the diene units relative to the zirconacyclopentadiene rings
of 3. Related dynamic behavior was observed for a trimeric
9,10-anthrylene ethynylene macrocycle.14
To further probe conformational changes in macro-
cycles 2 and 3, VT-NMR studies were performed. High
temperature NMR studies of 2 revealed only decomposi-
tion of the macrocycle at temperatures above 40 °C. This
suggests that decoupling of the zirconocene-containing
macrocycle requires less energy than rotation of the an-
thracene moieties. This lability is consistent with the ob-
served, facile fragmentation of the macrocycle observed by
mass spectrometry. Interestingly, low-temperature NMR
spectra of cyclophane 3 in dichloromethane-d2 feature only
insignificant chemical shift changes from 20 to -80 °C.
This suggests that, despite the common, trimeric nature of
2 and 3, these macrocycles feature significant structural
differences.
Comparisons with analogous macrocycles, in metalated
and demetalated forms, indicate that the most dramatic
structural change associated with hydrolytic removal of
the zirconocene units is reflected in differences in torsion
angles about the butadiene/zirconacyclopentadiene “cor-
ners” of the macrocycles. For example, in the zirconocene-
based macrocycle resulting from cyclization of 4,40-bis-
(trimethylsilylethynyl)biphenyl, the phenylene substitu-
ents are highly eclipsed as reflected in a (C6H4)C-C;
C-C(C6H4) torsion angle of only 2° for the zirconacyclo-
pentadiene units. The corresponding cyclophane is con-
siderably twisted at the butadiene groups (analogous
torsion angle of 37.0°).5a Similar structural differences
were observed by DFT studies of the simple (1E,3E)-1,4-
bistrimethylsilyl-2,3-diphenylbuta-1,3-diene and its corre-
sponding zirconacycle, in which no ring strain influences
the arrangement of the phenyl substituents (for computa-
tional details, see Supporting Information).15 The opti-
mized structures (Figure 3a) feature torsion angles of 0.3°
and 50.5°, respectively. Due to steric pressure in the cyclic
structures, this differenceisless pronouncedin2 and 3. The
average torsion angle for the optimized structure of 2 was
found to be 9.66°, whereas the analogous torsion angle in
the crystal structure of 3 is 43.7°.
Figure 2. Molecular structure of cyclophane 3 (for crystallo-
graphic details, see the Supporting Information) including van
der Waals radii.
due to unfavorable steric interactions which prevent the
selective formation of one macrocyclic product.
Interestingly, the 1H NMR spectra of macrocycles 2 and
3 exhibit key differences in the region associated with the
anthracene protons (Supporting Information). Thus, the
zirconocene macrocycle 2 exhibits four clearly resolved
signals for the anthracene protons at 20 °C, as expected for
a structure with the three anthracene groups strongly
leaning toward the center of the macrocycle. In addition,
one proton resonance is significantly shifted upfield and
appears at 6.21 ppm in chloroform-d and at 6.51 ppm in
toluene-d8.12 This apparently reflects a stacking of the
anthrylene units in solution, in the manner observed for
3 in the crystal. Thus, the 2,3-anthrylene protons oriented
toward the macrocycle center experience ring currents that
lead to the observed upfield shift.13 In contrast to 2, the
1
The increased torsion angles associated with demetala-
tion result in a conformationally less rigid macrocycle. In
the case of compounds 2 and 3 this may explain the
demetalated compound 3 exhibits only two H NMR
signals for the anthracene protons at room temperature,
which indicatesa dynamicprocessthatexchanges the outer
rings of the anthracene groups, and increased flexibility of
1
observed differences in the H NMR spectra. In 3, the
phenylene substituents have more freedom of motion with
respect to one another, via lower rotational barriers about
the (C6H4)C-C(diene) and (C6H4)C-C(C6H4) bonds.
The rather bulky silyl groups also appear to play a role
in determining the conformational rigidity of the macro-
cycles. In particular, these groups should hinder rotation
about the (C6H4)C-C(zirconacycle) bonds of 2. This is
apparent from a space-filling model of the optimized struc-
ture of 2 (Figure 3b) and from distances between adjacent
phenylene and silylgroups in2 and 3. The distance between
(12) For examples of NMR studies on 9,10-substituted anthracenes,
€
see: (a) Nikitin, K.; Muller-Bunz, H.; Ortin, Y.; Muldoon, J.; J.
McGlinchey, M. Org. Lett. 2011, 13, 256. (b) Zehm, D.; Fudickar, W.;
Hans, M.; Schilde, U.; Kelling, A.; Linker, T. Chem.;Eur. J. 2008, 14,
11429. (c) Fairfull-Smith, K. E.; Bottle, S. E. Eur. J. Org. Chem. 2008,
5391. (d) Nakatsuji, S.; Matsuda, K.; Uesugi, Y.; Nakashima, K.;
Akiyama, S.; Fabian, W. J. Chem. Soc., Perkin Trans. 1 1992, 755. (e)
€
Schmidt, R.; Gottling, S.; Leusser, D.; Stalke, D.; Krause, A.-M.;
€
Wurthner, F. J. Mater. Chem. 2006, 16, 3708. (f) Valentini, L.; Bagnis,
D.; Marrocchi, A.; Seri, M.; Taticchi, A.; Kenny, J. M. Chem. Mater.
2008, 20, 32.
(13) For other examples of molecules exhibiting resonance upfield
shifting in 1H NMR spectra due to intramolecular π-π stacking and/or
ring current shielding effects in constrained structures, see: (a) Ting, Y.;
Lai, Y. H. J. Am. Chem. Soc. 2004, 126, 909. (b) Mei, X.; Wolf, C. J. Org.
Chem. 2005, 70, 2299. (c) Nandy, R.; Subramoni, M.; Varghese, B.;
Sankararaman, S. J. Org. Chem. 2007, 72, 938.
(14) Chen, S.; Yan, Q.; Li, T.; Zhao, D. Org. Lett. 2010, 12, 4784.
(15) Erker, G.; Zwettler, R.; Kruger, C.; Hyla-Kryspin, I.; Gleiter, R.
Organometallics 1990, 9, 524.
1156
Org. Lett., Vol. 13, No. 5, 2011