A CuI-Based Metallo-Supramolecular Gel-Like Material Built from a Library of Oligomeric Ligands Featuring Exotopic 1,10-Phenanthroline Units

Article abstract of DOI:10.1002/ejoc.201501201

A 22-membered cyclic alkene C₁ incorporating an exotopic 1,10-phenanthroline nucleus in the ring skeleton has been synthesized and subjected to ring-opening metathesis polymerization (ROMP) in dilute dichloromethane solution at varying monomer

Full text of DOI:10.1002/ejoc.201501201

FULL PAPER
DOI: 10.1002/ejoc.201501201
A CuI-Based Metallo-Supramolecular Gel-Like Material Built from a Library
of Oligomeric Ligands Featuring Exotopic 1,10-Phenanthroline Units
José Augusto Berrocal,[a] Simone Albano,[b] Luigi Mandolini,[b] and Stefano Di Stefano*[b]
Keywords: Macrocyclic ligands / Gels / Metathesis / Dynamic covalent chemistry / Copper
A 22-membered cyclic alkene C1 incorporating an exotopic
1,10-phenanthroline nucleus in the ring skeleton has been
complexes taking advantage of the strong affinity of 1,10-
phenanthroline derivatives for CuI. The formation of a gel-
synthesized and subjected to ring-opening metathesis poly- like material only upon addition of [Cu(CH3CN)4]PF6 to the
merization (ROMP) in dilute dichloromethane solution at
varying monomer concentrations (cmon). The resultant librar-
ies of macrocyclic oligomers were used as ligands for the
generation of main-chain metal–ligand oligomeric/polymeric
most concentrated library (cmon = 60 mM) was explained as
arising from the presence of significant amounts of trimeric
and higher oligomeric macrocycles Ci (i Ն 3), acting as cross-
linking components.
These macrocycles, as well as the vast multitude of
macrocyclic ligands that have found countless applications
Introduction
Since the first templated synthesis of catenanes reported in supramolecular chemistry over many decades, possess en-
by Dietrich-Buchecker and Sauvage et al.,[1] copper(I) com- dotopic donor sites and, consequently, undergo endocyclic
plexes of 1,10-phenanthroline derivatives have been exten- coordination with metal ions. In contrast, macrocyclic li-
sively used to assemble catenane and rotaxane structures for gands featuring exotopic binding sites are rare.[6] Recently,
use as components in a variety of discrete[2] and polymeric Lee and co-workers[7] showed that exo coordination of the
systems.[3]
sulfur donors of thiamacrocycles allows the construction of
In the course of our studies of dynamic covalent chemis- metallo-supermolecules and -supramolecular polymers not
try,[4] we recently reported that ring-opening metathesis accessible by endo-coordination strategies.
polymerization (ROMP) of the macrocyclic alkene 1 in the
As a contribution to the largely unexplored area of exo-
presence of second-generation Grubbs’ catalyst G2 gener- cyclic coordination, we focused on the macrocyclic alkene
ates a dynamic library (DL) of cyclic oligomers under fully C1 as a potential precursor of multitopic macrocyclic li-
reversible conditions (Figure 1).[5]
gands endowed with exotopic coordination sites. In this
Figure 1. Dynamic library of macrocyclic oligomers generated during ROMP of 1 under dilute conditions.
[a] Institute for Complex Molecular Systems, Eindhoven
work we report on the synthesis of C1, its subsequent sub-
University of Technology,
P. O. Box 513, 5600 MB Eindhoven, The Netherlands
[b] Dipartimento di Chimica, Sapienza Università di Roma and
Istituto CNR di Metodologie Chimiche (IMC-CNR), Sezione
Meccanismi di Reazione,
jection to ROMP in the presence of G2, and the effect of
the addition of CuI to the libraries thereby obtained as a
novel strategy for building a metallo-supramolecular gel-
like material. The present procedure differs significantly
from known procedures for synthesizing organometallic
and coordination polymers[8] based on ring-opening poly-
merization reactions of metal-containing cyclic monomers,
such as ferrocenophanes[9] and CuI–catenanes.[10] In all
P.le A. Moro 5, 00185 Rome, Italy
E-mail: stefano.distefano@uniroma1.it
http://www.chem.uniroma1.it/dipartimento/persone/
stefano-di-stefano
Supporting information and ORCID(s) from the author(s) for
this article are available on the WWW under http://dx.doi.org/
10.1002/ejoc.201501201.
7504
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7504–7510

A CuI-Based Metallo-Supramolecular Gel-Like Material
to ROMP (CD2Cl2, 30 °C) in the presence of G2. The initial
concentration of monomer C1 (cmon) was varied from 10 to
100 mm. Periodic monitoring of the reaction mixtures by 1H
NMR spectroscopy showed that equilibrium was reached in
all cases after 24 h, in accordance with results obtained in
the ROMP of 1 under the same conditions.[5] The expected
formation of mixtures of cyclic oligomers (Figure 2) was
confirmed by mass spectrometry (Figure 3).[16,17]
these cases the bonds involving the metal ion do not un-
dergo any change during polymerization.
Scheme 1.
Results and Discussion
Design, Synthesis of C1, and Equilibration Experiments
The choice to combine exocyclic coordination with Ru-
catalyzed olefin metathesis demanded stringent require-
ments in the molecular design of C1, namely the presence
of the two methyl groups at the 2- and 9-positions of the
1,10-phenanthroline moiety. Indeed, 1,10-phenanthroline
derivatives are typically employed as Ru ligands for the gen-
eration of bioactive complexes[11] or antenna systems,[12] to
name but a few. We hypothesized that the steric hindrance
of the two methyl groups minimizes the displacement of the
ligands bound to the Ru center of G1 and G2, therefore
Figure 2. Dynamic library of macrocyclic oligomers generated dur-
ing ROMP of C1.
1
ensuring the activity of both catalysts. Preliminary H and
31P NMR experiments in CD2Cl2 confirmed this hypothe-
sis. Clear spectral variations and color change (from purple
to dark brown) were observed upon treatment of a solution
of G2 with 1,10-phenanthroline, whereas neocuproine (2,9-
dimethyl-1,10-phenanthroline) acted as a co-existing, but
non-interacting species in solution (see Figures S1–S3 in the
Supporting Information).
Macrocyclic alkene C1 was prepared by a two-step pro-
cedure (Scheme 1) starting from 4,7-dihydroxyneocuproine
(2), which was synthesized according to a literature pro-
cedure[13] (see Scheme S1 in the Supporting Information).
Figure 3. ROMP of C1 in the presence of G2. Typical ESI-TOF
mass spectrum of a reaction mixture at equilibrium (m/z = 680 has
been attributed to the doubly charged species C3 + H+ + Cu+).
The DLs obtained in the above equilibration experiments
The reaction of compound 2 with 8-bromooctene in THF/ can be described as macrocyclization equilibria ruled by the
DMF (1:1) at room temperature in the presence of NaH Jacobson–Stockmayer (J-S) theory.[18] Theory predicts that
gave the doubly alkylated compound 3. A 10 mm dichloro- the concentration of each oligomer Ci increases upon in-
methane solution of 3 was then subjected to ring-closing creasing the total monomer concentration cmon until a criti-
metathesis by addition of first-generation Grubbs’ catalyst cal value cmon* is reached. Below cmon* the concentration
G1. The less robust G1 was preferred to G2[14] to minimize of linear polymers Pi is negligibly small and the DL is com-
the incursion of cross-metatheses involving C1.[15] Column posed of cyclic oligomers only [Equation (1), which holds
chromatography of the crude reaction product afforded when cmon Յ cmon*]. Above cmon* the concentration of lin-
macrocycle C1 as a mixture of cis and trans isomers, clearly ear species Pi is given by Equation (2), whereas the concen-
detected in the 1H NMR spectrum (see Figure S6). The un- tration of cyclic species Ci remains constant [Equation (3)].
resolved mixture of geometrical isomers of C1 was subjected The limiting value [Ci]* approached by [Ci] when cmon ap-
Eur. J. Org. Chem. 2015, 7504–7510
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7505

J. A. Berrocal, S. Albano, L. Mandolini, S. Di Stefano
FULL PAPER
proaches cmon* is a measure of the thermodynamic effective equilibria for which cmon* values between 0.13–0.20 m have
molarity EMi of Ci.[19]
been reported.[22] The limiting value of 10 mm approached
by the equilibrium concentration of C1 compares well with
the EM of 22 mm reported for 1 and is consistent with what
is expected on the basis of known EM values of large, low-
strain rings of comparable size.[23]
i[Ci] = cmon
(1)
(2)
(3)
͚
i
i[Pi] = cmon – c*mon
͚
i
i[Ci]* = c*mon
͚
Copper(I)-Induced Gelation Experiments
i
In view of the ability of the CuI ion to form very strong
1:2 complexes with neocuproine,[24] it was of interest to
study the effect of the addition of this metal ion to the li-
braries obtained during the ROMP of C1. Linear species
are virtually absent and only mono-, di-, and multitopic
macrocyclic ligands are available as building blocks for the
construction of main-chain metal–organic oligomeric/poly-
meric assemblies. Although dimeric rings C2 can function
as binding units in the formation of linear assemblies in
which monomeric rings C1 act as chain termini (Figure 6,
a), the multitopic nature of trimers C3 and higher cyclic
oligomers enables them to act as cross-linking agents (Fig-
ure 6, b).
1
Figure 4 shows typical H NMR spectra (aromatic re-
gion) of equilibrated solutions obtained from ROMP of C1
in the presence of 3mol-% G2 in CD2Cl2 at varying cmon
values. The singlets of the geometrical isomers of C1 appear
as well-resolved signals that were easily integrated to allow
the concentrations of C1 to be plotted as a function of cmon
(Figure 5). In agreement with J-S theory, the concentration
profile reaches a plateau value in the high-concentration re-
gion.[20,21] Although the value of cmon* cannot be precisely
determined on the basis of available data, the shape of the
concentration profile suggests that it should lie somewhere
in the range of 0.12–0.16 m. This estimation is in line with
literature data related to a large variety of macrocyclization
Figure 4. Aromatic portion of the 1H NMR spectra of three equili-
brated solutions at different cmon (CD2Cl2, 30 °C). The two singlets
at lower field are due to the aromatic protons Ha of the two cis
and trans isomers.
Figure 6. Schematic representation of a) linear and b) cross-linked
metallo-supramolecular aggregates formed upon addition of CuI to
a library obtained from ROMP of C1.
A stoichiometric amount of [Cu(CH3CN)4]PF6 was in-
troduced into the equilibrated solutions 4 d after the start
of ROMP. This ensured that CuI complexation with phen-
anthroline units occurred when the catalyst had totally de-
composed[25] and, consequently, oligomer distributions
were exactly those dictated by J-S theory. For the same
reasons, quenching of ROMP reactions with ethyl vinyl
ether[3b] was avoided because the introduction of end-
groups into a dynamic system would likely cause variations
in the library composition. The results of the addition of
CuI to the no longer dynamic, but static libraries of oligo-
meric macrocyclic ligands are illustrated in Figure 7. The
complexation of CuI was revealed by the immediate appear-
ance of a brown color. The more dilute samples (cmon = 20
and 30 mm) gave free-flowing homogeneous solutions, but
for cmon = 60 mm, the appearance of the brown color was
Figure 5. Concentration of total C1 (circles) in equilibrated reaction
mixtures as a function of cmon. Triangles and squares represent the
contributions of the individual geometrical isomers (3mol-% G2,
CD2Cl2, 30 °C).
7506
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7504–7510

A CuI-Based Metallo-Supramolecular Gel-Like Material
followed by the formation of a gel-like material after about Further Gelation Studies
3 h. As a consequence of the high stability of 2:1 complexes
Other salts of metal cations capable of forming com-
of neocuproine derivatives and CuI,[24] we highlight the ex-
cellent thermal stability of the gel-like material. Even when
heated at 65 °C[26] for prolonged periods (10–15 min), the
material never turned into a solution. Such behavior is in
stark contrast to typical supramolecular gels, in which a
temperature increase causes disaggregation and the subse-
quent transformation of gels into solutions.[27] As expected,
treatment of the gel-like material with Et4NCN (ten-fold
excess) restored a free-flowing solution due to the seque-
stration of CuI (see Figure S8 in the Supporting Infor-
mation).
plexes with 1,10-phenanthroline,[24,28] namely AgI, CoII,
ZnII, and CuII, were also tested in gelation experiments in
CH2Cl2 (Table 1). Stoichiometric amounts (0.5 mol-equiv.)
of metal salts were added to cmon = 60 mm static libraries,
and the mixtures were kept in darkness and without stir-
ring. With the CuII salt as the only exception, all the salts
are scarcely soluble in CH2Cl2 and immediately after ad-
dition remained undissolved at the bottom of the vial.
Eventually (1 d) an appreciable reduction of the amount of
undissolved salt (CoII, ZnII, and AgI) was observed, most
likely due to complexation. None of these metal cations
caused the gelation of solutions: all of them remained free-
flowing (see Figure S10 in the Supporting Information).
Only CuII led to a precipitate in the dark-brown solution,
but no gel was formed (see Figure S10). In addition to the
metal cation, the counter anion was also varied in this series
of experiments. In a previous report[10] it was shown that
–
–
the substitution of PF6 by CF3SO3 did not alter the out-
come of the gelation. Here, despite the pivotal role played
by cmon in determining the product distribution, and hence
Figure 7. Appearance of static libraries obtained by ROMP of C1
in CH2Cl2 at the given monomer concentrations after the addition cross-linking, an influence of the counter anion on the out-
of a stoichiometric amount of [Cu(CH3CN)4]PF6.
come of the gelation cannot be excluded.
As predicted by J-S theory and actually found in real
Table 1. Gelation studies, and related outcomes, performed at cmon
systems, the equilibrium concentration of the Ci oligomers = 60 mm in CH2Cl2 with different metal cations.
declines progressively with increasing i. This happens not
Metal cation
Counter anion
Gelation
only because the limiting concentration [Ci]* = EMi is pro-
portional to the –5/2 power of i,[18] but also because the
rate at which each oligomer Ci approaches its limiting con-
AgI
CF3CO2–
no
no
no
–
CoII
ZnII
CH3CO2–
CH3CO2
–
centration when cmon approaches cmon* varies inversely with CuII
the degree of polymerization i [Equation (4)].[18]
CF3SO3
no (precipitation)
The same strategy of adding CuI to static libraries ob-
tained at cmon = 60 mm was also used to investigate the
[Ci] = ([C1]/EM1)iEMi
(4)
As a consequence, the concentration of higher oligomers influence of the loading of this metal cation on the genera-
is very low in dilute solutions, that is, when cmon ϽϽ cmon
*
tion of the gel-like material. CuI loadings of 0.25 to 5 mol-
and [C1] ϽϽ EM1, but increases rapidly with increasing equiv. were tested (Table 2) under identical experimental
cmon. In line with the above arguments, it seems likely that conditions. Gelation was not observed, except for a partial
the concentration of oligomers Ci with i Ն 3 in the library gelation observed with a loading of 1 mol-equiv. CuI. In
generated at cmon = 30 mm is still too low and cross-links this case roughly 40% of the solution was gelated, with the
are not formed in a significant amount upon addition of rest free-flowing (see Figure S11 in the Supporting Infor-
CuI. However, when cmon is increased to 60 mm, the concen- mation). Loadings of CuI equal to 0.25 and 0.75 gave films
tration of oligomers with i Ն 3 is believed to be high on the walls of the vials as minor products, together with a
enough and, consequently, the addition of CuI results in the predominant colored free-flowing solution. A large excess
formation of a metallo-supramolecular gel-like network. It of CuI (5 equiv.) provoked precipitation of dark aggregates
is apparent that the exotopic orientation of the phen- surrounded by an even darker solution (see Figure S11).
anthroline moiety plays a crucial role in the formation of These results clearly highlight the dramatic influence of the
the gel-like material. When the static library obtained from cmon/[CuI] ratio on the outcome of the gelation process.
ROMP of macrocyclic monomer 1 under identical condi- Moreover, it is evident that a 2:1 stoichiometry is highly
tions (cmon = 60 mm, CH2Cl2, 30 °C, 3 mol-% G2) was desirable for the obtainment of a homogeneous gel-like ma-
treated with 30 mm [Cu(CH3CN)4]PF6, a red, free-flowing terial.
solution was obtained (see Figure S9 in the Supporting In-
CuI-induced gelation was also attempted in different or-
formation). It appears that in this case the conformational ganic solvents screened on the basis of polarity and boiling-
flexibility of the larger macrocycles is not large enough for point criteria. However, the selective solubility of the static
the phenanthroline moieties to be oriented outwards for an libraries or [Cu(CH3CN)4]PF6 in most of the solvents
effective cross-linking to take place.
studied forced the choice towards halogenated solvents,
Eur. J. Org. Chem. 2015, 7504–7510
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7507

J. A. Berrocal, S. Albano, L. Mandolini, S. Di Stefano
FULL PAPER
Table 2. Gelation studies, and related outcomes, performed at cmon
= 60 mm in CH2Cl2 with different loadings of CuI.
Conclusions
In this paper we have introduced a simple procedure to
build a metallo-supramolecular gel-like material. The key
polymerization reaction was the coordination between CuI
and 1,10-phenanthroline units. A novel feature of the pro-
cedure is the use of a library composed of a virtually infinite
number of oligomeric ligands instead of a single ligand.[29]
Another distinctive feature is the presence of an exotopic
1,10-phenanthroline moiety inserted into the macrocyclic
frame of the monomeric olefin C1, ROMP of which in di-
lute dichloromethane solutions afforded libraries solely
composed of macrocyclic oligomers. Although the libraries
were obtained under dynamic conditions, the subsequent
polymerization reactions produced static libraries because
treatment with [Cu(CH3CN)4]PF6 was carried out after
complete inactivation of the Ru catalyst.[30] A gel-like mate-
rial was obtained from the most concentrated library (cmon
= 60 mm) upon addition of CuI. The tris-ligand macrocycle
C3 and higher oligomers are believed to play a key role as
cross-linking agents in the gel-forming process.
CuI loading [mol-equiv.]
Gelation
0.25
0.75
1
no
no
roughly 40%
no
5
namely chloroform, 1,2-dichloroethane, and tetrachloro-
ethane (Table 3). However, halogenated aromatic solvents,
for example, o-dichlorobenzene, could also not be used be-
cause of the insolubility of [Cu(CH3CN)4]PF6 and the static
libraries.
Table 3. Gelation studies, and related outcomes, performed at cmon
= 60 mm in different aliphatic halogenated solvents with 0.5 mol-
equiv. of CuI.
Solvent
Solubility of
static library
Solubility of
[Cu(CH3CN)4]PF6
Gelation
The presence and loading of CuI appears to be crucial for
the obtainment of the gel-like material. Indeed, the static
libraries are a powdery mixture of organic molecules in the
absence of CuI, and only in the presence of exactly 0.5 mol-
equiv. of CuI was a gel-like material (in the wet state) or a
film (in the dry state) formed.
CHCl3
ClCH2CH2Cl
Cl2CHCHCl2
poor
soluble
poor
poor
poor
poor
no
partial
no
However, in contrast to dichloromethane, none of the
above three aliphatic halogenated solvents gave gels. Only a
thin film on the walls of the vial surrounded by a colored
solution was observed in CHCl3 (see Figure S12 in the Sup-
porting Information) and 1,1,2,2-tetrachloroethane gave a
massive precipitate surrounded by an almost colorless solu-
tion (see Figure S12). Minor signs of gelation were ob-
served, however, in 1,2-dichloroethane (see Figure S12). On
performing a qualitative upside-down vial test, a small
amount of material did not flow, but stayed attached to the
bottom, but the amount of gelated solution was so small
that 1,2-dichloroethane could not be used for large-scale gel
preparation.
Experimental Section
Instruments and General Methods: NMR spectra were recorded
with 300 and 400 MHz spectrometers at room temperature unless
otherwise stated and were internally referenced to the residual pro-
ton solvent signal. Mass spectra were recorded with an ESI-TOF
mass spectrometer. UV/Vis spectra were recorded with a single-ray
spectrophotometer using a standard quartz cell (path length: 1 cm)
at room temperature. Scanning electron microscopy was performed
with an FEI Quanta 600F instrument under high vacuum (electron
beam acceleration voltage of 1 kV) without any further treatment
of the sample.
Because gelation only occurs with reliable and scalable
quantities in dichloromethane, a very low-boiling point sol-
vent, rheological studies could not be carried out: Solvent
would evaporate and measurements would be performed on
a dry material. On the other hand, performing plate–plate
rheology directly on a pre-dried material could represent an
alternative and valid approach. Thus, gels were air-dried
and dark-colored films were obtained. These dry films were
extremely brittle and any small mechanical deformation
manually exerted caused partial fractures or breaks. Such
behavior may be in contrast to the previously reported high
thermal stability of the gel. However, it is in line with our
previous report,[10] in which CuI complexation resulted in a
higher thermal stability of the material, but also on an en-
hanced rigidity due to restriction of conformational mo-
tions. Morphological studies performed by SEM (see p. S17
in the Supporting Information) offered an explanation to
this brittleness and showed that the dry, metalated material
was not resistant enough to undergo plate–plate rheological
studies.
Materials: All reagents and solvents were purchased at the highest
commercial quality and were used without further purification un-
less stated otherwise. The glassware was either flame- or oven-
dried. 2,9-Dimethyl-4,7-dihydroxy-1,10-phenanthroline (2) was
prepared according to a literature procedure[13] (see Scheme S1 in
the Supporting Information). 2,9-Dimethyl-4,7-bis(oct-7-enyloxy)-
1,10-phenanthroline (3) was synthesized as shown below, adapting
a literature procedure.[31] The NaH dispersion in mineral oil was
washed with hexane immediately prior to use. THF was distilled
from sodium and benzophenone. The dichloromethane used in the
olefin metatheses reactions was filtered through basic alumina just
before use. Chloroform, 1,2-dichloroethane, and tetrachloroethane
used in solvent gelation studies were flashed through basic alumina
prior to use.
2,9-Dimethyl-4,7-bis(oct-7-en-1-yloxy)-1,10-phenanthroline (3): 2,9-
Dimethyl-4,7-dihydroxy-1,10-phenanthroline (2; 2.16 g, 9 mmol)
was added to a suspension of NaH (1.44 g, 60% in mineral oil,
36 mmol) in THF/DMF (1:1, 104 mL) at 0 °C and the mixture was
stirred for 30 min under an inert atmosphere. Then 8-bromo-1-oc-
tene (6 mL, 36 mmol) and NaI (540 mg, 3.6 mmol) were added and
7508
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7504–7510

A CuI-Based Metallo-Supramolecular Gel-Like Material
the mixture was stirred overnight at room temperature. The reac-
tion was then quenched with water and the resulting mixture ex-
tracted three times with CH2Cl2. The organic phases were col-
lected, washed with brine, dried with Na2SO4, and filtered. The
solvent was evaporated and the crude obtained was subjected to
column chromatography (basic Al2O3, ethyl acetate/hexane, 8:2) to
second-generation Grubbs’ catalyst G2 (3 mol-%) was added. Reac-
tion mixtures were left without stirring at room temperature for
4 d.
Solutions were dried under a flow of N2 and a weighed amount of
the given salt was introduced into the vial, followed by a calculated
amount of the desired solvent to reach cmon = 60 mm. Suspensions
were sonicated and left without stirring.
1
afford pure 3 (1.2 g, 2.61 mmol, 29% yield), m.p. 137–140 °C. H
NMR (300 MHz, CDCl3): δ = 1.48–1.59 (m, 8 H), 1.87–1.96 (m, 4
H), 2.02–2.07 (m, 4 H), 2.84 (s, 6 H), 4.16 [t, 3J(H,H) = 6 Hz, 4
H], 4.96 (m, 4 H), 5.80 (m, 2 H), 6.81 (s, 2 H), 8.06 (s, 2 H) ppm.
13C NMR (75 MHz, CDCl3): δ = 25.86, 26.22, 28.67, 28.69, 28.81,
33.56, 66.35, 103.29, 114.26, 117.87, 119.38, 138.18, 145.41, 160.01,
161.72 ppm. HRMS (ESI-TOF): calcd. for C30H41N2O2 [M + H]+
461.3168; found 461.3179. UV/Vis (CH3OH): λmax (ε) = 256
(36210), 297 (9590), 308 (9390), 325 (2070), 341 nm
(1250 dm3 mol–1 cm–1).
General Procedure for CuI Loading Gelation Experiments: Cyclic
monomer C1 or 1 was weighed in a screw-cap vial and the appropri-
ate amount of CH2Cl2 (200–250 μL), dried on basic alumina just
before use, was added to prepare solutions of the desired concentra-
tion and then second-generation Grubbs’ catalyst G2 (3 mol-%)
was added. The reactions were run at 30 °C for 24 h. After cooling
to room temperature, the mixture was held in the vial for an ad-
ditional 72 h and a weighed amount of [Cu(CH3CN)4]PF6 (de-
pending on the number of mol-equiv. with respect to initial C1) was
added.
Cyclic Monomer C1: Compound 3 (1.06 g, 2.31 mmol) was dis-
solved in CH2Cl2 (230 mL) and the solution was degassed by
freeze–pump–thaw cycles and first-generation Grubbs’ catalyst G1
(285 mg, 0.35 mmol) was added. The resulting mixture was stirred
at 30 °C for 3 d under an inert atmosphere, monitoring the reaction
by ESI-TOF MS. Then the solution was filtered through a short
path of basic alumina, the solvent was evaporated, and the crude
material was subjected to column chromatography (basic Al2O3,
ethyl acetate/hexane, 6:4) to give C1 as a mixture of cis and trans
isomers (150 mg, 0.35 mmol, 15% yield), m.p. 193–197 °C. 1H
NMR (300 MHz, CD2Cl2): δ = 1.45–1.67 (m, 16 H), 1.93–2.11 (m,
General Procedure for the Gelation Experiments with Other Sol-
vents: Cyclic monomer C1 was weighed in a vial and the appropri-
ate amount of CH2Cl2 (500–600 μL), dried on basic alumina just
before use, was added to prepare solutions at cmon = 60 mm. Then
second-generation Grubbs’ catalyst G2 (3 mol-%) was added. Reac-
tion mixtures were left without stirring at room temperature for
4 d.
Solutions were dried under a flow of N2 and a weighed amount of
the given salt was introduced into the vial, followed by a calculated
amount of the desired solvent to reach cmon = 60 mm. Suspensions
were sonicated and left without stirring.
3
16 H), 2.80 (s, 12 H), 4.31 (m, 8 H), 5.32 [t, J(H,H) = 6 Hz, 2 H],
5.53 (m, 2 H), 6.88 (s, 2 H), 6.89 (s, 2 H), 8.13 (s, 2 H), 8.16 (s, 2
H) ppm. 13C NMR (75 MHz, CD2Cl2): δ = 25.71, 27.14, 27.35,
27.42, 27.59, 27.72, 28.83, 29.34, 29.39, 29.57, 32.66, 65.46, 69.53,
103.30, 103.65, 119.06, 119.12, 119.66, 119.81, 129.79, 130.34,
145.85, 145.92, 159.50, 161.64, 161.67 ppm. HRMS (ESI-TOF):
calcd. for C28H37N2O2 [M + H]+ 433.2834; found 433.2831. UV/
Vis (CH3OH): λmax (ε) = 256 (27530), 298 (6960), 308 (6830), 325
(1470), 341 nm (975 dm3 mol–1 cm–1).
Scanning Electron Microscopy: A solution collected at the bottom
of a gel-containing upside-down vial was drop-cast on to a glass
cover slip (12 mm diameter) and air-dried. The sample was imaged
without any further treatment at an electron beam acceleration
voltage of 1 kV.
General Procedure for Ring-Opening Metathesis: Cyclic monomer
C1 was weighed in a NMR tube and the appropriate amount of
CD2Cl2 (500–600 μL), dried on basic alumina just before use, was
added to prepare solutions of the desired concentration. Then sec-
ond-generation Grubbs’ catalyst G2 (3 mol-%) was added. The re-
actions were run at 30 °C and monitored by 1H NMR spectroscopy.
Acknowledgments
Thanks are due to the Ministero dell’Istruzione, dell’Università e
della Ricerca (MIUR) (PRIN 2010CX2TLM) and Sapienza Uni-
versità di Roma through the Ricerca Scientifica Anno 2014 pro-
gram. The authors thank Chiara Biagini and Rachele Caruso for
technical assistance in the synthetic work. Matilde Putti is acknowl-
edged for SEM imaging.
General Procedure for the Gelation Experiments: Cyclic monomer
C1 or 1 was weighed in a screw-cap vial and the appropriate
amount of CH2Cl2 (200–250 μL), dried on basic alumina just be-
fore use, was added to prepare solutions of the desired concentra-
tion and then second-generation Grubbs’ catalyst G2 (3 mol-%)
was added. The reactions were run at 30 °C for 24 h. After cooling
to room temperature, the mixture was held in the vial for additional
72 h and a weighed amount of [Cu(CH3CN)4]PF6 (0.5 equiv. with
respect to initial C1) was added. Gelation was observed after 3 h in
the case of cmon = 60 mm.
[1] C. O. Dietrich-Buchecker, J.-P. Sauvage, J.-P. Kintzinger, Tetra-
hedron Lett. 1983, 24, 5095–5098.
[2] a) J. F. Stoddart, Chem. Soc. Rev. 2009, 38, 1802–1820; b) J. E.
Beves, B. A. Blight, C. J. Cambell, D. A. Leigh, R. T. McBur-
ney, Angew. Chem. Int. Ed. 2011, 50, 9260–9327; c) J.-C. Cham-
bron, J.-P. Sauvage, New J. Chem. 2013, 37, 49–57.
[3] a) B. J. Holliday, T. M. Swager, Chem. Commun. 2005, 23–36;
b) S. M. Kang, M. Cetin, R. Jiang, E. S. Clevenger, M. F.
Mayer, J. Am. Chem. Soc. 2014, 136, 12588–12591, and refer-
ences cited therein.
[4] a) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor,
J. K. M. Sanders, S. Otto, Chem. Rev. 2006, 106, 3652–3711; b)
J. Reek, S. Otto (Eds.), Dynamic combinatorial chemistry,
Wiley-VCH, Weinheim, Germany, 2010; c) B. L. Miller (Ed.),
Dynamic combinatorial chemistry in drug discovery, bioorganic
chemistry and materials science, Wiley, Hoboken, NJ, 2009; d)
S. Ladame, Org. Biomol. Chem. 2008, 6, 219–226; e) Y. Jin, C.
Yu, R. J. Denman, W. Zhang, Chem. Soc. Rev. 2013, 42, 6634–
6654; f) J. Li, P. Nowak, S. Otto, J. Am. Chem. Soc. 2013, 135,
General Procedure for the Demetalation Experiments: A ten-fold
excess tetraethylammonium cyanide (Et4NCN) solution in CH2Cl2
(not degassed) was added to the CuI-complexed gelated material.
Two phases were clearly observed immediately after mixing. Upon
gentle shaking, the mixture became homogeneous, giving a clear
indication of CuI sequestration by CN–.
General Procedure for Gelation Experiments with Other Cations:
Cyclic monomer C1 was weighed in a vial and the appropriate
amount of CH2Cl2 (500–600 μL), dried on basic alumina just be-
fore use, was added to prepare solutions at cmon = 60 mm. Then,
Eur. J. Org. Chem. 2015, 7504–7510
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
7509

J. A. Berrocal, S. Albano, L. Mandolini, S. Di Stefano
FULL PAPER
9222–9239; g) A. Herrmann, Chem. Soc. Rev. 2014, 43, 1899–
1933; h) S. Ulrich, P. Dumy, Chem. Commun. 2014, 50, 5810–
5825.
respectively, possibly arising from trace amounts of CuI in the
methanol used in the preparation of the sample.
[18] a) H. Jacobson, W. H. Stockmayer, J. Chem. Phys. 1950, 18,
1600–1606; b) G. Ercolani, L. Mandolini, P. Mencarelli, S.
Roelens, J. Am. Chem. Soc. 1993, 115, 3901–3908; c) S. Di Ste-
fano, J. Phys. Org. Chem. 2010, 23, 797–805.
[5]
[6]
J. A. Berrocal, M. M. L. Nieuwenhuizen, L. Mandolini, E. W.
Meijer, S. Di Stefano, Org. Biomol. Chem. 2014, 12, 6167–6174.
A very ingenious use of ZnII coordination with a combination
of an exo-bidentate bipyridyl ligand and an endo-tridentate di-
imminopyridyl ligand resulted in precise control of the six
crossover points in a successful self-assembly of molecular bor-
romean rings from 18 components, see: K. S. Chichak, S. J.
Cantrill, A. R. Pease, S.-H. Chiu, G. W. V. Cave, J. L. Atwood,
J. F. Stoddart, Science 2004, 304, 1308–1312.
[19] a) L. Mandolini, Adv. Phys. Org. Chem. 1986, 22, 1–111; b) M.
Ciaccia, I. Tosi, L. Baldini, R. Cacciapaglia, L. Mandolini, S.
Di Stefano, C. A. Hunter, Chem. Sci. 2015, 6, 144–151.
[20] Data points could be fitted with good precision to a single
exponential function, as found in previous works (see ref.[21]).
We can offer no theoretical explanation for such behavior.
[21] a) J. A. Berrocal, R. Cacciapaglia, S. Di Stefano, Org. Biomol.
Chem. 2011, 9, 8190–8194; b) J. A. Berrocal, R. Cacciapaglia,
S. Di Stefano, L. Mandolini, New J. Chem. 2012, 36, 40–43.
[22] a) K. Ito, Y. Hashizuka, Y. Yamashita, Macromolecules 1977,
10, 821–824; b) Y. Yamashita, J. Mayumi, Y. Kawakami, K.
Ito, Macromolecules 1980, 13, 1075–1080; c) A. T. ten Cate, H.
Kooijman, A. L. Spek, R. P. Sijbesma, E. W. Meijer, J. Am.
Chem. Soc. 2004, 126, 3801–3808; d) R. Cacciapaglia, S.
Di Stefano, L. Mandolini, J. Am. Chem. Soc. 2005, 127, 13666–
13671; e) R. Cacciapaglia, S. Di Stefano, L. Mandolini, P.
Mencarelli, F. Ugozzoli, Eur. J. Org. Chem. 2008, 186–195; f)
R. Cacciapaglia, S. Di Stefano, G. Ercolani, L. Mandolini,
Macromolecules 2009, 42, 4077–4083.
[7]
a) S. Park, S. Y. Lee, K.-M. Park, S. S. Lee, Acc. Chem. Res.
2012, 45, 391–403; b) L. F. Lindoy, K.-M. Park, S. S. Lee,
Chem. Soc. Rev. 2013, 42, 1713–1727; c) S. Kim, L. F. Lindoy,
S. S. Lee, Coord. Chem. Rev. 2014, 280, 176–202.
[8]
[9]
K. A. Williams, A. J. Boydston, C. W. Bielawski, Chem. Soc.
Rev. 2007, 36, 729–744.
a) P. F. Brandt, T. B. Rauchfuss, J. Am. Chem. Soc. 1992, 114,
1926–1927; b) C. E. Stanton, T. R. Lee, R. H. Grubbs, N. S.
Lewis, J. K. Pudelski, M. R. Callstrom, M. S. Erickson, M. L.
McLaughlin, Macromolecules 1995, 28, 8713–8721; c) M. A.
Buretea, T. D. Tilley, Organometallics 1997, 16, 1507–1510; d)
P. Nguyen, P. Gomez-Elipe, I. Manners, Chem. Rev. 1999, 99,
1515–1548.
[23] A reasonable estimate of the EM of a strainless ring of compar-
able size is in the order of 0.050 m (see ref.[19a]). This implies
that the strain energy of C1 should not exceed 1 kcal/mol.
[24] H. Irving, D. H. Mellor, J. Chem. Soc. 1962, 5237–5245.
[10]
J. A. Berrocal, L. M. Pitet, M. M. L. Nieuwenhuizen, L. Man-
dolini, E. W. Meijer, S. Di Stefano, Macromolecules 2015, 48,
1358–1363.
[11] L. Xu, N.-J. Zhong, Y.-Y. Xie, H.-L. Huang, G.-B. Jiang, Y.-J. [25] In a control experiment, the ring-opening metathesis of 30 mm
Liu, PloS ONE 2014, 9, e96082.
[12] G. Accorsi, A. Listorti, K. Yoosaf, N. Armaroli, Chem. Soc.
Rev. 2009, 38, 1690–1700.
C1 initiated by the addition of G2 afforded an equilibrated li-
brary Ci after 24 h. After 72 h, 30 mm 3 was added to the mix-
ture and no transformation of the latter was observed, thereby
confirming the inactivation of the catalyst.
[13] R. A. Altman, S. L. Buchwald, Org. Lett. 2006, 8, 2779–2782.
[14] The half-lives of methylidene intermediates derived from G1
and G2 are 40 and 340 min, respectively, in benzene at 55 °C,
see: S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day,
R. H. Grubbs, J. Am. Chem. Soc. 2007, 129, 7961–7968; for
seminal G1-based syntheses of macrocycles containing the
phenanthroline moiety, see also: a) M. Weck, B. Mohr, J.-P.
Sauvage, R. H. Grubbs, Angew. Chem. Int. Ed. Engl. 1997, 36,
[26] Heating of the sample was performed in a closed-cap vial to
avoid solvent evaporation.
[27] G. Yu, X. Yan, C. Han, F. Huang, Chem. Soc. Rev. 2013, 42,
6697–6722.
[28] H. Irving, D. H. Mellor, J. Chem. Soc. 1962, 5222–5237.
[29] J. B. Beck, S. Rowan, J. Am. Chem. Soc. 2003, 125, 13922–
13923.
1308–1310; Angew. Chem. 1997, 109, 1365; b) M. Weck, B. [30] Hydrogel formation upon addition of MgCl2 to a small DL of
Mohr, J.-P. Sauvage, R. H. Grubbs, J. Org. Chem. 1999, 64,
5463–5471.
macrocycles was recently reported by Otto and co-workers. In
this case the library was still dynamic during polymerization
because treatment with MgCl2 induced the sorting of the li-
brary, significantly altering the oligomer distribution, see: J. Li,
I. Cvrtila, M. Colomb-Delsuc, E. Otten, S. Otto, Chem. Eur. J.
2014, 20, 15709–15714.
[15] G. C. Vougioukalakis, R. H. Grubbs, Chem. Rev. 2010, 110,
1746–1787.
[16] Each member Ci of the DL will be present as a mixture of
geometrical isomers because each double bond can have either
a cis or trans configuration. This means that two isomers are
possible for C1, three for C2, four for C3, six for C4 and so on.
[17] The peaks at m/z = 680, 927, and 1359 in Figure 3 have been
assigned to C3 + H+ + Cu+, C2 + Cu+, and C3 + Cu+ species,
[31] X. Liu, X. Li, Y. Chen, Y. Hu, Y. Kishi, J. Am. Chem. Soc.
2012, 134, 6136–6139.
Received: September 16, 2015
Published Online: October 29, 2015
7510
www.eurjoc.org
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2015, 7504–7510