Organic Letters
Letter
Cγ on the phenanthrene backbone as the only site for the
protonation. The next step is the electrophilic attack of the
ortho (Cβ) or para (Cα) position by the adjacent biphenyl
moiety. After the first C−C bond formation, the same pathway
is followed by the other side of the molecule (Scheme 1).
Indeed, the DFT calculations suggest that C−C bond
formation at the carbons Cα has a lower transition-state
energy (ΔΔG ≈ 3 kcal/mol, calculated at the CAM-B3LYP-
D3BJ/6-31+G**+PCM(CH2Cl2) level of theory; Figures S5
and S6, SI) and should be the major product. This functional
and basis set are used throughout this Letter unless otherwise
noted. (Computational details are provided in the SI.)
However, one may have predicted that the formation of
sterically unstrained M1 should be the major product because
it is ∼11.9 kcal/mol more stable than the corresponding
helicene H1 (Figure S9, SI). Note that the cation radical
mechanism showed similar regioselectivity but higher barriers
(∼8 kcal/mol; Figures S7 and S8, SI), and thus we believe that
the arenium ion mechanism is a more feasible reaction
pathway.
Scheme 2. Synthetic Approaches for (A) Mono [7]helicene
1, (B) Double [7]helicenes 2 and 3, and (C) Controlled
Molecule M2
To assess our hypothesis that cyclization at Cα is preferred
despite the buildup of strain energy, we followed a systematic
synthetic approach. The strategy for the preparation of
helicene 1 starts with the seven-step synthesis of 3,6-
dibromo-9,10-dibutyl-phenanthrene 11 from the readily
available 2-bromobenzoyl chloride in ∼55% overall yield.
suitable coupling partner for the Suzuki−Miyaura cross-
coupling reaction, the bromines were converted to boronate
ester (BPin) 12 in 69% yield and reacted with 2-bromo-4,4′-di-
tert-butylbiphenyl S121 using Pd(PPh3)4 as the catalyst to
afford compound 15 in 80% yield (Scheme 2A). Note that t-
butyl groups are added to increase the solubility of the product.
Finally, the oxidative C−C bond formation reaction was
completed using 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ) and MeSO3H in CH2Cl2 at 0 °C for <1 h.22 In
agreement with our hypothesis, only the helicene 1 was
observed as the product in excellent isolated yield (∼84%),
and no planar isomer (M1) was observed. The structure of π-
extended [7]helicene 1 was confirmed by NMR and single-
crystal X-ray crystallography. (See the crystallography section
Encouraged by these results, we considered the possibility of
the synthesis of a double [7]helicene. The simple retro-
synthetic analysis of [7]helicene (Scheme 1) suggests that
double analogs can be synthesized from a dibenzo[g,p]-
chrysene (DBC) derivative, which is essentially a combination
of mirror images of phenanthrene. The desired double
[7]helicene is much higher in energy (ΔG = 25.6 kcal/mol;
Figure S10, SI) than the thermodynamically more stable
isomer M2 (Scheme 2C). To address this challenge, we
undertook the synthesis.
identified as the P,P/M,M (2) and M,P/P,M (3) config-
urations, respectively; note that 3 is ∼4.7 kcal/mol less stable
than 2 (Figure S10, SI). To prove that no nonhelical isomer
was present in our mixture, the synthesis of compound M2 has
been accomplished starting from the 2,7,10,15-
tetrabromodibenzo[g,p]chrysene S3.23 As provided in the SI
(Figure S4), the NMR spectrum of this compound is distinct,
and no trace of this chemical shift pattern was observed in the
double-helicene reaction.
The crystal structures of mono (1) and double [7]helicenes
(2 and 3) are depicted in the Figure 1. In the mono
[7]helicene 1, the center-to-center distance between the
overlapped terminal phenyl rings is ∼3.8 Å which is more
than twice the van der Waals radius of a carbon atom (∼1.7
Å). Comparing these values with the crystal structure of
normal [7]helicene (i.e., neither extended nor substituted, 3.8
Å)24 shows that the extension of the π system or substitution
did not affect the general geometry of 1. The distance between
the CH3 carbon of the t-butyl group and the center of the
extended phenyl ring is as short as ∼3.5 Å for 1. Also, the
attached biphenyl moieties undergo contortion after oxidative
C−C bond formation reaction. Analyzing the crystal structure
of 2 shows that the center-to-center distance of the terminal
rings increases to ∼4.2 Å (average of two helicenes); this value
is ∼4.1 Å for 3. The bigger values, compared with 1, can be
attributed to the twisted nature of the DBC framework, which
is not planar like the phenanthrene in 1. This effect also reveals
itself in the t-Bu CH3 carbon-to-phenyl ring distance, which
increases to ∼3.7 Å for both 2 and 3.
DBC has been brominated in quantitative yield to afford the
3,6,11,14-tetrabromodibenzo[g,p]chrysene S2.23 The Suzuki−
Miyaura cross-coupling reaction of S2 with four equivalents of
biphenyl boronic ester 4 has generated the important precursor
13 in 67% yield, which allowed us to carry out the final step,
oxidative C−C bond formation using DDQ/MeSO3H in
CH2Cl2 at 0 °C, as shown in Scheme 2B. As expected by
looking at the structure of the double [7]helicene derivatives,
two different relative configurations (P,P/M,M and M,P/P,M)
are synthetically feasible. Fortunately, we were able to isolate
and identify them. The major and minor products were
5171
Org. Lett. 2021, 23, 5170−5174