Highlighting gyroscopic motion in crystals in 13C CPMAS spectra by specific isotopic substitution and restricted cross polarization

Article abstract of DOI:10.1039/b409744k

The temperature-dependent exchange rate and signal coalescence in the ¹³C CPMAS NMR spectrum of a crystalline molecular gyroscope are exposed by specifically deuterating the overlapping static carbons.

Full text of DOI:10.1039/b409744k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
13
Highlighting gyroscopic motion in crystals in C CPMAS spectra by
specific isotopic substitution and restricted cross polarization
Steven D. Karlen and Miguel A. Garcia-Garibay*
Received (in Columbia, MO, USA) 26th June 2004, Accepted 21st September 2004
First published as an Advance Article on the web 24th November 2004
DOI: 10.1039/b409744k
In order to implement this strategy, we developed a very simple
labelling procedure and we tested the concept with relatively well-
characterized crystals of compound 1.
The temperature-dependent exchange rate and signal coales-
cence in the 13C CPMAS NMR spectrum of a crystalline
molecular gyroscope are exposed by specifically deuterating the
overlapping static carbons.
The synthesis of 1,4-bis(3,3,3-tri-d5-phenylpropynyl)benzene (1-
d30) was accomplished in three steps from NMR grade d6-benzene
as indicated in Scheme 1. A simple Friedel–Crafts reaction with
CCl4 and AlCl3 gave chloro-tris(d5-phenyl)methane 2 in 76%
yield.6 The trityl chloride 2 was treated with ethynylmagnesium
bromide to give 3,3,3-d15-triphenylpropyne 3 in 95% yield.
Finally, a double coupling with 1,4-diiodobenzene using 10%
Pd(PPh3)2Cl2 in piperidine converted alkyne 3 to 1-d30 in 30%
isolated yield.{
The spectroscopic and physical properties of 1-d30 were con-
sistent with those of unlabeled 1.1 Crystallizations occur in cen-
trosymmetric structures (Ci) both in solvent-free crystals and in a
benzene clathrate, from CH2Cl2 and C6H6, respectively. We had
previously shownthatthe 13C CPMASNMRsignals ofthe pheny-
lene carbons C1 and C3 in the benzene clathrate of 1 at 131.6
and 130.8 ppm were amenable to coalescence analysis. A rate of
exchange of 130 s21 and a barrier of rotation of 12.8 kcal mol21
were estimated at the coalescence temperature of 255 K. In
contrast, measurements carried out with solvent-free crystals
showed severe overlap between the carbons of the phenylene
rotor and those of the triphenylmethyl groups, rendering
coalescence analysis impossible.
Recent efforts from our group have been directed to the design,
synthesis and dynamic characterization of organic structures that
display internal rotary motion while in the solid state.1,2 Some of
the desired structures emulate the functions of macroscopic
compasses and gyroscopes. Illustrated in Fig. 1 with 1,4-
bis(3,3,3-triphenylpropynyl)benzene (1), they consist of a rotary
group, such as a p-phenylene (shown in red), linked to a shielding
triarylmethyl frame by a barrierless dialkyne axle (shown in blue).1
One of the most general methods to analyze the gyroscopic motion
of any central rotor in derivatives of 1 could be based on the use of
coalescence analysis by variable temperature 13C CPMAS NMR.3
The coalescence method takes advantage of the crystallographic
and magnetic non-equivalence of sites related by rotation about
the dialkyne axis, and requires the exchanging signals to be
spectrally resolved.4 Ideally, the rates of motion are varied as a
function of temperature and changes in lineshape are analyzed in
terms of exchange rates.
Given that structures analogous to 1 crystallize with low
molecular symmetries (either C1 or Ci) with no less than four and
as many as seven non-equivalent aromatic groups, the resolution
of the carbons assigned to the phenylene rotor is difficult.5 In order
to highlight the dynamics of the rotor one may take advantage of
specific deuteration to ‘‘remove’’ all the interfering signals from the
triphenylmethyl frame.
With samples of 1-d30 now available, we set out to determine its
dynamics. The 13C CPMAS NMR spectrum of crystals grown
from CH2Cl2 with a contact time of 8 ms at 214 K is shown in
Fig. 2 (top). The spectrum of 1-d30 was analogous to that
previously measured for the natural abundance samples. Signals
Fig. 1 (Left) A gyroscope is a device consisting of a spinning mass, or
rotor (gold), with a spinning axis positioned through the center of the
mass, and mounted within a rigid frame (silver). (Right) Analogous to the
macroscopic object, molecular gyroscope 1 contains a phenylene rotor
(red) with an alkyne axle and shielding groups that encapsulate it (blue).
*mgg@chem.ucla.edu
Scheme 1
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 189–191 | 189

View Article Online
between 1 and 1200 Hz.8 Small deviations in the simulated spectra
arise from contributions due to residual signals of the non-
protonated phenylene carbon at 121.6 ppm and one of the trityl
carbons at 126.3 ppm. These are both labelled with an asterisk in
Fig. 3.
With the rate of exchange derived from the simulation and
temperature data we produced an Arrhenius plot to calculate
an activation barrier of 11.3 kcal mol21 and a pre-exponential
factor of 2.9 6 1011 s21. This barrier is 1.5 kcal mol21 lower
than that previously determined for the benzene clathrate
(12.8 kcal mol21) and 3.3 kcal mol21 lower than that determined
for the same crystal form (Ea 5 14.6 kcal mol21) using
quadrupolar echo 2H NMR between 300 and 480 K.1 The
differences in activation energies for the same sample with two
different methods may result from the use of different tempera-
tures, different isotopologues (1-d4 vs. 1-d30), or the accumulated
experimental errors of the two techniques. While different barriers
at different temperatures would be indicative of a phase transition,
an explanation based on different isotopes would reflect a well-
documented steric effect arising from the different vibrational
amplitudes and effective sizes of the C–1H and C–2H bonds.9
While we estimate the error from the data shown in Fig. 3 to be
¡1 kcal mol21, analysis of the 2H NMR data published
Fig. 2 13C CPMAS NMR spectrum of 1-d30 with a contact time of 8 ms
(top) and 50 ms (bottom). The longer contact times allow for transfer of
polarization to all carbon atoms in the structure. Very short contact times
only allow for polarization to the phenylene carbons, which are directly
1
bound to the only H in the structure.
corresponding to the saturated trityl carbons occur at 55.0 ppm
and the alkyne signals at 83.7 and 97.1 ppm. The ipso-carbons
from the three non-equivalent phenyl groups of the trityl frame
resonate at 140.9, 142.5 and 148.6 ppm. Phenylene and protonated
trityl signals overlap between 121.6 and 132.3 ppm.
It should be noted that signal intensities in the CPMAS
experiment depend on the extent of 1H–13C cross polarization,
which in the case of 1-d30 originates exclusively from the 1H in the
central phenylene rotor. Since the strength of cross polarization
previously suggests a much larger error of ¡2.5 kcal mol21
.
2
Clearly, more precise values from H NMR experiments will be
required to verify the magnitude of the suggested barrier
difference.
1
depends on the distance-dependent H–13C dipole–dipole interac-
tion,7 signal intensities from carbons that are closer to the
In conclusion, a simple method was developed to prepare
molecular gyroscopes with hydrogenated rotors and per-
deuterated triphenylmethyl frames. Using 1-d30 as a test and
controlling the extent of cross polarization in the CPMAS
experiment, we were able to uncover the 13C signals involved
in chemical exchange, revealing a coalescence process that
was not detectable using samples with natural isotopic
abundance.{
1
phenylene H are more intense. The two alkyne signals illustrate
this clearly as the carbon attached to the central ring at 83.1 ppm is
twice as intense as that of the alkyne carbon at 97.1 ppm, which is
one bond length further away.
When the same sample was analyzed with a contact time of
50 ms, only the signals corresponding to C1 and C3 were detected
at 128.5 and 132.3 ppm (Fig. 2 bottom). Having successfully
removed all the signals of the aromatic trityl group by specific
deuteration, we were able to record changes in lineshape as a
function of temperature (Fig. 3) between 214 and 308 K.
Lineshape analysis of the isolated signals at eight temperatures
was carried with the program g-NMR assuming exchange rates
Steven D. Karlen and Miguel A. Garcia-Garibay*
Department of Chemistry and Biochemistry, University of California,
Los Angeles, California, 90095-1569, USA. E-mail: mgg@chem.ucla.edu
Notes and references
{ Selected spectroscopic data for 1-d30, 1H NMR (300 MHz): d 5 7.42 ppm
(s), 13C NMR (75 MHz) same as 1 except that carbon signals at d 5 128.69,
127.51, and 127.4 ppm are triplets with J(C–D) 5 14 Hz. IR: the stretch at
3030–3083 in 1 changes to an sp2-C–D stretching at 2356. A small peak at
3073 cm21 corresponds to the phenylene group.
{ Simulations were performed at 75 MHz with exchange between signals at
128.5 and 132.3 ppm assuming a Gaussian lineshape. A static line width of
ca. 115 Hz was estimated from measurements carried out at the lowest
temperature.
1 (a) Z. Dominguez, H. Dang, M. J. Strouse and M. A. Garcia-Garibay,
J. Am. Chem. Soc., 2002, 124, 7719–7727; (b) Z. Dominguez, H. Dang,
M. J. Strouse and M. A. Garcia-Garibay, J. Am. Chem. Soc., 2002, 124,
2398–2399.
2 (a) C. E. Godinez, G. Zepeda and M. A. Garcia-Garibay, J. Am. Chem.
Soc., 2002, 124, 4701–4704; (b) C. E. Godinez, G. Zepeda, C. J. Mortko,
H. Dang and M. A. Garcia-Garibay, J. Org. Chem., 2004, 69,
1652–1662.
3 (a) A. D. Bain, Prog. Nucl. Magn. Reson. Spectrosc., 2003, 43, 63; (b)
F. G. Riddel, K. S. Cameron, S. A. Holmes and J. H. Strange, J. Am.
Chem. Soc., 1997, 119, 7555; (c) L. Lunazzi, A. Mazzanti, D. Casarini,
O. D. Lucchi and F. Fabris, J. Org. Chem., 2000, 65, 883.
Fig. 3 Variable temperature 13C CPMAS NMR of 1-d30 with a contact
time of 50 ms (black), simulations from g-NMR in blue. The stars indicate
residual signals from the non-protonated phenylene carbon and one of the
trityl signals.
190 | Chem. Commun., 2005, 189–191
This journal is ß The Royal Society of Chemistry 2005

View Article Online
4 Horst Friebolin, Basic One- and Two-Dimensional NMR Spectroscopy,
Wiley-VCH, Weinheim, Germany, 1998.
7 C. A. Fyfe, Solid State NMR for Chemists; C.F.C. Press: Guelph,
Ontario, 1983.
5 Z. Doninguez, T.-A. V. Khuong, H. Dang, C. N. Sanrame, J. E. Nun˜ez
and M. A. Garcia-Garibay, J. Am. Chem. Soc., 2003, 125, 8827–8837.
6 C. R. Hauser and B. E. Hudson, Jr., Org. Synth. Coll. Vol. III, 1955,
824.
8 gNMR v. 5.0, Adept Scientific, Inc., Bethesda, Maryland, 2003.
9 (a) N. E. Heimer and D. L. Mattern, J. Am. Chem. Soc., 1993, 115,
2217–2220; (b) A. B. Jaffe, D. S. Malament, E. P. Slisz and J.
M. McBride, J. Am. Chem. Soc., 1972, 94, 8515–8521.
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 189–191 | 191