A supramolecular sorting hat: Stereocontrol in metal-ligand self-assembly by complementary hydrogen bonding

Article abstract of DOI:10.1002/anie.201405242

A combination of self-complementary hydrogen bonding and metal-ligand interactions allows stereocontrol in the self-assembly of prochiral ligand scaffolds. A unique, non-tetrahedral M₄L₆ structure is observed upon multicomponent self-assembly of 2,7-diaminofluorenol with 2-formylpyridine and Fe(ClO₄)₂. The stereochemical outcome of the assembly is controlled by self-complementary hydrogen bonding between both individual ligands and a suitably sized counterion as template. This hydrogen-bonding-mediated stereoselective metal-ligand assembly allows the controlled formation of nonsymmetric discrete cage structures from previously unexploited ligand scaffolds. Abracadabra: The stereoselective self-assembly of an unsymmetrical metal-ligand cage can be controlled by self-complementary hydrogen bonding between alcohol-containing ligands as well as between ligands and suitable anion guests.

Full text of DOI:10.1002/anie.201405242

.
Angewandte
Communications
DOI: 10.1002/anie.201405242
Self-Assembly
A Supramolecular Sorting Hat: Stereocontrol in Metal–Ligand Self-
Assembly by Complementary Hydrogen Bonding**
Michael C. Young, Lauren R. Holloway, Amber M. Johnson, and Richard J. Hooley*
Abstract: A combination of self-complementary hydrogen
bonding and metal–ligand interactions allows stereocontrol in
the self-assembly of prochiral ligand scaffolds. A unique, non-
tetrahedral M4L6 structure is observed upon multicomponent
self-assembly of 2,7-diaminofluorenol with 2-formylpyridine
and Fe(ClO4)2. The stereochemical outcome of the assembly is
controlled by self-complementary hydrogen bonding between
both individual ligands and a suitably sized counterion as
template. This hydrogen-bonding-mediated stereoselective
metal–ligand assembly allows the controlled formation of
nonsymmetric discrete cage structures from previously unex-
ploited ligand scaffolds.
of a second controlling element that favors the formation of
self-assembled constructs: complementary hydrogen bonds.
Hydrogen bonding is well-precedented as a strategy of
multicomponent self-assembly,[7] but the combination of both
hydrogen bonding and metal–ligand interactions as integral
components in controlling self-assembly is rare. A limited
number of polygons[8] and larger aggregates[9] have been
formed, and self-assembled cages are known that use hydro-
gen bonding to aid guest binding.[10] Other contributions such
as steric hindrance[11] and guest templation can dramatically
alter the outcome of the self-assembly process,[10] but the use
of hydrogen bonds to control and alter the stereochemical
outcome of complex metal-mediated self-assembly is unex-
plored. Here we describe the formation and anion-binding
properties of a novel polyhedral structural motif that forms
through both metal–ligand- and hydrogen-bond-mediated
self-assembly.
Using hydrogen bonding as a control element necessarily
limits the scope of the metal–ligand contacts used to mediate
assembly: cage complexes that are only soluble in water or
DMSO cannot exploit weak H bonds due to solvent com-
petition. Fortunately, the assembly of 2-iminopyridine ligands
with FeII centers occurs readily in aprotic solvents such as
acetonitrile. We recently reported the formation of a self-
assembled M2L3 meso-helicate complex derived from diami-
no-dibenzosuberol that contained alcohol groups on the
interior of the assembly.[12] The bent suberol scaffold formed
an assembly of M2L3 stoichiometry with controlled stereo-
chemistry. An M2L3 assembly is quite simple in structure, and
with only a small cavity, stereocontrol is relatively easy. On
the other hand, formation of an M4L6 assembly is vastly more
complex if a more nonsymmetrical ligand such as 2 is used. In
this case, one could imagine more than 100 different isomers if
a truly tetrahedral M4L6 structure occurred (Figure 1; see the
Supporting Information (SI) for a more detailed analysis).
These isomers arise from the possibility of metal-centered
isomerism (individual fac and mer possibilities lead to fac4,
fac3·mer1 isomers, etc.) as well as D or L stereoisomerism. In
addition, further isomerism is possible by orientation of the
CHOH ligand face “in” and “out” of the cage, as well as by
T
he central tenet in strategically planning the construction of
molecular polyhedra by metal–ligand self-assembly is to
maximize the symmetry and rigidity of the constituent
ligands.[1] To apply a greater scope of function to self-
assembled cages, however, ligands with appended functional
groups are required. Functionalized highly symmetric ligands
are accessible,[2] but there are few simple scaffolds that are
capable of both self-assembly and function, especially with
regard to their reactivity once assembled.[3] If less symmet-
rical ligands could be employed in self-assembly, a toolbox of
functionalized species for use in the creation of biomimetic
self-assembled architectures would be achievable.[4] One
roadblock to this goal is that many methods of metal–
mediated self-assembly involve octahedral metals as struc-
tural vertices, and this introduces the problem of metal-
centered isomerism.[5] For example, symmetrical linear imi-
nopyridine ligands such as 1 form Fe4L6 tetrahedra that
consist of four fac metal centers. The relative stereochemistry
is controlled: short, rigid ligands form only one isomer upon
assembly, and longer, more flexible ligands form a mixture of
three isomers, where both D and L configurations are
observed at the four metal centers, leading to cages of T, S4,
and C3 symmetry.[5a] This isomerism can be controlled in
certain circumstances,[6] but the use of nonlinear ligands that
can display multiple orientations upon assembly introduces
daunting complexity to the analysis of a system. Here we
describe a solution to this problem through the introduction
À
the introduction of possible diastereomers at the CH OH
center upon self-assembly into a chiral environment. Obvi-
ously, not all of the possible isomers would form, but a highly
complex, essentially unassignable mixture of numerous cages
could be expected upon self-assembly.
[*] M. C. Young, L. R. Holloway, A. M. Johnson, Prof. R. J. Hooley
University of California Riverside, Department of Chemistry
Riverside, CA 92521 (USA)
E-mail: richard.hooley@ucr.edu
To answer the question of whether simple alcohols could
provide control of the assembly by self-complementary
hydrogen bonding, the dianiline precursor A was targeted,
allowing the analysis of the assembly properties of ligand 2
(Figure 2, see SI for synthetic procedures). Selective bis-
[**] This research was supported by the NSF (CHE-1151773 to R.J.H.).
M.C.Y. and A.M.J. acknowledge UC Riverside for support. Prof.
Byron Purse is thanked for stimulating discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201405242.
9832
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9832 –9836

Angewandte
Chemie
Figure 1. Possibilities for the metal-mediated assembly of both symme-
trical[5a] and desymmetrized FeII-iminopyridine cages [Fe4·L6]8+
.
nitration[13] of fluorenone gave the 2,7-dinitro compound in
41% yield, followed by exhaustive reduction with Raney
nickel and hydrazine to give A in 61% yield.[14] Multicompo-
nent self-assembly of A was performed by treatment with 2-
formylpyridine and iron(II)perchlorate. After heating in
CH3CN at 608C for 3 h, a deep purple solution was formed,
which upon treatment with diethyl ether led to the precip-
itation of a purple solid. The resulting complex was soluble in
CD3CN and could be analyzed by NMR spectroscopy. The
1H NMR spectrum of the assembly is shown in Figure 2a, and
although complex, the peaks are sharp and correspond to
a cage assembly rather than to nondiscrete aggregates. The
peaks displayed a single diffusion constant in a DOSY NMR
experiment, indicating that only one stoichiometry of assem-
bly is formed, and the presence of an M4L6 complex was
observed by ESI-MS. The NMR spectrum did not appreciably
change upon varying the temperature from 758C to À408C.
X-ray quality crystals of [Fe4·26]8+ were obtained by slow
evaporation from a saturated solution in CD3CN, and the
solid-state structure was determined by X-ray diffraction
analysis (Figure 2b,c). The solution of the structure was
challenging: the crystals were rotationally twinned and
contained seven highly disordered perchlorate counterions,
not to mention numerous disordered solvent molecules. The
structure of the M4L6 cage complex is quite unusual; a true
tetrahedral arrangement is not formed and the observed cage
assembly bears a striking resemblance to a wizardꢀs hat. This
analogy is strengthened by the presence of a single perchlo-
rate ion at the base of the cage: the assembly rests atop of it.
The complex displays two different types of ligand position-
ing: the “base” ligands are on the outside of the structure,
whereas the “axial” ligands occupy the interior. The four FeII
centers show different stereochemical arrangements (as
opposed to the identical arrangements common in standard
tetrahedra):[15] the iron center at the “peak” displays a L-fac
orientation, whereas the other three centers at the base
display the rarely observed D-mer configuration.[16] The three
ligands at the lower base of the assembly are oriented so that
the alcoholic groups point upwards toward the “peak”,
whereas the three vertical ligands are positioned inside the
base, and their alcohol units point toward the central cavity.
Figure 2. Characterization of [Fe4·26]8+: a) 1H NMR and 2D DOSY spec-
tra of [(ClO4)ꢀFe4·26]7+ (CD3CN, 400 MHz, 298 K; D=100 ms,
d=2.6 ms, diffusion coefficient=4.8ꢀ10À10 m2sÀ1). *=minor peaks
from other isomers. b) Crystal structure of the [(ClO4)ꢀFe4·26]7+ cage
showing the presence of both L-fac and D-mer iron(II) centers and the
bound perchlorate guest. c) Cropped views of the X-ray crystal struc-
ture of [(ClO4)ꢀFe4·26]7+ showing: hydrogen bonding between ligand
alcohols and the bound perchlorate and hydrogen bonds between
different ligand alcohol groups. d) NOE contacts observed by NOESY
analysis of the solution-phase structure (see SI for full assignment).
The structure displays pseudo-C3 symmetry: whereas at first
glance the structure appears to possess a C3 rotation access
Angew. Chem. Int. Ed. 2014, 53, 9832 –9836
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9833

.
Angewandte
Communications
through the L-fac Fe center, the three Fe–Fe distances in the
base are not identical in the solid-state structure.
diffusion coefficient as the parent complex, and most likely
À
belong to different orientations of the CH OH groups in the
Not only is the stereochemistry at the metal centers
controlled, but also that of the prochiral ligands. In the chiral
environment of the assembly, the CHOH group in the center
of the fluorenol ligands becomes prochiral, and each ligand
could orient the hydroxy group either “in” or “out” of the
cage core. This does not occur in this assembly: each OH
group is oriented toward the center of the prism structure, and
this provides an explanation for the stereoselectivity of the
assembly.
The cavity of the prism is not particularly large, and the
internally pointing OH groups are capable of self-comple-
mentary hydrogen bonding with other OH groups on adjacent
ligands. In addition, the single bound perchlorate ion on the
interior of the cavity exerts a templating effect through
hydrogen bonding with the alcohols at the base.
ligands of [(ClO4)ꢀFe4·26]7+. Even so, considering the daunt-
ing number of isomer possibilities for this complex assembly
and the weak nature of the directing hydrogen bond contacts,
this selectivity is remarkable.
It is noteworthy that the ClO4À ion does not completely fill
the cage interior; although it occupies the base of the cage, the
peak appears unfilled. Cocrystallized solvent molecules of
acetonitrile were highly disordered in the solid-state struc-
ture, and were not observed in the cavity. In solution,
however, there is evidence from diffusion analysis for
a weak association of the acetonitrile solvent with the cage
(see SI). The 2D DOSY NMR spectrum of [(ClO4)ꢀFe4·26]7+
shows two diffusion constants for acetonitrile: one corre-
sponding to the bulk solvent and one that codiffuses with the
cage, suggesting either co-encapsulation of solvent with the
anion, or at least a strong interaction between the assembly
and a small number of solvent molecules.
Figure 2c shows the two different types of H bonding
present in the structure: the view from below shows three of
À
the oxygen atoms from the bound ClO4 ion hydrogen
The presence of a bound perchlorate in the cavity is
unsurprising, considering that polycationic cages are well-
precedented to show strong anion binding, especially in the
solid state.[17] The structure of the perchlorate anion allows
a strong templating effect, as it is perfectly positioned to
interact with the internal alcohol functions. Other counterions
were not as effective, however. When A was treated with
Fe(OTf)2 and 2-formylpyridine, a discrete complex was not
observed in solution. In contrast, a broad, undefined spectrum
was observed upon 1H NMR analysis in CD3CN (Figure 3b).
bonding with the OH groups at the base of the cage. The H–O
distances are 2.2 ꢁ and the perchlorate ion fully occupies the
base of the assembly. The hydrogen-bonding network is not
merely an anion templation effect: the axial ligands are in
close proximity to each other, and additional hydrogen bonds
are present between OH groups on the axial and base ligands,
rather than to the bound perchlorate. This combination of
both self-complementary and templating hydrogen bonds can
only occur in the conformation shown by [Fe4·26]8+. Other
conformations would point the OH groups away from the
center of the cage, preventing favorable hydrogen bonds. In
the case of a truly tetrahedral arrangement, the resultant
cavity would be too large to allow templation by small anions.
While the majority of the energy of formation upon self-
assembly comes from favorable M–L interactions in the Fe-
iminopyridine units, the discrimination between different
isomers comes from favorable hydrogen bonding.
Whereas only one isomer was observed in the solid state,
this could conceivably be an artifact of favorable crystalliza-
tion of the major product. At first glance, the complex NMR
spectrum may suggest the presence of an equal distribution of
multiple isomers formed upon assembly of ligand 2, but
further analysis indicates otherwise. Even though the
1H NMR spectrum of [(ClO4)ꢀFe4·26]7+ is complex, it can
be assigned. The nonsymmetric nature of the assembly leads
to most of the ligand protons occupying different magnetic
environments and the significant majority of peaks in Fig-
ure 2a correspond to individual protons in one assembly.
2D NMR experiments (COSY, TOCSY, HSQC, and
NOESY) were performed to allow the assignment. The full
solution-phase assignment is described in the Supporting
Information, but the notable NOE contacts corroborating the
solid-state structure are shown in Figure 2d. There are some
minor peaks in the 1H NMR spectrum corresponding to other
minor isomers (these peaks are denoted by * in Figure 2a),
indicating that the assembly is not 100% selective, but most of
the sample upon self-assembly exists as the single diastereo-
mer shown in Figure 2. It is challenging to determine the
structure of these minor isomers, but they display the same
Figure 3. 1H NMR spectra showing anion-templation studies of
[Fe4·26]8+ (CD3CN, 400 MHz, 298 K): a) [(ClO4)ꢀFe4·26](ClO4)7;
b) [Fex·2y](OTf)z; c) [(ClO4)ꢀFe4·26](OTf)7.
The triflate anion is evidently too weak a hydrogen-bond
acceptor to template the cage formation and no discrete
assembly is observed. When this undefined aggregate (here-
inafter referred to as [Fex·2y]·(OTf)z for simplicity) was
treated with Bu4NClO4 at room temperature, no change to
the nondiscrete aggregate was observed. This aggregate
displays similar solubility properties to [(ClO4)ꢀFe4·26]7+,
and has the characteristic purple color of an FeII-iminopyr-
idine complex, so is most likely a mixture of nondiscrete
oligomeric aggregates. Upon heating [Fex·2y]·(OTf)z in the
presence of Bu4NClO4 in CD3CN with ultrasonication for 3 h,
the undefined oligomeric aggregates reverted to the discrete
cage [(ClO4)ꢀFe4·26]7+ (Figure 3c).
9834 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9832 –9836

Angewandte
Chemie
À
If the self-assembly process was performed in a protic
solvent such as ethanol, rather than acetonitrile, no discrete
cage was formed. Ligand A was heated in EtOH with 2-
formylpyridine and Fe(ClO4)2, followed by removal of EtOH
in vacuo. Redissolution in CD3CN and subsequent NMR
analysis showed only broad, undefined peaks similar to those
of the anion exchange, the binding of BF4 could be
monitored by 19F NMR spectroscopy. The BF4-bound cage
[(BF4)ꢀFe4·26](OTf)7 was formed by heating the [Fex·2y]-
(OTf)z aggregate with one equivalent of Bu4NBF4. The anion-
exchange process was then evaluated by titrating excess
Bu4NBF4 into that solution. Fast anion exchange was
observed, as evidenced by a shift in the 19F resonance
À
of the [Fex·2y]·(OTf)z oligomers. This oligomer contains ClO4
À
ions, however, and upon heating in CD3CN, the discrete cage
could be reformed (see SI for spectra). The self-assembly
process, though slow, is reversible, and the addition of suitable
templating anions allows the formation of the cage assembly
from either starting materials or an oligomeric “unformed”
aggregate. If the hydrogen-bonding network is interrupted by
either a poorly hydrogen-bonding counterion, or a protic
assembly medium, the discrete cage is not formed, but the
“misfolding” can be easily corrected by conversion to
conditions favorable to hydrogen-bonding control.
toward that of free BF4
.
The association constant between the BF4À anion and the
[(BF4)ꢀFe4·26](OTf)7 species was determined to be 430 Æ
7mÀ1 based on a titration experiment tracked by 19F NMR
spectroscopy. The low association constant is likely due to
competition between the anion and the weakly hydrogen-
bonding acetonitrile solvent. When a similar experiment was
performed on the [(ClO4)ꢀFe4·26](ClO4)7 assembly using the
BF4À anion, an even lower association constant was observed
(15 Æ 2mÀ1). In addition to competing with acetonitrile, the
tetrafluoroborate is now competing with the perchlorate
anion, which can also bind inside the cage. The cage is
kinetically stable throughout the exchange: although templa-
tion is required for its formation, the template can dissociate
from the cage without disrupting the structure. This is
unsurprising: as the cage rests atop the templating anion,
complete encapsulation does not occur, and there is a low
barrier to anion exchange.
In conclusion, we have shown that the presence of
hydrogen-bonding groups, even those as weak as alcohols
can have profound implications on the stereochemistry of the
self-assembly in metal–organic cage complexes. Intramolec-
ular hydrogen bonding can template the formation of a non-
tetrahedral M4L6 self-assembled cage structure, and the
orientation of the hydrogen bonds sorts the assembly process,
favoring one diastereomer over multiple other possibilities,
while conferring selectivity for suitable anion templates. We
are currently exploiting these assemblies for controlled
biomimetic reactivity and the discovery of unprecedented
supramolecular architectures.
The scope of anions that are suitable templates for the
assembly was explored by treating the oligomeric aggregate
[Fex·2y]·(OTf)z with different organic-soluble salts as de-
scribed above for the perchlorate displacement. [Fex·2y]-
(OTf)z was heated with ultrasonication in acetonitrile with
various Bu4N+XÀ salts and the presence of discrete assemblies
was studied by 1H NMR analysis. As would be expected from
the crystal structure of [(ClO4)ꢀFe4·26]7+ only select anions
,
are suitable templates for assembly, determined by the size
and shape of the anion as well as H-bond-acceptor ability.
Oxoanions such as perchlorate, nitrate, and hydrogen sulfate
all lead to successful templation. These oxygen-containing
anions display the correct size and geometry to both fit inside
the cavity of [Fe4·26]8+ and provide hydrogen-bond-donor
groups that can interact with the internally oriented OH
À
groups of the ligand. Whereas the tetrahedral ClO4 shows
hydrogen bonding with at least four OH groups, anions with
fewer donors such as nitrate are tolerated as well. The lone-
pair donors need not be oxygen-based: treatment of [Fex·2y]-
À
(OTf)z with the BF4 anion resulted in templated cage
formation. Interestingly, although one could imagine triflate
providing suitable H-bond-acceptor sites for binding, it is not
an effective template, nor are longer oxoanions such as
S2O42À. The much larger Ph3SiF2À and PF6À anions are also too
large to fit inside of the cage, and no templation is observed.
All halide anions are ineffective templates as well, as they are
either too small to effectively engage in the requisite hydro-
gen bonding or cause decomposition of the cage upon heating.
Also, oxidizing anions were incompatible: periodate and
nitrite led to decomposition of the cages and precipitation of
an insoluble brown solid. The 1H NMR spectra of these other
[anionꢀFe4·26] assemblies all show slightly different chemical
shifts to those of [(ClO4)ꢀFe4·26]7+, as would be expected
from the nonsymmetric nature of the assembly and the close
interactions between ligand and template. The differences are
minor, however, and correspond to the same overall assembly
structure.
Received: May 13, 2014
Revised: June 12, 2014
Published online: July 17, 2014
Keywords: diastereoselectivity · hydrogen bonding ·
.
ligand effects · self-assembly · supramolecular chemistry
[1] a) T. R. Cook, Y.-R. Zheng, P. J. Stang, Chem. Rev. 2013, 113,
734; b) M. M. J. Smulders, I. A. Riddell, C. Browne, J. R.
Nitschke, Chem. Soc. Rev. 2013, 42, 1728.
[2] a) J. L. Bolliger, A. M. Belenguer, J. R. Nitschke, Angew. Chem.
Int. Ed. 2013, 52, 7958; Angew. Chem. 2013, 125, 8116; b) D.
Fujita, K. Suzuki, S. Sato, M. Yagi-Utsumi, Y. Yamaguchi, N.
Mizuno, T. Kumasaka, M. Takata, M. Noda, S. Uchiyama, K.
Kato, M. Fujita, Nat. Commun. 2012, 3, 1093; c) P. Mal, D.
Schultz, K. Beyeh, K. Rissanen, J. R. Nitschke, Angew. Chem.
Int. Ed. 2008, 47, 8297; Angew. Chem. 2008, 120, 8421; d) P. Mal,
B. Breiner, K. Rissanen, J. R. Nitschke, Science 2009, 324, 1697;
e) D. C. Caskey, T. Yamamoto, C. Addicott, R. K. Shoemaker, J.
Vacek, A. M. Hawkridge, D. C. Muddiman, G. S. Kottas, J.
Michl, P. J. Stang, J. Am. Chem. Soc. 2008, 130, 7620; f) S. Sato, J.
Although the presence of a single equivalent of anion
template is sufficient to promote the formation of the desired
cage, the anion is not irreversibly bound in the cavity.
Whereas the 1H NMR spectrum of both [(ClO4)ꢀFe4·26]7+
and [Fex·2y]·(OTf)z was too complex for an accurate analysis
Angew. Chem. Int. Ed. 2014, 53, 9832 –9836
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9835

.
Angewandte
Communications
Iida, K. Suzuki, M. Kawano, T. Ozeki, M. Fujita, Science 2006,
313, 1273.
[10] a) S. Yi, V. Brega, B. Captain, A. E. Kaifer, Chem. Commun.
2012, 48, 10295; b) R. Custelcean, P. V. Bonnesen, N. C. Duncan,
X. Zhang, L. A. Watson, G. Van Berkel, W. B. Parson, B. P. Hay,
J. Am. Chem. Soc. 2012, 134, 8525; c) C. Klein, C. Gꢂtz, M.
[3] a) M. Frank, J. Hey, I. Balcioglu, Y.-S. Chen, D. Stalke, T.
Suenobu, S. Fukuzumi, H. Frauendorf, G. H. Clever, Angew.
Chem. Int. Ed. 2013, 52, 10102; Angew. Chem. 2013, 125, 10288;
b) R. Chakrabarty, P. J. Stang, J. Am. Chem. Soc. 2012, 134,
14738; c) M. Wang, W.-J. Lan, Y.-R. Zheng, T. R. Cook, H. S.
White, P. J. Stang, J. Am. Chem. Soc. 2011, 133, 10752.
[4] K. Mahata, M. L. Saha, M. Schmittel, J. Am. Chem. Soc. 2011,
133, 10752.
[5] a) W. Meng, J. K. Clegg, J. D. Thoburn, J. R. Nitschke, J. Am.
Chem. Soc. 2011, 133, 13652; b) J. Xu, T. N. Parac, K. N.
Raymond, Angew. Chem. Int. Ed. 1999, 38, 2878; Angew.
Chem. 1999, 111, 3055; c) A. J. Metherell, W. Cullen, A.
Stephenson, C. A. Hunter, M. D. Ward, Dalton Trans. 2014, 43,
71.
´
Bogner, F. Topic, K. Rissanen, A. Lꢂtzen, Angew. Chem. Int. Ed.
2014, 53, 3739; Angew. Chem. 2014, 126, 3814.
[11] a) D. J. Cooke, J. M. Cross, R. V. Fennessy, L. P. Harding, C. R.
Rice, C. Slater, Chem. Commun. 2013, 49, 7785; b) A. M.
Johnson, R. J. Hooley, Inorg. Chem. 2011, 50, 4671.
[12] M. C. Young, A. M. Johnson, R. J. Hooley, Chem. Commun.
2014, 50, 1378.
[13] X. Zhang, J.-B. Han, P.-F. Li, X. Ji, Z. Zhang, Synth. Commun.
2009, 39, 3804.
[14] H.-L. Pan, T. L. Fletcher, Synthesis 1972, 192.
[15] K. Mahata, P. D. Frischmann, F. Wꢂrthner, J. Am. Chem. Soc.
2013, 135, 15656.
[6] a) M. C. Young, A. M. Johnson, A. S. Gamboa, R. J. Hooley,
Chem. Commun. 2013, 49, 1627; b) N. Ousaka, S. Grunder, A. M.
Castilla, A. C. Whalley, J. F. Stoddart, J. R. Nitschke, J. Am.
Chem. Soc. 2012, 134, 15528; c) T. N. Parac, D. L. Caulder, K. N.
Raymond, J. Am. Chem. Soc. 1998, 120, 8003; d) W. Meng, T. K.
Ronson, J. R. Nitschke, Proc. Natl. Acad. Sci. USA 2013, 110,
10531.
[7] T. Heinz, D. M. Rudkevich, J. Rebek, Jr., Nature 1998, 394, 764.
[8] L. J. Marshall, J. de Mendoza, Org. Lett. 2013, 15, 1548.
[9] M. C. Young, E. Liew, J. Ashby, K. E. McCoy, R. J. Hooley,
Chem. Commun. 2013, 49, 6331.
[16] a) I. A. Riddell, M. M. J. Smulders, J. K. Clegg, Y. R. Hristova,
B. Breiner, J. D. Thoburn, J. R. Nitschke, Nat. Chem. 2012, 4,
751; b) R. W. Saalfrank, R. Burak, A. Breit, D. Stalke, R. Herbst-
Irmer, J. Daub, M. Porsch, E. Bill, M. Mꢂther, A. X. Trautwein,
Angew. Chem. Int. Ed. Engl. 1994, 33, 1621; Angew. Chem. 1994,
106, 1697.
[17] a) S. Ma, M. M. J. Smulders, Y. R. Hristova, J. K. Clegg, T. K.
Ronson, S. Zarra, J. R. Nitschke, J. Am. Chem. Soc. 2013, 135,
5678; b) I. A. Riddell, Y. R. Hristova, J. K. Clegg, C. S. Wood, B.
Breiner, J. R. Nitschke, J. Am. Chem. Soc. 2013, 135, 2723.
9836 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 9832 –9836