*4.2. The Curious Case of the 2-Indenyltriptycene Dimer*

In a closely related system, the oxidative coupling of 2-indenyltriptycene produces *rac*-2,2 -di(9 triptycyl)-1,1 -biindenyl, 51, whose structure appears as Figure 25. However, the molecule is not strictly C2 symmetric in the solid state; although the 1,1 -biindenyl core maintains its local C2 axis, the Cs-type orientation of the intermeshed triptycene blades breaks this symmetry, as depicted in the space-fill representation. Remarkably, despite the obvious steric congestion, there is no evidence, even at 193 K, of slowed paddlewheel rotation on the NMR time-scale. Evidently, in solution, it behaves as a molecular gearing system that exhibits dynamic C2 symmetry implying that the triptycyl units undergo correlated disrotatory motion with concomitant interplanar bending to compensate for the unusually large range of angles (from 103◦ to 135◦) between the paddlewheel blades [31].

**Figure 25.** Molecular structure of *rac*-2,2 -di(9-triptycyl)-1,1 -biindenyl, **51**, and a space-fill view showing the gear meshing of the triptycyl groups [31].

#### **5. Hindered Rotations in Phenyl-Anthracenes**

Many ligands widely used in catalytic asymmetric syntheses, such as BINAP, depend for their activity on the phenomenon of non-interconverting atropisomers that give rise to *R* and *S* enantiomers because of steric hindrance between bulky aromatic ring systems. We were, therefore, intrigued by the computational prediction that 9-phenylanthracene should adopt an orthogonal orientation of the two ring systems and exhibit a rotational barrier of ca. 84 kJ mol−<sup>1</sup> about the axis connecting them [32]. To probe such an assertion experimentally, one must lower the *C*2v symmetry of 9-phenylanthracene (or the *D*2h symmetry of 9,10-diphenylanthracene), but not introduce any additional steric perturbations.

Once again, a palladium-mediated procedure appeared to be viable. Our initial approach involved the incorporation of *meta*-fluoro substituents in 9,10-diphenylanthracene, via the Suzuki cross-coupling (in 97% yield) of 9,10-dibromoanthracene with 3-fluorophenylboronic acid, catalysed by (dppf)PdCl*<sup>2</sup>* in 1,4-dioxane in the presence of tetrabutylammonium hydroxide [33]. The interconversion of the *syn* and *anti* rotamers of 9,10-di(3-fluorophenyl)anthracene, 52, was monitored by variable-temperature 13C-and 19F-NMR spectroscopy and yielded a barrier of 88 <sup>±</sup> 2 kJ <sup>×</sup> mol−1, gratifyingly close to the calculated prediction. The structure of *anti*-52, appears in Figure 26, showing that in the solid state the fluorophenyl groups are orientated at 85◦ to the plane of the anthracene.

**Figure 26.** Structures of *anti*-9,10-di(3-fluorophenyl)anthracene, **52**, and of 9-(1-naphthyl)-10 phenylanthracene, **53**.

Nevertheless, even this minor perturbation does not really provide an answer for the unsubstituted phenyl group, which requires that the symmetry of the environment of the phenyl be broken rather than the symmetry of the phenyl itself. This was accomplished by the synthesis and structural characterisation of 9-(1-naphthyl)-10-phenylanthracene, 53, by using the same Suzuki procedure to couple 9-iodo-10-phenylanthracene and 1-naththylboronic acid in 55% yield. Since it is known that the rotational barrier in 9-(1-naphthyl)anthracene is ~160 kJ mol−<sup>1</sup> [32], this provides a rigid mirror-symmetric framework against which the dynamic behaviour of an unsubstituted phenyl ring can be monitored. The structure of 53 is also shown in Figure 26 and reveals that the dihedral angles of the phenyl and naphthyl rings relative to the plane of the anthracene are 80◦ and 88◦, respectively. Moreover, even at 303 K, on a 600 MHz spectrometer the *ortho* proton resonances of the phenyl ring already show non-equivalence, indicating slowed rotation on the NMR time-scale, and a barrier of at least 85 kJ mol−<sup>1</sup> [33]. Overall, these observations provide an experimental validation of the computational prediction.
