Enantioselective synthesis of chiral porphyrin macrocyclic hosts and kinetic enantiorecognition of viologen guests

Article abstract of DOI:10.1039/d0sc05233g

The construction of macromolecular hosts that are able to thread chiral guests in a stereoselective fashion is a big challenge. We herein describe the asymmetric synthesis of two enantiomericC₂-symmetric porphyrin macrocyclic hosts that thread and bind different viologen guests. Time-resolved fluorescence studies show that these hosts display a factor 3 kinetic preference (ΔΔG?on= 3 kJ mol^?1) for threading onto the different enantiomers of a viologen guest appended with bulky chiral 1-phenylethoxy termini. A smaller kinetic selectivity (ΔΔG?on= 1 kJ mol^?1) is observed for viologens equipped with small chiralsec-butoxy termini. Kinetic selectivity is absent when theC₂-symmetric hosts are threaded onto chiral viologens appended with chiral tails in which the chiral moieties are located in the centers of the chains, rather than at the chain termini. The reason is that the termini of the latter guests, which engage in the initial stages of the threading process (entron effect), cannot discriminate because they are achiral, in contrast to the chiral termini of the former guests. Finally, our experiments show that the threading and de-threading rates are balanced in such a way that the observed binding constants are highly similar for all the investigated host-guest complexes,i.e.there is no thermodynamic selectivity.

Full text of DOI:10.1039/d0sc05233g

Showcasing research from Professor Roeland Nolte’s
laboratory, Institute for Molecules and Materials, Radboud
University, Nijmegen, The Netherlands.
As featured in:
Enantioselective synthesis of chiral porphyrin macrocyclic
hosts and kinetic enantiorecognition of viologen guests
Chiral guests display kinetic stereoselective threading
through the cavity of a new chiral porphyrin cage (center)
if their chirality (yellow and red helices) is located at the
chain ends and not in the centers, supporting the previously
reported entron effect of threading.
See Johannes A. A. W. Elemans,
Roeland J. M. Nolte et al.,
Chem. Sci., 2021, 12, 1661.
rsc.li/chemical-science
Registered charity number: 207890

Chemical
Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
Enantioselective synthesis of chiral porphyrin
macrocyclic hosts and kinetic enantiorecognition
Cite this: Chem. Sci., 2021, 12, 1661
of viologen guests†
All publication charges for this article
have been paid for by the Royal Society
of Chemistry
Pieter J. Gilissen, ‡a Annemiek D. Slootbeek, ‡a Jiangkun Ouyang,
a
b
a
a
*
Nicolas Vanthuyne,
Rob Bakker,
Johannes A. A. W. Elemans
a
*
and Roeland J. M. Nolte
The construction of macromolecular hosts that are able to thread chiral guests in a stereoselective fashion
is a big challenge. We herein describe the asymmetric synthesis of two enantiomeric C2-symmetric
porphyrin macrocyclic hosts that thread and bind different viologen guests. Time-resolved fluorescence
studies show that these hosts display a factor 3 kinetic preference (DDG‡on ¼ 3 kJ molꢀ1) for threading
onto the different enantiomers of a viologen guest appended with bulky chiral 1-phenylethoxy termini. A
smaller kinetic selectivity (DDG‡on ¼ 1 kJ molꢀ1) is observed for viologens equipped with small chiral sec-
butoxy termini. Kinetic selectivity is absent when the C2-symmetric hosts are threaded onto chiral
viologens appended with chiral tails in which the chiral moieties are located in the centers of the chains,
rather than at the chain termini. The reason is that the termini of the latter guests, which engage in the
initial stages of the threading process (entron effect), cannot discriminate because they are achiral, in
contrast to the chiral termini of the former guests. Finally, our experiments show that the threading and
de-threading rates are balanced in such a way that the observed binding constants are highly similar for
all the investigated host–guest complexes, i.e. there is no thermodynamic selectivity.
Received 22nd September 2020
Accepted 22nd December 2020
DOI: 10.1039/d0sc05233g
rsc.li/chemical-science
trimethylammonium compounds.14 The reported receptors
showed different degrees of thermodynamic enantior-
Introduction
ecognition (up to 12-fold difference in binding constant Kassoc
for the binding of the two enantiomers). Studies on kinetic
enantiorecognition, such as differences in threading rate
constants (kon-values), however, have rarely been reported.15
Our research aims at encoding information into synthetic
polymers using a macrocyclic porphyrin catalyst based on the
glycoluril framework, i.e. Mn1 (Fig. 1) that can thread onto
a polymer containing alkene double bonds, e.g. a polybutadiene
chain, and processively epoxidize these double bonds.16–18 Since
we intend to write a binary code on said polymer in the form of
chiral epoxides ((R,R)-epoxide ¼ digit 0, (S,S)-epoxide ¼ digit 1),
our desired writing system requires a chiral variant of Mn1 that
is capable of performing sequential processive threading and
catalysis. Two important requirements for sequential processive
catalysis are spatiotemporal control and unidirectionality,19
both of which are determined by kinetic factors. We hypothe-
sized that unidirectional threading of polymer chain through
a catalytic machine might be accomplished by aligning chiral
structural information in both the catalyst and the polymer.
Therefore, it is crucial that the catalytic machine displays
a kinetic preference for one of the enantiomers of a chiral
(polymeric) guest. In earlier work, we reported on the post-
modication of the achiral porphyrin cage H2120 by providing
it with a nitro function, yielding a racemic mixture of planar
Chiral recognition and selection of substrate (guest) molecules
are well-established features of enzymes and natural recep-
tors.1,2 Inspired by these naturally occurring chiral hosts,
chemists have developed a variety of synthetic chiral macro-
molecular architectures that function as receptors. Examples
include macrocyclic arenes,3–8 cyclodextrins,9 and metal–
organic cages.10–13 Calixarenes have been reported that display
enantiorecognition towards chiral carboxylates3 and chiral
amines.4–6 Amino acid recognition has been achieved with cal-
ixarenes,4 cyclodextrins,9 and metal–organic cages.10 Chiral
metal–organic cages have also been used for the enantiosepa-
ration of small alcohols and carboxylic acids,11,12 and for the
separation of atropisomeric compounds, such as 1,10-bi-2-
naphthol (BINOL) derivatives.11 Triptycene-based hosts have
been
developed
for
the
recognition
of
chiral
aInstitute for Molecules and Materials, Radboud University, Heyendaalseweg 135,
6525 AJ Nijmegen, The Netherlands. E-mail: r.nolte@science.ru.nl; j.elemans@
science.ru.nl
bAix Marseille Univ, CNRS, Centrale Marseille, iSm2, Marseille, France
† Electronic supplementary information (ESI) available: Experimental procedures,
characterization data, kinetic data, DFT calculations, and copies of NMR spectra
of all new compounds. See DOI: 10.1039/d0sc05233g
‡ These authors contributed equally to this work.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci., 2021, 12, 1661–1667 | 1661

View Article Online
Chemical Science
Edge Article
reactions on 2-alkoxyphenols employing enantiopure
lactates.25–27
Hence, we designed a stepwise protocol for the introduction
of the chiral centers. The Mitsunobu reaction of 2-benzylox-
yphenol with ethyl (S)-lactate or ethyl (R)-lactate afforded esters
(R)-5 and (S)-5 in 62% and 72% yield, respectively. Aer quan-
titative removal of the benzyl protecting group, the other chiral
center was installed with a second Mitsunobu reaction. In this
way, diesters (R,R)-7 and (S,S)-7 were obtained in reasonable
yields (57% and 74%, respectively) and with excellent diaste-
reomeric ratios (>98 : 2). Both diesters 7 were reduced with
lithium aluminum hydride, to afford the corresponding
primary alcohols 8 in excellent yields. The primary alcohols
were then transformed into tosylate leaving groups, affording
(R,R)-9 (83%) and (S,S)-9 (99%). According to the established
procedure for the synthesis of achiral cage compound H21,28 the
N,O-acetals of compound 10 were activated with zinc chloride in
the presence of thionyl chloride, and the resulting N-acylimi-
nium intermediates were reacted with electron-rich arenes
(R,R)-9 and (S,S)-9 to afford chiral clip molecules (R,R,R,R)-11
(38%) and (S,S,S,S)-11 (36%). The syntheses of the two enan-
tiomers of H23 were completed by reacting chiral clips 11 under
highly dilute basic conditions with 1 equivalent of porphyrin
tetraol H212. Aer two chromatographic purication steps
(alumina followed by 60H silica), the enantiopure cage
compounds (R,R,R,R)-H23 (8%) and (S,S,S,S)-H23 (5%) were
obtained as purple solids. Electronic circular dichroism (ECD)
measurements (Fig. 2B) showed that the chirality of the exible
spacers is clearly transferred to the porphyrin, as opposite signs
of the CD signals were observed for the Soret bands (lmax ¼ 416
nm) of the enantiomeric hosts (R,R,R,R)-H23 and (S,S,S,S)-H23.
1H NMR analysis (Fig. 2C) revealed that the methyl groups of the
C2-symmetric macrocycles are in uncommon – highly shielded –
chemical environments, as their protons resonate at negative
chemical shi (assigned as protons a and b). This effect can be
explained by the close proximity of the shielding area of the
aromatic porphyrin ring, indicating that the methyl groups are
pointing into the cavity of the cage. 2D ROESY experiments
indicated the close proximity of the methyl groups a and b, as
we observed mutual NOE contacts. DFT calculations (Fig. 2D)
Fig. 1 Chemical structures of porphyrin macrocycles.
chiral nitro-functionalized porphyrin cage H22.21 The enantio-
mers of H22 could be resolved by chiral HPLC22 and by installing
a chiral auxiliary on the corresponding amino-functionalized
host.23 The enantiopure hosts investigated earlier were shown
to have small differences in affinity for chiral viologen guests
(up to 3-fold difference in binding constant Kassoc).22,23 In this
paper we report an alternative approach to synthesize a chiral
C2-symmetric porphyrin cage compound, i.e. a host in which
the porphyrin macrocycle is linked to the glycoluril framework
via chiral linkers (H23, Fig. 1). Furthermore, we show that the
enantiomers of H23 display a kinetic preference for the
threading and binding of viologen guests equipped with a chiral
head group.
Results and discussion
Synthesis and characterization
The synthetic route to the novel porphyrin macrocyclic hosts conrmed that the minimum energy structure of H23 is the one
(R,R,R,R)-H23 and (S,S,S,S)-H23 is depicted in Fig. 2. The key with the methyl groups pointing into the cavity of the porphyrin
synthesis step was the introduction of the chiral centers in macrocyclic host.
intermediate 7. Burkard and Effenberger reported the synthesis
of (R,R)-7 in one step from pyrocatechol and the very reactive
triate ester of ethyl (S)-lactate under basic conditions.24
Host–guest binding and threading studies
According to their communicated procedure, the resulting We reported earlier that H21 is an excellent host for the binding
diester (R,R)-7 was obtained diastereomerically pure. In our of viologen derivatives.20 Hence, we investigated the effect of the
hands, however, this procedure was not scalable and it resulted chiral spacers in the C2-symmetric hosts H23 on the binding
in an inseparable 94 : 6 diastereomeric mixture of (R,R)-7 and thermodynamics and the threading kinetics (Table 1) of various
the meso-compound (R,S)-7 (via epimerization of one of the achiral and chiral viologen guests (Fig. 3). We studied threading
chiral centers). To prevent epimerization, we resided to the by using approach-to-equilibrium uorescence quenching
Mitsunobu reaction. Initially, we attempted a double Mitsu- spectroscopy, in which we follow the uorescence intensity of
nobu reaction on pyrocatechol with ethyl (S)-lactate to afford the host as a function of time, as it is quenched upon binding
diester (R,R)-7 in a single step, which however failed. In contrast the viologen moiety (Fig. 4, for details see the ESI, Fig. S8–S55†).
to the lack of literature procedures for Mitsunobu reactions on The initial stages of this process follow second-order
pyrocatechol, there are successful examples of Mitsunobu kinetics.29,30 Subsequently, de-threading of the viologen guests
1662 | Chem. Sci., 2021, 12, 1661–1667
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
Fig. 2 Synthesis and characterization of H23. (A) Synthesis of H23; reagents and conditions: (a) ethyl (S)-lactate (for (R)-5) or ethyl (R)-lactate (for
(S)-5), DIAD, PPh3, toluene, 0 / 20 ꢂC; (b) H2, Pd/C, EtOAc, 20 ꢂC; (c) ethyl (S)-lactate (for (R,R)-7) or ethyl (R)-lactate (for (S,S)-7), DIAD, PPh3,
toluene, 0 / 20 ꢂC; (d) LiAlH4, THF, 0 / 20 ꢂC; (e) TsCl, pyridine, CH2Cl2, 0 / 20 ꢂC; (f) ZnCl2, SOCl2, CH2Cl2, 20 ꢂC; (g) K2CO3, CH3CN, reflux.
For clarity, the synthetic scheme only displays the structures of all (R)-derivatives; yields of all (S)-derivatives are indicated in square brackets. (B)
1
ECD spectra of (R,R,R,R)-H23 and (S,S,S,S)-H23 (c ¼ 3 ꢃ 10ꢀ5 M in CHCl3/CH3CN, 1 : 1, v/v). (C) Negative chemical shift region of the H NMR
spectrum of (R,R,R,R)-H23 (400 MHz, c ¼ 10ꢀ3 M in CDCl3/CD3CN, 1 : 1, v/v) with assignment of protons a–c. (D) Front view of the DFT-
optimized structure of (R,R,R,R)-H23 at the B3LYP/6-311+G(d) level. Color code: white ¼ hydrogen, grey ¼ carbon, blue ¼ nitrogen, red ¼
oxygen.
from the hosts was studied by dilution-induced uorescence Waals interactions between the bound guest and the spacer
recovery, for which the initial stages follow rst-order kinetics methyl groups of the host.
(see the ESI, Fig. S56–S103†).29,30 Binding constants were ob-
We then studied the threading of achiral polymer 14 through
tained by dividing the threading kon-values by the de-threading macrocyclic hosts H2130 and H23 (Table 1, entries 4–6). The kon
-
koff-values. For the smallest possible viologen, i.e. methyl viol- value for the threading of polymer 14 through H21 was 20 times
ogen 13, the kinetics of threading and de-threading were too higher than that for the threading through the two enantiomers
fast to measure, and therefore the binding constants of this of H23. In addition, the koff-value for the de-threading of poly-
guest in the enantiomers of host H23 were obtained by uo- mer 14 from H21 was 14 times higher than that for the de-
rescence titrations (see the ESI, Tables S4–S6†). The obtained threading of 14 from the enantiomers of H23. Even though
binding constants with guest 13 were slightly higher than the the kinetic processes of both threading through and de-
binding constant between 13 and the parent porphyrin threading from H21 are a lot faster compared to those deter-
macrocyclic host H21 (Table 1, entries 1–3).20 This result implies mined for H23, the on- and off-rates are balanced, resulting in
that the methyl groups, which are inside the cavity, do not have similar Kassoc values. The NMR spectra of the host–guest
a negative effect on the thermodynamics of the binding process. complexes of H23 with 14 again showed deshielding of methyl
In fact, a 1H NMR titration of (S,S,S,S)-H23 with viologen 13 protons a and b by ꢁ2 ppm units (see the ESI, Fig. S106†).
showed a remarkably large effect on the methyl protons a and
To investigate whether enantiomeric viologens with chiral
b of the chiral spacers (see the ESI, Fig. S104†). Upon binding N-substituents would exhibit stereoselective threading
the guest, the signals of these protons were deshielded by behavior through the cavities of the enantiomers of H23, we
ꢁ2 ppm units, indicating that the methyl groups orient them- examined the threading and de-threading processes of chiral
selves out of the cavity, thereby facilitating binding of the guest viologens (R,R)-15 and (S,S)-15 (Table 1, entries 7–12).23 The
inside the cavity (see the ESI, Fig. S105†). Apparently, this threading experiments (Fig. 4A and B) revealed that the enan-
structural reorganization of the host has no negative effect on tiomers of H23 display no kinetic preference for the binding of
the binding strength of 13. An explanation for the slightly either (R,R)-15 or (S,S)-15. Also the de-threading (see the ESI,
enhanced binding may be the presence of stabilizing van der Fig. S60–S67†) of these guests from both enantiomers of H23
occurred at the same rate. As a consequence, there is no
© 2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci., 2021, 12, 1661–1667 | 1663

View Article Online
Chemical Science
Edge Article
Table 1 Kinetic and thermodynamic data for porphyrin macrocyclic host–guest complexes acquired by fluorescence spectroscopy in CHCl3/
CH3CN (1 : 1, v/v) at 298 K
kon (Mꢀ1 sꢀ1
)
DG‡on (kJ molꢀ1
)
DDG‡on (kJ molꢀ1
)
koff (sꢀ1
)
DG‡off (kJ molꢀ1
)
Kassoce (Mꢀ1
)
DGassoc (kJ molꢀ1
)
d
d
Entry Guest
Host
1a
2b
3b
4c
13
13
13
14
14
14
H21
—
—
—
—
—
—
46.7
54.1
54.1
—
46.6
46.5
—
46.5
46.4
47.0
50.8
51.9
46.9
52.0
51.0
54.4
62.4
62.4
54.3
62.4
62.4
52.1
62.4
59.3
51.8
59.6
62.6
—
—
—
—
—
—
—
—
—
—
6.0 ꢃ 105
1.9 ꢃ 106
1.6 ꢃ 106
2.9 ꢃ 107
1.5 ꢃ 107
1.7 ꢃ 107
8.6 ꢃ 106
1.1 ꢃ 107
1.1 ꢃ 107
1.1 ꢃ 107
1.2 ꢃ 107
1.1 ꢃ 107
1.5 ꢃ 107
1.7 ꢃ 107
1.3 ꢃ 107
1.5 ꢃ 107
1.4 ꢃ 107
1.6 ꢃ 107
1.2 ꢃ 107
2.3 ꢃ 106
2.5 ꢃ 106
1.3 ꢃ 107
2.5 ꢃ 106
2.4 ꢃ 106
1.7 ꢃ 107
6.6 ꢃ 106
1.1 ꢃ 107
2.1 ꢃ 107
9.2 ꢃ 106
4.8 ꢃ 106
ꢀ33.0
ꢀ35.8
ꢀ35.4
ꢀ42.5
ꢀ41.0
ꢀ41.2
ꢀ39.6
ꢀ40.1
ꢀ40.2
ꢀ40.1
ꢀ40.4
ꢀ40.2
ꢀ41.0
ꢀ41.2
ꢀ40.6
ꢀ41.0
ꢀ40.8
ꢀ41.1
ꢀ40.4
ꢀ36.3
ꢀ36.5
ꢀ40.6
ꢀ36.5
ꢀ36.4
ꢀ41.3
ꢀ38.9
ꢀ40.1
ꢀ41.8
ꢀ39.7
ꢀ38.1
(R,R,R,R)-H23
(S,S,S,S)-H23
H21
4.0 ꢃ 104
1.4 ꢃ 10ꢀ3 89.3
1.3 ꢃ 10ꢀ4 95.2
1.2 ꢃ 10ꢀ4 95.3
5
(R,R,R,R)-H23 2.0 ꢃ 103
+0.0
6
(S,S,S,S)-H23 2.0 ꢃ 103
7b
8
(R,R)-15 H21
—
—
+0.1
—
—
(R,R)-15 (R,R,R,R)-H23 4.2 ꢃ 104
4.0 ꢃ 10ꢀ3 86.7
9
(R,R)-15 (S,S,S,S)-H23 4.4 ꢃ 104
3.9 ꢃ 10ꢀ3 86.7
10b
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
(S,S)-15 H21
—
—
+0.1
—
—
(S,S)-15 (R,R,R,R)-H23 4.4 ꢃ 104
3.7 ꢃ 10ꢀ3 86.9
4.1 ꢃ 10ꢀ3 86.6
2.3 ꢃ 10ꢀ3 88.0
4.7 ꢃ 10ꢀ4 92.0
3.7 ꢃ 10ꢀ4 92.6
2.4 ꢃ 10ꢀ3 87.9
3.4 ꢃ 10ꢀ4 92.8
4.4 ꢃ 10ꢀ4 92.1
1.5 ꢃ 10ꢀ4 94.9
3.1 ꢃ 10ꢀ5 98.7
2.9 ꢃ 10ꢀ5 98.9
1.4 ꢃ 10ꢀ4 94.9
2.8 ꢃ 10ꢀ5 99.0
3.0 ꢃ 10ꢀ5 98.8
2.6 ꢃ 10ꢀ4 93.4
1.1 ꢃ 10ꢀ5 101.3
2.3 ꢃ 10ꢀ5 99.4
2.4 ꢃ 10ꢀ4 93.6
2.4 ꢃ 10ꢀ5 99.3
1.4 ꢃ 10ꢀ5 100.7
(S,S)-15 (S,S,S,S)-H23 4.5 ꢃ 104
(R,R)-16 H21
3.6 ꢃ 104
—
ꢀ1.1
(R,R)-16 (R,R,R,R)-H23 7.8 ꢃ 103
(R,R)-16 (S,S,S,S)-H23 4.9 ꢃ 103
(S,S)-16 H21
3.7 ꢃ 104
—
+1.0
(S,S)-16 (R,R,R,R)-H23 4.7 ꢃ 103
(S,S)-16 (S,S,S,S)-H23 7.0 ꢃ 103
(R,R)-17 H21
1.8 ꢃ 103
—
+0.0
(R,R)-17 (R,R,R,R)-H23 7.2 ꢃ 101
(R,R)-17 (S,S,S,S)-H23 7.1 ꢃ 101
(S,S)-17 H21
1.9 ꢃ 103
—
+0.0
(S,S)-17 (R,R,R,R)-H23 7.1 ꢃ 101
(S,S)-17 (S,S,S,S)-H23 7.2 ꢃ 101
(R,R)-18 H21
4.5 ꢃ 103
—
+3.0
(R,R)-18 (R,R,R,R)-H23 7.2 ꢃ 101
(R,R)-18 (S,S,S,S)-H23 2.5 ꢃ 102
(S,S)-18 H21
5.1 ꢃ 103
—
ꢀ2.9
(S,S)-18 (R,R,R,R)-H23 2.2 ꢃ 102
(S,S)-18 (S,S,S,S)-H23 6.7 ꢃ 101
a
b
c
d
e
Values taken from ref. 20. Determined by uorescence titration. Values taken from ref. 30. Estimated error 20%. Kassoc ¼ kon/koff, unless
stated otherwise, estimated error 30%. See the ESI for details.
similar to those of H21$15 (see the ESI, Tables S7–S13†). The 1H
NMR spectra of all combinations of host–guest complexes
H23$15 were nearly fully superimposable (see the ESI,
Fig. S107†). The spectra again showed deshielding of the
methyl groups of the chiral spacers of the hosts, as we also
observed for the host–guest complexes of H23 with achiral
guests. We propose that the lack of kinetic selectivity of chiral
hosts H23 for threading onto chiral guests 15 arises from the
absence of chirality at the termini of the viologen substituents,
i.e. the part of the guest molecule that has to initiate the
threading process. Previously, we have shown that the thread-
ing of polymeric viologens through H21 involves a kinetically
favorable ‘entron’ effect, which involves an internal lling of
the cavity of the porphyrin macrocyclic host by the rst 5–8
atoms of the polymer chain.30 Starting from the chain termini,
the stereogenic centers of viologens 15 only appear at the 6th
chain atom. We hypothesize that the remote location of the
Fig. 3 Chemical structures of viologen guests 13–18.
chirality in the chain inhibits the kinetic stereoselectivity of the
threading process, and that the initial entron effect is only
governed by the threading of the achiral isopropyl termini of
the chains of 15, which are identical for both enantiomers.
Alternatively, the lack of stereoselectivity could be due to the
lack of steric bulk near the chiral centers of guest 15.
thermodynamic preference for the binding of the chiral guests.
The binding constants of all complexes H23$15 were veried
independently by uorescence titrations and the values were
1664 | Chem. Sci., 2021, 12, 1661–1667
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
Fig. 4 Threading of porphyrin macrocyclic hosts H23 onto chiral viologens 15–18 in CHCl3/CH3CN (1 : 1, v/v) at 298 K. Normalized fluorescence
intensity of hosts H23 as a function of time after the addition (t ¼ 50 s) of 1 equivalent of guest (A) (R,R)-15 (c ¼ 2.5 ꢃ 10ꢀ7 M), (B) (S,S)-15 (c ¼ 2.5 ꢃ
10ꢀ7 M), (C) (R,R)-16 (c ¼ 3 ꢃ 10ꢀ6 M), (D) (S,S)-16 (c ¼ 3 ꢃ 10ꢀ6 M), (E) (R,R)-17 (c ¼ 10ꢀ5 M), (F) (S,S)-17 (c ¼ 10ꢀ5 M), (G) (R,R)-18 (c ¼ 10ꢀ5 M), and
(H) (S,S)-18 (c ¼ 10ꢀ5 M).
To investigate these hypotheses further, we designed and was threaded (Fig. 4H). The results indicate that the chiral
synthesized chiral guest molecules 16–18 (see the ESI, Scheme environment of the chain termini attached to the viologen
S1†). Guest 16 contains small sec-butoxy chiral groups at the moiety determines the extent of kinetic selectivity. Interestingly,
chain termini and guest 17 contains bulky 1-(4-octylphenyl) de-threading of guests 16–18 from hosts H23 (see the ESI,
ethoxy chiral groups in the center of the chains. Guest 18 Fig. S68–S103†) followed the same trends as observed for the
contains a viologen binding station appended at the chain corresponding threading processes, i.e. a kinetic preference in
termini with chiral 1-phenylethoxy moieties.
the threading process is associated with a kinetic preference in
We found that guests 16 (Table 1, entries 13–18) display the de-threading process. As a result, the thermodynamic
a kinetic preference of a factor 1.6 (DDG‡on ¼ 1 kJ molꢀ1) for selectivity for all diastereomeric complexes of guests 16–18 with
threading through the enantiomers of H23, i.e. (R,R)-16 prefers hosts H23 is low, i.e. all binding constants are similar. More-
(R,R,R,R)-H23 over (S,S,S,S)-H23 and (S,S)-16 prefers (S,S,S,S)- over, all the host–guest complexes H23$16, H23$17, and H23$18
1
H23 over (R,R,R,R)-H23 (Fig. 4C and D). The small selectivity is display similar H NMR spectra (see the ESI, Fig. S108–S110†).
attributed to the location of the small chiral moieties at the The methyl protons a and b are again deshielded by ꢁ2 ppm
termini of the guests.
units. In addition, the aromatic viologen protons are shielded
Then, we investigated guests 17 (Table 1, entries 19–24), by 2–4 ppm units. The complexation induced shis (CIS-values)
which contain bulky chiral moieties in locations remote from for the protons of the macrocyclic hosts H23 and the protons of
the chain termini. We found that threading of hosts H23 onto the guests 16–18 near the viologen binding station are nearly
guests 17 proceeded very slowly, i.e. 1–3 orders of magnitude identical for all host–guest complexes H23$16, H23$17, and
slower than measured for the other investigated guests. This H23$18 (see the ESI, Tables S65–S67†). Finally, for all the
deceleration is caused by the steric bulk in the chains of the investigated guests 16–18, the kinetic events of threading and
guests. The threading experiments (Fig. 4E and F) also indicated de-threading involving H21 occur faster than those involving
that guests 17 do not display kinetic selectivity, which is line H23, but the resulting thermodynamic equilibria remain mostly
with the experiments involving guests 15. Hence, for such unaffected.
selectivity the location of the chiral moiety at the terminus of
the guest is pivotal.
Conclusions
In a nal set of experiments, we investigated the threading of
guests 18 (Table 1, entries 25–30), in which the bulky chiral
moieties are positioned at the chain termini. The threading
rates of guests 17 and 18 were similar, which is expected based
We have successfully synthesized chiral C2-symmetric
porphyrin macrocyclic hosts (R,R,R,R)-H23 and (S,S,S,S)-H23 in
an enantioselective fashion from the naturally abundant
on the similar degree of steric bulk. More interestingly, (R,R)-18
compounds ethyl (S)-lactate and ethyl (R)-lactate, respectively.
showed a factor 3 (DDGo‡n ¼ 3 kJ molꢀ1) kinetic preference for
The chiral information present in the exible spacers of
threading host (S,S,S,S)-H23 over (R,R,R,R)-H23 (Fig. 4G). The
(R,R,R,R)-H23 and (S,S,S,S)-H23 is effectively transferred
opposite kinetic selectivity was observed when guest (S,S)-18
throughout the entire macrocycle up to the porphyrin, as was
© 2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci., 2021, 12, 1661–1667 | 1665

View Article Online
Chemical Science
Edge Article
evidenced by CD spectroscopy and conrmed by DFT calcula-
tions. The macrocyclic hosts display the characteristic features
of the achiral macrocycle H21, such as binding and threading of
viologen guests. While the thermodynamics of binding achiral
guests in H21 and H23 are very similar, the underlying kinetics
are completely different, since threading of the guests through
H23 is signicantly slower than threading through H21. This
deceleration allows for a more controlled threading process. As
we aim to store data on polymers via catalytic reactions on
a polymer chain, gaining spatio-temporal control is highly
desired. Another step in this direction is the realization of
diastereoselectivity in the threading process. Chiral viologen
guests 16 and 18, equipped with chiral sec-butoxy and 1-phe-
nylethoxy termini, respectively, display a signicant kinetic
preference (DDG‡on ¼ 1–3 kJ molꢀ1) for binding in one enan-
tiomer of H23 over the other. In contrast, dihydrocitronellyl-
2 M. K. Gilson and H.-X. Zhou, Annu. Rev. Biophys. Biomol.
Struct., 2007, 36, 21–42.
3 Y.-S. Zheng and C. Zhang, Org. Lett., 2004, 6, 1189–1192.
4 L. Mutihac, J. H. Lee, J. S. Kim and J. Vicens, Chem. Soc. Rev.,
2011, 40, 2777–2796.
5 J. Luo, Q.-Y. Zheng, C.-F. Chen and Z.-T. Huang, Tetrahedron,
2005, 61, 8517–8528.
6 Y. Kubo, S. Maeda, S. Tokita and M. Kubo, Nature, 1996, 382,
522–524.
´
´
7 M. Tlust´y, P. Slavık, M. Kohout, V. Eigner and P. Lhotak, Org.
Lett., 2017, 19, 2933–2936.
8 S. J. Nemat, H. J˛edrzejewska, A. Prescimone, A. Szumna and
K. Tiefenbacher, Org. Lett., 2020, 22, 5506–5510.
9 Q. Huang, L. Jiang, W. Liang, J. Gui, D. Xu, W. Wu, Y. Nakai,
M. Nishijima, G. Fukuhara, T. Mori, Y. Inoue and C. Yang, J.
Org. Chem., 2016, 81, 3430–3434.
appended viologens 15 and 1-(4-octylphenyl)ethoxy-appended 10 W. Xuan, M. Zhang, Y. Liu, Z. Chen and Y. Cui, J. Am. Chem.
viologens 17 do not display such a kinetic preference. The Soc., 2012, 134, 6904–6907.
reason for this difference in kinetic stereoselectivity is that the 11 K. Wu, K. Li, Y.-J. Hou, M. Pan, L.-Y. Zhang, L. Chen and
chirality of guests 16 and 18 is located at the termini of their C.-Y. Su, Nat. Commun., 2016, 7, 10487.
chain substituents, which are involved in the initial interaction 12 T. Liu, Y. Liu, W. Xuan and Y. Cui, Angew. Chem., Int. Ed.,
with the chiral cavities of hosts H23. The rate of the subsequent 2010, 49, 4121–4124.
threading process is dictated by the relative stereochemistry of 13 L.-J. Chen, H.-B. Yang and M. Shionoya, Chem. Soc. Rev.,
host and guest. The termini of guests 15 and 17, however, are 2017, 46, 2555–2576.
achiral, and therefore the initial interactions of these termini 14 G.-W. Zhang, P.-F. Li, Z. Meng, H.-X. Wang, Y. Han and
with the cavities of the enantiomers of hosts H23 are similar. C.-F. Chen, Angew. Chem., Int. Ed., 2016, 55, 5304–5308.
Therefore, we conclude that chirality at the chain terminus of 15 B.-Y. Wang, S. Stojanovic, D. A. Turner, T. L. Young,
´
´
the viologen guest is a prerequisite for realizing kinetic stereo-
C. M. Hadad and J. D. Badjic, Chem.–Eur. J., 2013, 19,
selectivity. This conclusion ts with the previously observed
4767–4775.
‘entron’ effect, which states that an initial binding event 16 S. Varghese, J. A. A. W. Elemans, A. E. Rowan and
between the chain terminus of the guest inside the cavity of the R. J. M. Nolte, Chem. Sci., 2015, 6, 6050–6058.
host dictates the rate of the threading process.30 A bulky chiral 17 M. G. T. A. Rutten, F. W. Vaandrager, J. A. A. W. Elemans and
moiety at the chain terminus further enhances the kinetic R. J. M. Nolte, Nat. Rev. Chem., 2018, 2, 365–381.
stereoselectivity, either by steric interactions or by favourable 18 J. A. A. W. Elemans and R. J. M. Nolte, Chem. Commun., 2019,
supramolecular interactions of the phenyl substituent. We are 55, 9590–9605.
currently investigating manganese(III) derivatives of (R,R,R,R)- 19 X. Wang, Q. Gan, B. Wicher, Y. Ferrand and I. Huc, Angew.
H23 and (S,S,S,S)-H23, which may potentially be used as proc- Chem., Int. Ed., 2019, 58, 4205–4209.
essive catalysts for the stereoselective epoxidation of double 20 J. A. A. W. Elemans, M. B. Claase, P. P. M. Aarts, A. E. Rowan,
bond-containing chiral polymer chains.
A. P. H. J. Schenning and R. J. M. Nolte, J. Org. Chem., 1999,
64, 7009–7016.
21 S. Varghese, B. Spierenburg, A. Swartjes, P. B. White,
P. Tinnemans, J. A. A. W. Elemans and R. J. M. Nolte, Org.
Lett., 2018, 20, 3719–3722.
Conflicts of interest
There are no conicts to declare.
22 J. Ouyang, A. Swartjes, M. Geerts, P. J. Gilissen, D. Wang,
P. C. P. Teeuwen, P. Tinnemans, N. Vanthuyne,
S. Chentouf, F. P. J. T. Rutjes, J.-V. Naubron, J. Crassous,
J. A. A. W. Elemans and R. J. M. Nolte, Nat. Commun.,
2020, 11, 4776.
23 S. Varghese, B. Spierenburg, J. P. J. Bruekers, A. Swartjes,
P. B. White, J. A. A. W. Elemans and R. J. M. Nolte, Eur. J.
Org. Chem., 2019, 3525–3533.
Acknowledgements
This work was funded by the European Research Council (ERC
Advanced Grant No. 74092 to R. J. M. N.) and by the Dutch
Ministry of Education, Culture, and Science (Gravitation
program 024.001.035). We thank Prof. Floris P. J. T. Rutjes for
discussions and his interest in this work.
24 U. Burkard and F. Effenberger, Chem. Ber., 1986, 119, 1594–
1612.
¨
25 S. Muller, C. M. Afraz, R. De Gelder, G. J. A. Ariaans,
References
B. Kaptein, Q. B. Broxterman and A. Bruggink, Eur. J. Org.
Chem., 2005, 1082–1096.
1 A. Stank, D. B. Kokh, J. C. Fuller and R. C. Wade, Acc. Chem.
Res., 2016, 49, 809–815.
1666 | Chem. Sci., 2021, 12, 1661–1667
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
26 Y. Nishiyama, Y. Han-ya, S. Yokoshima and T. Fukuyama, J. 29 R. G. E. Coumans, J. A. A. W. Elemans, R. J. M. Nolte and
Am. Chem. Soc., 2014, 136, 6598–6601.
27 S. Chacko, H. I. M. Boshoff, V. Singh, D. M. Ferraris,
A. E. Rowan, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
19647–19651.
D. R. Gollapalli, M. Zhang, A. P. Lawson, M. J. Pepi, 30 A. B. C. Deutman, C. Monnereau, J. A. A. W. Elemans,
A. Joachimiak, M. Rizzi, V. Mizrahi, G. D. Cuny and
L. Hedstrom, J. Med. Chem., 2018, 61, 4739–4756.
G. Ercolani, R. J. M. Nolte and A. E. Rowan, Science, 2008,
322, 1668–1671.
28 P. J. Gilissen, A. Swartjes, B. Spierenburg, J. P. J. Bruekers,
P. Tinnemans, P. B. White, F. P. J. T. Rutjes, R. J. M. Nolte
and J. A. A. W. Elemans, Tetrahedron, 2019, 75, 4640–4647.
© 2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci., 2021, 12, 1661–1667 | 1667