probes for dynamics in crystalline solids and rigid media.12
2
In brief, the H NMR method relies on the orientation
dependence of the quadrupolar coupling and its modulation
by molecular motions. Although powder samples with slowly
moving or rigid molecules give rise to broad symmetric
spectra known as a Pake pattern, molecular motion and
reorientation give rise to spectral variations that depend on
the rate and type of the deuteron motion. The dynamic range
covered by line shape analysis techniques varies from ca.
104 to 108 s-1, which correspond to the slow and fast
exchange limits of the method, respectively. For our studies,
we take advantage of the well-characterized spectral changes
occurring upon rotation of phenylene groups about their 1,4-
axis, typically by 180° angular displacement, which are
commonly referred to as phenylene flips.13,14
An isotopically labeled sample of 3-d4 for dynamic NMR
was prepared as shown in Scheme 1 using 1,4-dibromoben-
zene-d4. The spectra were acquired with a quadrupole echo
pulse sequence15 with a 90° pulse of 2.25 µs, with echo and
refocusing delays of 50 and 40 µs, respectively. The delay
between pulses was 30 s. Exchange rates were determined
by visually matching the experimental data to spectra
simulated using the program reported by Nishikiori et al.14
The simulated spectra were obtained using a quadrupolar
coupling constant (QCC) measured from spectra acquired
in the slow exchange regime and a model that considers site
Figure 6. (Left) Experimental (thin line) and simulated (thick line)
wide-line 2H NMR spectrum of 1 in a solvent-free structure. (Right)
Arrhenius plot of the exchange data.
Although we do not know the structure of solvent-free 3,
we note that this interesting result is consistent with the
smaller packing coefficient for the solvated crystals of hybrid
structure 3 (0.69) as compared to those of the trityl compound
1 (0.74) and the static triptycyl compound 2 (0.85). Interest-
2
exchange of the H nuclei by 180° rotations. The only
adjustable parameter used in the simulation was the site-
exchange rate krot, which was varied from 104 to 108 Hz.
The experimental and simulated spectra for the solvent-
free sample are shown in Figure 6 for data recorded between
300 and 448 K. The best visual agreement was obtained using
rates of rotation between 0.5 × 106 and 50 × 106 s-1.
Arrhenius analysis of the data gave a barrier for phenylene
flipping of 7.9 ( 1.6 kcal/mol and a preexponential factor
of 2.75 × 1011 Hz.16 The relatively linear plot suggests that
the solvent-free structure is a single phase and not a mixture
of polymorphs. Notably, the barrier in 3 is 6 kcal/mol lower
than that previously measured by the same method for
solvent-free samples of 1 within a similar temperature range.7
The lower barrier results in faster rotation rates, which are
desirable to control the orientation of dipolar analogues with
external electric and magnetic fields.
2
ingly, although an attempt to measure the H NMR of the
solvate gave a heterogeneous spectrum with two or more
crystalline components, we made a qualitative observation
that the rates of phenylene motion in the solvate are slower
than those of the solvent-free structure, suggesting that
desolvation occurs without collapse of the lattice (Supporting
Information).
In conclusion, molecular gyroscope 3 with an asymmetric
triptycyl-trityl stator structure retains the relatively high
melting point of the two symmetric structures, 1 or 2, but it
displays largely improved solubility and dynamic properties.
The packing structure of the CHCl3 clathrate retains the
parallel alignment observed in both 1 and 2, with phenylene-
triptycene intermolecular contacts similar to those present
in 2. We believe that compound 3 represents an interesting
lead that highlights the potential properties of solid-state
molecular gyroscopes with asymmetric structures.
(12) Hoatson, G. L.; Vold, R. L. NMR Basic Princ. Prog. 1994, 32, 1-67.
(13) (a) Cholli, A. L.; Dumais, J. J.; Engel, A. K.; Jelinski, L. W.
Macromolcules 1984, 17, 2399-2404. (b) Rice, D. M.; Wittebort, R. J.;
Griffin, R. G.; Meirovich, E.; Stimson, E. R.; Meinwald, Y. C.; Freed, J.
H.; Scheraga, H. A. J. Am. Chem. Soc. 1981, 103, 7707-7710. (c) Rice,
D. M.; Meinwald, Y. C.; Scheraga, H. A.; Griffin, R. G. J. Am. Chem. Soc.
1987, 109, 1636-1640. (d) Rice, D. M.; Blume, A.; Herzfeld, J.; Wittebort,
R. J.; Huang, T. H.; DasGupta, S. K.; Griffin, R. G. Biomol. Stereodyn.,
Proc. Symp. 1981, 2, 255-270.
Acknowledgment. This work was supported by the
National Science Foundation through grants DMR0307028
and DGE-0114443 (NSF IGERT: Materials Creation
Training Program (MCTP)).
(14) Nishikiori, S.; Soma, T.; Iwamoto, T. J. Inclusion Phenom. 1997,
27, 233-243.
Supporting Information Available: Synthesis and char-
2
(15) The 2H NMR spectra were acquired on a Bruker Avance spectrom-
eter at a frequency of 46.073 MHz with a single channel solenoid probe
containing a 5 mm insert. Measurements were made from 300 to 448 K
with the temperature calibrated using a 195Pb shift standard.
(16) The uncertainty in the barrier was estimated from simulations with
the highest and lowest exchange rates that reasonably matched the
experimental spectra at each temperature value.
acterization of compounds 3 and 5, TGA traces, H NMR
of partially desolvated samples, and crystallographic infor-
mation (CIF) for 3. This material is available free of charge
OL060894D
3420
Org. Lett., Vol. 8, No. 16, 2006