Angewandte
Research Articles
Chemie
increase in the fluorescence emission intensity but a slight
decrease of the FF value to 24.3 Æ 2.6%. The sequential
addition of Na+ to remove the acetate anions resulted in the
recovery of the emission intensity (Figure 4e and Figure S39),
indicating that such a reversible switching process did
influence the AIE behaviors of [3]rotaxane 1. Interestingly,
for both 1 and 2, as revealed by the fluorescence lifetimes that
were recorded by the time-correlated single-photon counting
(TCSPC) technique (Figure 4 f), a decrease of the fluores-
cence lifetimes upon the addition of TBAA (for 1, from
3.20 ns to 1.50 ns; for 2, from 3.22 ns to 2.74 ns, Figure S40 and
Table S1) was observed, suggesting anion-induced co-confor-
mational changes that influenced the energy of the excited
states. It is also worth noting that the fluorescence lifetimes
agree well with the mono-exponential decay model, indicat-
ing that the fluorescence originates from the sole excited state
and no competing radiative deactivation process occurs.[17]
In order to provide more information for aforementioned
stimuli-responsive AIE behaviors of 1, dynamic light scatter-
ing (DLS) was further employed to investigate the different
aggregation states of 1. According to the DLS results, no
aggregation of 1 was observed in pure CH2Cl2 solution with an
average hydrodynamic diameter (Dh) of 1.5 Æ 0.1 nm. When fh
was increased to 95%, the average Dh remarkably increased
to 65.5 Æ 6.3 nm, clearly suggesting the formation of nano-
scale aggregates. The sequential addition of TBAA to trigger
the switching of 1 led to a further increase of the average Dh to
122.0 Æ 8.6 nm, again confirming the anion-induced tuning of
the AIE behavior of 1. As expected, sequentially adding Na+
to drive the reversed switching process resulted in the almost
complete regeneration of the original aggregation state with
an average Dh of 72.5 Æ 4.7 nm (Figures S41).
Figure 5. a) Cartoon representation of the diastereospecific formation
of [3]rotaxane 1, b) chiral HPLC traces, and c) CD spectra (50 mM in
CH2Cl2) of the three fractions that were assigned to pS-1, meso-1, and
pR-1, respectively.
larger cavity of DEP[5]A in the (pS, pS, pS, pS, pS) and (pR,
pR, pR, pR, pR) conformers.[14e]
More importantly, according to both the CD spectra and
previous reports,[11b,14e] the absolute configurations of the
[3]rotaxane in the first fraction (pS-1) and the third fraction
(pR-1) were assigned to be pS and pR, respectively, in which
both wheels threaded on the axle are in (pS, pS, pS, pS, pS) or
(pR, pR, pR, pR, pR) conformers, respectively. While for the
[3]rotaxane in the second fraction (meso-1), as revealed by its
X-ray single crystal structure described above, one (pS, pS, pS,
pS, pS) wheel and one (pR, pR, pR, pR, pR) wheel were
located on the axle. Notably, due to the existence of the axle
in the cavity of DEP[5]A that blocks the free rotations of the
paraphenylene methylene units, all rotaxanes displayed high
configuration stability. No racemization was observed even
after prolonged heating (Figure S51), such stability makes
them excellent candidates for novel CPL materials.
Resolution of [3]Rotaxane 1
In principle, due to the existence of eight conformers (i.e.,
four diastereomers two enantiomers) of DEP[5]A, there
should be 36 isomers of [3]rotaxane 1 (Figure S45). However,
as revealed by the chiral HPLC measurements, three peaks
with a ratio of peak areas of 24.2:51.1:24.6 were observed,
preliminarily indicating the existence of only three isomers in
[3]rotaxane 1 (Figure 5a,b). After isolation of these fractions
1
using chiral preparative HPLC column, H NMR analysis of
each fraction was first performed. As shown in Figure S50, the
1H NMR spectrum of the first fraction is almost the same as
that of the third fraction, both of which are only slightly
different from that of the second fraction, suggesting the
isolation of the [3]rotaxane enantiomers. The CD analysis
further confirmed such hypothesis, as the CD peak patterns of
the first fraction and the third fraction mirrored each other in
both form and intensity, which is consistent with an enantio-
meric relationship. In addition, the second fraction revealed
no CD signal, indicating that it is the meso form of [3]rotaxane
1 (Figure 5c). The ratio of these three isomers was approx-
imately about 1:2:1, which matches the statistical distribu-
tions. According to these results, the formation of [3]rotaxane
1 was proven to be diastereospecifical to symmetric diaste-
reoisomers, possibly attributed to both the lower energy and
Switchable CPL of Chiral [3]Rotaxanes pS-1 and pR-1
With the targeted chiral [3]rotaxanes in hand, their
chiroptical properties were then evaluated. The solution-state
CD spectra of both pS-1 and pR-1 indicated that there was no
obvious chirality information transfer even after the addition
of acetate anions to trigger the DEP[5]A macrocycles to
relocate near to the DSA unit, as revealed by the fact that
only slight changes in the intensity of the bands at both
250 nm and 310 nm, which could be mainly attributed to the
DEP[5]A macrocycles, were observed (Figure S52). More-
over, as suggested by the CD spectra of the cast film of pS-
1 and pR-1 (Figure S53), bands at 250 nm and 310 nm were
observed, which is consistent with the solution state CD
spectra. However, a weak band at around 400 nm, which was
Angew. Chem. Int. Ed. 2021, 60, 9507 –9515
ꢀ 2021 Wiley-VCH GmbH
9511