3
-H‚PF
6
, while the smaller resonance arises from 4-H‚PF
6
.
observed in the spectra is the decrease in intensity of the
peak at δ ) 8.62 ppm (diimine dumbbell 5-H‚PF ) and the
corresponding growth of a new resonance at δ ) 8.37 ppm
(diimine [2]rotaxane [5-H‚DB24C8]‚PF ). It is apparent,
therefore, that since the diimine dumbbell 5-H‚PF cannot
thread through the cavity of DB24C8, the [2]rotaxane is
being generated via a process in which consumption of 5-H‚
Two singlets, one each for 4-H‚PF
6
and 5-H‚PF , are also
6
6
observed in the CHN region of the spectrum. During the
course of the reaction, both CHO signals decrease in their
intensities, concomitant with the increasing intensities ob-
served for the CHN peaks. Moreover, comparison of the
spectra shows that, although initially intermediate mono-
6
6
aldehyde/monoimine 4-H‚PF
ally, after 60 min, diimine 5-H‚PF
6
is the major product, eventu-
starts to predominate.
PF
6
results in the formation of either a monoaldehyde/
6
monoimine (4-H‚PF
6
) or dialdehyde (3-H‚PF ) thread which,
6
At equilibrium (1440 min ) 1 day), integration of the CHO/
CHN signals allows the product distribution, dialdehyde 3-H‚
in turn, can pass through the cavity of the macrocyclic
polyether and subsequently react to form the diimine [2]-
PF
6
(4%), monoaldehyde/monoimine 4-H‚PF
6
(23%), and
rotaxane [5-H‚DB24C8]‚PF
6
. At equilibrium (t ) 6200 min
1
5
diimine dumbbell 5-H‚PF
6
(73%), to be calculated.
) ca. 4.5 days), the product distribution is [5-H‚DB24C8]‚
The implication of this equilibrium composition is that,
on addition of 1 equiv of DB24C8 to the equilibrated
mixture, no more than 27% of it can be rotaxanated unless
thermodynamic control is operating. The results are recorded
in Figure 2. At t ) 0 min, with no crown ether present, the
PF
4-H‚PF
6
(47%), 5-H‚PF
6
(16%); [4-H‚DB24C8]‚PF
6
(26%),
6
(1%).
6
(6%); [3-H‚DB24C8]‚PF
6
(4%), 3-H‚PF
The same equilibrium proportions of these six species can
be reachedsbut much more quicklysin 2125 min (Figure
3) when DB24C8 is present in the reaction mixture from
1
Figure 3. Partial H NMR spectra (400 MHz, CD
3
CN, 300 K),
(20 mM), 3,5-
di-tert-butylaniline (40 mM), and DB24C8 (20 mM). Peaks
1
Figure 2. Partial H NMR spectra (400 MHz, CD
3
CN, 300 K),
(20 mM) and
,5-di-tert-butylaniline (40 mM), which was allowed to reach
equilibrium (spectrum at t ) 0 min) prior to the addition of DB24C8
recorded over time, of an initial mixture of 3-H‚PF
6
recorded over time, of an initial mixture of 3-H‚PF
6
3
+
corresponding to species containing an NH
2
center not bound/
occupied by a DB24C8 ring are represented in black while those
which are bound/occupied are represented in red. The resonances
associated with the dialdehyde, monoaldehyde, and diimine probe
protons are highlighted as described in Figure 1.
+
(
2
20 mM). Peaks corresponding to species containing an NH center
not bound/occupied by a DB24C8 ring are represented in black
while those which are bound/occupied are represented in red. The
resonances associated with the dialdehyde, monoaldehyde, and
diimine probe protons are highlighted as described in Figure 1.
the outset. This observation is yet another indication that
thermodynamic control is operating in this system, as the
same equilibrium composition is being reached, irrespective
of the starting point of the reaction. By utilizing the
secondary dialkylammonium ion/crown ether recognition
motif to bring about the noncovalent synthesis of a [2]-
pseudorotaxane, in conjunction with reversible imine bond
formation to stopper covalently the [2]pseudorotaxane and
produce a dynamic [2]rotaxane, we have demonstrated that
thermodynamic control can be used in the construction of
interlocked molecular compounds that are wholly organic
in nature. A high degree of control of superstructure and
structure can be envisaged in such “slowly” equilibrating
systems. For example, in Scheme 2, by judiciously altering
spectrum is identical with the one recorded in Figure 1 at
equilibrium (t ) 1440 min). Upon addition of DB24C8, any
species with an aldehyde group at either of its termini (i.e.,
3
-H‚PF
crown ether’s macrocyclic cavity, generating the correspond-
ing [2]pseudorotaxanesi.e., [3-H‚DB24C8]‚PF and [4-H‚
DB24C8]‚PF , respectively. This effect is seen immediately
t ) 10 min) and, subsequently, resonances corresponding
to 3-H‚PF , 4-H‚PF , [3-H‚DB24C8]‚PF , and [4-H‚DB24C8]‚
PF remain substantially unchanged throughout the course
of the experiment. By far the most predominant change
6
and 4-H‚PF
6
) is capable of threading through the
6
6
(
6
6
6
6
(15) Although by no means quantitative, the positive-ion FAB mass
1
6
spectrum of this equilibrium mixture tallies approximately with the ratios
solvent polarities, the equilibria in the “vertical” direction
1
of the components observed by H NMR spectroscopy in solution. The
intensities of the signals in the mass spectrum, corresponding to [5-H‚
+
+
+
+
DB24C8] :5-H :[4-H‚DB24C8] :4-H , occur in the ratio 100:32:11:5.
Signals are not observed for the minor solution components, namely [3-H‚
(16) Ashton, P. R.; Chrystal, E. J. T.; Glink, P. T.; Menzer, S.; Schiavo,
C.; Spencer, N.; Stoddart, J. F.; Tasker, P. A.; White, A. J. P.; Williams,
D. J. Chem. Eur. J. 1996, 2, 709-728.
+
+
DB24C8] and 3-H , respectively.
Org. Lett., Vol. 1, No. 9, 1999
1365