th December 2022
Optical resolution of pseudo-para-disubstituted [2.2]paracyclophane was achieved using the diastereomer method. Optically active π-stacked molecules consisting of phenanthrene and benzo[c]phenanthrene were synthesized using the enantiopure pseudo-para-disubstituted [2.2]paracyclophanes as chiral building blocks. Helicity was induced on the stacked phenanthrene and benzo[c]phenanthrene moieties by the planar chiral [2.2]paracyclophane, and the molecule formed an optically active S-shaped structure. The S-shaped molecule emitted circularly polarized luminescence, which was compared with that of the corresponding optically active U-shaped molecule prepared from pseudo-meta-disubstituted [2.2]paracyclophane.
As mentioned above, we have developed optical resolution methods for disubstituted [2.2]paracyclophanes such as pseudo-ortho-disubstituted5a and pseudo-meta-disubstituted5b [2.2]paracyclophanes. Pseudo-para-disubstituted [2.2]paracyclophane becomes a planar chiral molecule depending on the substituents; in the case of different substituents, it exhibits planar chirality. To the best of our knowledge, there are a few reports on the optical resolution of pseudo-para-disubstituted [2.2]paracyclophane using the kinetic resolution method.7 Akiyama et al. attempted kinetic resolution of racemic pseudo-para-bromohydroxy[2.2]paracyclophane using chiral phosphoric acid as the catalyst.7a Benedetti and Micouin reported the ruthenium-catalyzed asymmetric transfer hydrogenation8 of pseudo-para-diformyl[2.2]paracyclophane to obtain the corresponding enantiopure pseudo-para-formyl-hydroxymethyl[2.2]paracyclophane.7b Rowlands reported the synthesis of chiral pseudo-para-disubstituted [2.2]paracyclophane with an oxazoline unit by C–H activation.7c The development of an optical resolution method for pseudo-para-disubstituted [2.2]paracyclophane is required for its use as a chiral building block in the synthesis of various optically active π-stacked molecules. This paper reports the optical resolution of pseudo-para-disubstituted [2.2]paracyclophane using the diastereomer method and the syntheses of optically active π-stacked molecules such as para-arylenevinylene-stacked molecules (PAV-stacked molecules) and helicene-stacked S-shaped molecules.
Scheme 1 (i) Optical resolution of racemic pseudo-para-disubstituted [2.2]paracyclophane rac-2. (ii) Synthesis of target molecule (Rp)-10. isomers. |
Fig. 2 (A) Molecular structures of (Rp)-10. (B) Molecular structures of (Rp)-14. Thermal ellipsoids are at the 30% probability level. Hydrogen atoms are omitted for clarity. |
The synthetic route to the helicene-stacked U-shaped molecule from pseudo-meta-disubstituted [2.2]paracyclophane is shown in Scheme 2. The starting compound (Sp)-11 was prepared previously and reacted with styrene 6 in a Pd(OAc)2/P(o-tol)3 catalytic system to afford (Sp)-12 in 77% isolated yield. The Heck–Mizoroki cross-coupling of (Sp)-12 with 8 afforded the corresponding PAV-stacked molecule (Sp)-13 in 35% isolated yield. Oxidative cyclization of (Sp)-13 in the presence of I2 under ultraviolet (UV) irradiation (λ = 365 nm) produced the target U-shaped molecule (Rp)-14 in 9% isolated yield. The X-ray crystal structure of (Rp)-14 is shown in Fig. 2B and Fig. S24 (ESI†).
As shown in Fig. 2A and B, the originally planar phenanthrene moieties in (Rp)-10 and 14 are slightly twisted by the bridging methylene of the planar chiral [2.2]paracyclophane unit. The torsion angles of the phenanthrene units in (Rp)-10 and 14 are 15.5° and 19.6°, respectively. Left-handed and right-handed helicities11,12 (M- and P-helicities) were induced to the phenanthrenes in (Rp)-10 and 14, respectively. The helicities of both the benzo[c]phenanthrene moieties in (Rp)-10 and 14 were right-handed helicities of (P)-[4]helicenes, with torsion angles of 25.4° and 27.5°, respectively. As shown in Fig. S34(B) (ESI†), when [4]helicene moiety in (Rp)-10 was forced to be flipped from (P)-helicity to (M)-helicity, its ΔE reached 22 kcal mol−1, indicating that configuration of [4]helicene moiety was (P)-helicity in solution, as well. On the other hand, flipping of the phenanthrene moiety is seemed to be easy in the solution sate (Fig. S34A, ESI†). The same phenomena were shown in (Rp)-14.
Ultraviolet-visible (UV-vis) absorption, photoluminescence (PL), circular dichroism (CD), and CPL spectra of both enantiomers of 9, 10, 13, and 14 were obtained from their dilute CHCl3 solutions (1.0 × 10−5 M). Those of the PAV-stacked molecules 9 and 13 are shown in Fig. S30 and S31 (ESI†), respectively. Fig. 3 shows the UV-vis absorption spectrum in a dilute CHCl3 solution of (Rp)-10 with the oscillator strengths simulated using the time-dependent density functional theory (TD-DFT) calculations at the TD-CAM-B3LYP(CHCl3)/6-31G(d)//CAM-B3LYP(CHCl3)/6-31G(d) level of theory. Fig. S28 (ESI†) shows the simulated UV-vis absorption spectrum of (Rp)-10 based on the oscillator strengths. The absorption peak top of (Rp)-10 was observed at 285 nm, and a tailing band was observed at the absorption edge (Fig. 3A). According to the simulated oscillator strengths, the tailing band consisted of weak absorption bands in the longer wavelength region (Fig. 3B). Fig. 4A and B show the UV-vis absorption spectrum and the calculated oscillator strength of (Rp)-14. The absorption bands of (Rp)-14 were almost the same as those of (Rp)-10. The molecular orbitals of (Rp)-10 and (Rp)-14 are shown in Fig. S32 and S33 (ESI†), respectively, in addition to those of phenanthrene and benzo[c]phenanthrene, which are monomeric π-electron systems. Phenanthrene and benzo[c]phenanthrene are stacked at the terminal benzene moieties without orbital hybridization and perturbation, leading to narrow energy gaps between the HOMO (highest occupied molecular orbital), HOMO−1, and HOMO−2, as well as between the LUMO (lowest unoccupied molecular orbital) and LUMO+1. The major S1 transitions for both (Rp)-10 and (Rp)-14 were determined to be HOMO/LUMO+1 (Tables S6 and S7 (ESI†), respectively) with small oscillator strengths of f = 0.0020 for (Rp)-10 and 0.0014 for (Rp)-14 (insets in Fig. 3B and 4B, respectively). The chiroptical properties of 10 and 14 in the ground state were investigated, and their CD spectra in CHCl3 solutions (1.0 × 10−5 M) are shown in Fig. 5A and 6A, respectively. Fig. 5B and 6B shows the CD spectra of the (Rp)-isomers, (Rp)-10 and (Rp)-14, respectively, and Fig. 5C and 6C shows the respective TD-DFT simulated rotatory strengths of the (Rp)-isomers.13 Mirror-image CD spectra with large Cotton effects were observed for 10 and 14 (Fig. 5A and 6A, respectively), the maximum absolute molar ellipticity |[θ]| values of 10 and 14 were found to be 3.8 × 105 and 3.0 × 105 deg cm2 dmol−1, respectively. As shown in Fig. 5A and B, small Cotton effects were observed in the long wavelength region of the CD spectra of (Rp)-10; the [θ] values were 0.068 × 105 and −0.311 × 105 deg cm2 dmol−1 at 380 nm and 352 nm, respectively. These small Cotton effects were supported by the TD-DFT calculations, and the corresponding rotatory strengths were calculated to be +1.8 × 10−40, −5.2 × 10−40, and −11.3 × 10−40 esu2 cm2 (Fig. 5C). Thus, although the rotatory strengths were very small, positive and negative Cotton effects appeared in the long wavelength region. As shown in Fig. 6B, the same CD sign pattern was observed for (Rp)-14; thus, small positive and negative Cotton effects were observed in the CD spectrum, which was also reproduced by the rotatory strength simulation (Fig. 6C).
The chiroptical properties of 10 and 14 in the excited state were investigated by CPL spectroscopy.14Fig. 7 and 8 show the CPL spectra of 10 and 14 in CHCl3 solutions (1.0 × 10−5 M), respectively, along with their glum charts.15,16 Molecule 10 emitted PL with a low quantum efficiency (ΦPL) of 0.06, and its PL spectrum exhibited a vibrational structure (Fig. 7). The CPL signals of both isomers were observed in the emitting region, and the signs of the (Rp)- and (Sp)-isomers were positive and negative, respectively. Their signs were identical to those of the first Cotton effects of the (Rp)- and (Sp)-isomers (Fig. 5A). The |glum| value of 10 was estimated to be 1.4 × 10−3.
As shown in Fig. 8, the PL spectrum of 14 was almost identical to that of 10, and clear CPL signals were observed in the emission band. The CPL signs of (Rp)- and (Sp)-14 were positive and negative, respectively, which were the same as those of (Rp)- and (Sp)-10 (Fig. 7) as well as the signs of their first Cotton effects (Fig. 6A). The |glum| value of 14 was estimated to be 2.6 × 10−3, which was larger than that of 10 (1.4 × 10−3).
Conventional wisdom suggests that the |glum| value of 14 was higher than that of 10 because of the helicities of the stacked phenanthrene and [4]helicene induced by the planar chiral [2.2]paracyclophane, as shown in Fig. 2. Those in S-shaped (Rp)-10 were (M)-helicity and (P)-helicity, respectively (Fig. 2A), whereas those in U-shaped (Rp)-14 were (P)-helicity and (P)-helicity (Fig. 2B). Theoretical investigations were carried out to understand their PL and CPL behaviors using the TD-DFT calculations; the results are shown in Fig. 9 and 10. In both (Rp)-10 and (Rp)-14, emission occurred as a result of the LUMO to the HOMO transition of the [4]helicene moieties. Regardless of the orientation between the stacked phenanthrene and [4]helicene, the emitting species were the [4]helicene units in (Rp)-10 and (Rp)-14, leading to the almost identical PL spectra (Fig. 7 and 8).
The glum value can be estimated using electronic and magnetic transition dipole moments (μ and m, respectively), and it is expressed by the following equation: glum,calcd = 4|μ||m|cosθ/(|μ|2 + |m|2), where θ is the angle between μ and m.16 The results of the glum simulations for (Rp)-10 and (Rp)-14 are presented in Fig. 9 and 10, respectively. The μ of (Rp)-10 lies along the long axis of the S-shaped structure (Fig. 9), while that of (Rp)-14 lies along the center line of the U-shaped structure (Fig. 10). The θ angles of (Rp)-10 and (Rp)-14 are almost the same (69.6° and 65.5°, respectively). The μ of (Rp)-10 was longer than that of (Rp)-14, and the m of (Rp)-10 was shorter than that of (Rp)-14, indicating that the glum,calcd value (+3.1 × 10−3) of (Rp)-10 was lower than that (+7.4 × 10−3) of (Rp)-14. Although, the calculated glum,calcd values were overestimated in comparison with the observed glum,obsd values (Fig. 7 and 8), their order of magnitude and positive signs of the simulation were consistent with the observed values.
(RP,1S,4R)-4:1H NMR (CDCl3, 500 MHz) δ 1.15 (s, 3H), 1.18 (s, 3H), 1.21 (s, 3H), 1.77−1.84 (m, 1H), 1.99–2.06 (m, 1H), 2.19–2.26 (m, 1H), 2.56–2.63 (m, 1H), 2.72–2.86 (m, 2H), 2.93–3.03 (m, 2H), 3.07–3.21 (m, 3H), 3.46–3.53 (m, 1H), 6.04 (d, J = 1.7 Hz, 1H), 6.42 (d, J = 8.0 Hz, 1H), 6.48 (d, J = 1.7 Hz, 1H), 6.52 (d, J = 8.0 Hz, 1H), 6.94 (dd, J = 1.7, 1.7 Hz, 1H), 7.12 (dd, J = 1.7, 1.7 Hz, 1H) ppm; 13C{1H} NMR (CDCl3, 125 MHz) δ 9.8, 16.8, 17.0, 28.9, 30.9, 31.1, 32.7, 33.5, 35.3, 54.6, 55.0, 90.9, 126.7, 127.7, 127.8, 127.8, 129.1, 130.4, 134.1, 134.1, 134.4, 137.5, 138.6, 141.3, 148.5, 165.6, 178.3 ppm. HRMS (ESI+) calcd for C26H27BrO4 + Na+: 505.0985, found 505.0987. [α]D25 = −55.28 (c 0.20, CHCl3).
(SP,1S,4R)-4:1H NMR (CDCl3, 500 MHz) δ 1.17 (s, 3H), 1.20 (s, 3H), 1.21 (s, 3H), 1.79–1.86 (m, 1H), 2.00–2.08 (m, 1H), 2.23–2.31 (m, 1H), 2.56–2.64 (m, 1H), 2.71–2.87 (m, 2H), 2.93–3.22 (m, 5H), 3.46–3.53 (m, 1H), 6.07 (d, J = 1.7 Hz, 1H), 6.45 (d, J = 8.0 Hz, 1H), 6.48 (d, J = 1.7 Hz, 1H), 6.52 (d, J = 8.0 Hz, 1H), 6.94 (dd, J = 1.7, 1.7 Hz, 1H), 7.12 (dd, J = 1.7, 1.7 Hz, 1H) ppm; 13C{1H} NMR (CDCl3, 125 MHz) δ = 9.8, 16.9, 16.9, 29.0, 31.1, 31.2, 32.7, 33.5, 35.3, 54.5, 55.0, 90.9, 126.8, 127.7, 127.8, 129.1, 130.4, 134.1, 134.4, 137.5, 138.6, 141.2, 141.3, 148.5, 165.5, 178.1 ppm. HRMS (ESI+) calcd for C26H27BrO4 + Na+: 505.0985, found 505.0969. [α]D25 = +55.48 (c 0.20, CHCl3).
R f = 0.30 (CHCl3/hexane = 1/2 v/v). 1H NMR (CDCl3, 500 MHz): δ 1.91–1.99 (m, 1H), 2.86–2.90 (m, 1H), 3.27–3.29 (m, 2H), 3.48–3.51 (m, 1H), 3.69–3.73 (m, 2H), 4.62–4,66 (m, 1H), 5.25 (d, J = 6.9 Hz, 1H), 5.60 (d, J = 6.9 Hz, 1H), 6.42 (dd, J = 6.9, 6.9 Hz, 2H), 7.54–7.63 (m, 5H), 7.75 (d, J = 8.6 Hz, 1H), 7.87 (d, J = 8.6 Hz, 1H), 7.90–7.97 (m, 4H), 8.01 (dd, J = 1.7, 1.2 Hz, 1H), 8.57 (d, J = 8.0 Hz, 1H), 8.80–8.81 (m, 1H) ppm; 13C{1H} NMR (CDCl3, 125 MHz): δ 32.0, 32.1, 37.7, 38.4, 124.1, 124.5, 125.5, 125.7, 125.8, 126.0, 126.5, 126.5, 127.1, 127.2, 127.7, 127.9, 128.4, 128.6, 128.8, 129.3, 129.7, 130.5, 130.5, 131.2, 131.4, 132.1, 132.5, 132.5, 133.4, 134.2, 134.4, 135.4, 136.0, 136.6 ppm. HRMS (APCI): m/z calcd for C36H26 + H+: 459.2107; found: 459.2092. [α]D25 = +304.43 (c 0.060, CHCl3).
(Sp)-10 was obtained in 6% yield by the same procedure of (Rp)-10. HRMS (APCI) calcd for C36H26 + H+: 459.2107, found 459.2114. [α]D25 = −304.23 (c 0.060, CHCl3).
R f = 0.39 (CH2Cl2/hexane = 1/2 v/v). 1H NMR (CDCl3, 500 MHz) δ 2.70–2.75 (m, 1H), 2.94–3.00 (m, 1H), 3.12–3.27 (m, 3H), 3.72–3.78 (m, 1H), 3.90–3.98 (m, 2H), 5.44 (d, J = 8.0 Hz, 1H), 5.86 (d, J = 7.5 Hz, 1H), 6.04 (d, J = 8.0 Hz, 1H), 6.13 (d, J = 7.5 Hz, 1H), 7.46–7.59 (m, 4H), 7.79 (d, J = 8.6 Hz, 1H), 7.83 (d, J = 7.45 Hz, 1H), 7.92–8.01 (m, 5H), 8.23 (d, J = 8.02 Hz, 1H), 8.44 (d, J = 8.59 Hz, 1H) ppm; 13C{1H} NMR (CDCl3, 125 MHz) δ 32.6, 32.7, 36.6, 38.0, 123.4, 124.4, 125.2, 125.5, 125.6, 125.66, 125.73, 125.9, 126.5, 127.2, 127.5, 127.6, 128.1, 128.2, 128.4, 128.7, 128.8, 129.2, 129.5, 130.2, 130.3, 130.4, 130.5, 132.2, 132.3, 132.8, 133.0, 134.2, 134.4, 134.6, 136.0, 137.8 ppm. HRMS (APCI) calcd for C36H26 + H+: 459.2107, found 459.2100. [α]D25 = +508.4 (c 0.042, CHCl3).
(Sp)-14 was obtained in 10% yield by the same procedure of (Rp)-14. HRMS (APCI) calcd for C36H26 + H+: 459.2107, found 459.2106. [α]D25 = −508.3 (c 0.02, CHCl3).
Footnote |
† Electronic supplementary information (ESI) available: Experimental details including synthetic procedures, characterizations, NMR and HRMS spectra. The results of X-ray crystallography, UV, CD, PL, CPL spectroscopy, and theoretical studies. CCDC 2214406–2214410. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d2tc04652k |
This journal is © The Royal Society of Chemistry 2023 |