.
Angewandte
Communications
(M,R,R)-4[11] in a high yield. Note that the opposite helicity
(M) prevailed in this case.
Encouraged by this observation, we focused on exploring
the scope and limitations of this diastereoselective cyclo-
isomerization. Therefore, the chiral triynes (R,R)-7–12, 19,
and 20[11] with tolyl-terminated alkyne units were cyclized to
the corresponding heterohelicenes with an excellent 100:0
diastereoselectivity in favor of the (M,R,R) stereoisomers of
2, 13–18, and 21 (Scheme 2). The stereochemical outcome of
all of these reactions was sensitive to neither the reaction
conditions used nor the presence of different substituents
(CH3O, Br, and CO2CH3 in the different positions) nor
heterocyclic subunit(s), namely pyridine. Regardless of the
length of the helical scaffold, this method allowed for the
construction of [5]-, [6]-, and [7]heterohelicenes in optically
pure form. Mostly good to high preparative yields were
obtained under at least one of the reaction conditions
screened: A) [CpCo(CO)2]/PPh3 with simultaneous visible
light irradiation; B) [CpCo(CO)(fum)][12] (fum = dimethylfu-
marate) in combination with microwave irradiation; and
C) [Ni(cod)2]/PPh3. The bromo derivative (M,R,R)-15 was
obtained in only moderate yield (a two-step chromatographic
purification was necessary) and the cyclization of both the
pyridine-derived triynes required a stoichiometric amount of
the [CpCo(CO)(fum)] complex to afford the helical pyrido-
helicenes (M,R,R)-17 and (M,R,R)-18 (no other cyclization
conditions were effective).
The helicity of the key cyclized product (M,R,R)-4 could
not be unequivocally assigned by measuring NOE in the
1H NMR spectra, and we did not obtain suitable crystals for
a single-crystal analysis. However, the theoretical calculations
predicted its M helicity convincingly in the case of an R
configuration at stereogenic centers (see Figure 2). That
result was further supported by comparing the observed and
calculated electronic CD spectra and the chemical shifts of
(M,R,R)-4 in the 1H and 13C NMR spectra, which were
juxtaposed with the calculated parameters of the synthetically
inaccessible (P,R,R)-4 (for details and the experimental CD
spectra of (M,R,R)-3/(P,R,R)-3, (M,R,R)-4, (M,R,R)-2, and
(M,R,R)-21; see the Supporting Information).
Figure 2. The relative equilibrium free energies and barriers to epime-
rization of (M,R,R)/(P,R,R)-3 and (M,R,R)/(P,R,R)-4. These values were
calculated by DFT (B3LYP/cc-pVTZ), and the transition states TS were
localized by the QST3 method (Gaussian09).
21.1 kcalmolꢀ1 (13.6 or 13.8 kcalmolꢀ1 for 3). We hypothe-
sized that such a stereocontrol might operate in the higher
and functionalized derivatives as well.
In accord with the theoretical calculations, the stereo-
chemical outcome of [2+2+2] cycloisomerization of the
centrally chiral triyne (R,R)-5[11] with terminal alkyne units
was unsatisfactory, as a 34:66 mixture of (M,R,R)- and
(P,R,R)-3 was formed regardless of the reaction conditions
used (Scheme 1). As predicted, the presence of the tolyl
groups at the tethered alkyne units had a dramatic effect on
the stereochemical outcome of the reaction: The cyclization
of the chiral triyne (R,R)-6 afforded exclusively the helicene
The helicity of the cyclized products (M,R,R)/(P,R,R)-3
was assigned by measuring the NOE in the 1H NMR spectra.
Furthermore, we found a good agreement between the
observed and calculated chemical shifts in the 1H and
13C NMR spectra of the distinguishable diastereomers
(M,R,R)- versus (P,R,R)-3 (for details, see the Supporting
Information).
The dynamic equilibrium between (M,R,R)- and (P,R,R)-
3 allowed the determination of the corresponding barriers to
epimerization by the temperature-dependent 1H NMR meas-
urements. In accord with the theoretical calculations
(Figure 2), we determined the experimental barrier to be
13.0 kcalmolꢀ1 for the (P,R,R)-3!(M,R,R)-3 process and
12.8 kcalmolꢀ1 for the (M,R,R)-3!(P,R,R)-3 backward con-
version (the diastereomers differed in Gibbs energy by about
0.2 kcalmolꢀ1). A single-crystal analysis of the minor diaste-
reomer (M,R,R)-3, which crystallized preferentially from the
34:66 equilibrium mixture of (M,R,R)- and (P,R,R)-3, was
performed (for details, see the Supporting Information).[13]
Scheme 1. Structure optimization for diastereoselective [2+2+2] cyclo-
isomerization. [a] A: [CpCo(CO)2] (20 mol%), PPh3 (40 mol%), irradi-
ated by a halogen lamp, decane, 1408C, 1 h; B: [CpCo(CO)(fum)]
(20 mol%), 1-butyl-2,3-dimethylimidazolium tetrafluoroborate
(25 mLmLꢀ1 of the reaction solution), microwave reactor, THF, 1808C,
20 min; C: [Ni(cod)2] (20 mol%), PPh3 (40 mol%), THF, room temper-
ature, 30 min. fum=dimethylfumarate. Yields of isolated product after
column chromatography are given. The diastereomeric ratios were
1
determined by H NMR spectroscopy (the stereochemical outcomes
were independent of the cyclization procedure used). [a]2D2 values were
determined in chloroform (c=0.28–0.40).
2
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!