Pyridyl-decorated self-folding heptaamide cavitands as ligands in the rhodium-catalyzed hydrogenation of norbornadiene

Article abstract of DOI:10.1002/ejoc.201402385

The different binding geometries exhibited in solution by the Rh ^I cationic complexes of three regioisomeric self-folding heptaamide cavitands, each decorated with one pyridyl group at the upper rim, are taken into account to explain the diverse distributions of products obtained when these complexes are employed as catalysts for the hydrogenation of norbornadiene. Copyright

Full text of DOI:10.1002/ejoc.201402385

FULL PAPER
DOI: 10.1002/ejoc.201402385
Pyridyl-Decorated Self-Folding Heptaamide Cavitands as Ligands in the
Rhodium-Catalyzed Hydrogenation of Norbornadiene
[a]
[a,b]
ˇ
Sasa Korom and Pablo Ballester*
Keywords: Cavitands / Homogeneous catalysis / Hydrogenation / Rhodium
The different binding geometries exhibited in solution by the
RhI cationic complexes of three regioisomeric self-folding
heptaamide cavitands, each decorated with one pyridyl
group at the upper rim, are taken into account to explain the
diverse distributions of products obtained when these com-
plexes are employed as catalysts for the hydrogenation of
norbornadiene.
Introduction
The application of supramolecular concepts to catalysis
is targeted through different approaches.[1–7] Among them,
the encapsulation of catalytic metal centers within molec-
ular assemblies has proven to be highly efficient.[8–10] By
this approach, the stability,[11,12] reactivity,[13] and selectiv-
ity[11,14,15] normally observed for the catalytic metal center
free in solution can be radically altered and modified.
Encapsulation of a catalytic metal center has been
achieved successfully through ligand-template directed as-
sembly of the container.[16–18] Alternatively, an organome-
tallic complex can be used in the form of a guest in a molec-
ular or supramolecular container.[19] In the latter case, the
molecular vessel acts as a second coordination sphere li-
gand. Recently, one of us[11] reported that self-folding oc-
taamide cavitand 1[20,21] (Figure 1), based on a resorcin[4]-
arene scaffold, binds the organometallic complex
[Rh(NBD)2]·BF4 (3·BF4) with high affinity in dichloro-
methane solution. The binding process mainly produces a
1:1 inclusion complex 3+ʚ1. The inclusion process, how-
ever, also induces the partial dissociation of one of the
NBD ligands of 3+, affording a small quantity of another
inclusion complex – (NBD)Rh+ʚ1.
Interestingly, the mixture of the two inclusion cationic
RhI complexes was found to be catalytically competent in
the hydrogenation of NBD. When the hydrogenation is per-
formed on a laboratory scale, with the supramolecular sys-
tem 1 (1.2 mol-%) and 3+ (1.0 mol-%), the GC/FID analysis
of the crude product reveals that it is composed of nor-
[a] Institute of Chemical Research of Catalonia (ICIQ),
Av. Països Catalans 16, 43007 Tarragona, Spain
E-mail: pballester@iciq.es
http://www.iciq.es/portal/325/default.aspx
[b] Catalan Institution for Research and Advanced Studies
(ICREA)
Figure 1. Molecular structures of the self-folding cavitands octaam-
ide 1 and heptaamides 2a–d, together with the cationic RhI com-
plex 3+, norbornadiene (NBD, 4), and the monoamides 5a–d used
as model systems.
Pg. Lluís Companys 23, 08010 Barcelona, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201402385.
4276
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 4276–4282

Cavitands as Ligands in Rh-Catalyzed Hydrogenation
bornene (6, 72%), nortricycline (7, 3.7%), and dimer 8 and N-phenylbenzamide (5d) were produced as additional
(24.2%, Figure 2). This finding constitutes a remarkable re- reference systems (see Figure 1 for structures).
sult because the hydrogenation of NBD promoted by 3+
The X-ray structures of cavitands 2a, 2b, and 2d revealed
alone (1 mol-%) produces almost exclusively dimer 8 as the that they each adopted the vase conformation in the solid
result of the known reductive NBD dimerization.[22] We hy- state. This conformation is stabilized by the formation of
pothesized that the dimer 8 also detected in the NBD cata- an array of six intramolecular hydrogen bonds between the
lytic hydrogenation in the presence of the mixture of in- amide functions at the upper rim.[20] In the cases of 2a and
clusion complexes 3+ʚ1 and (NBD)Rh+ʚ1 must be pro- 2b, one molecule of the crystallization solvent is included
duced by the presence of a small quantity of 3+ that is re- in the aromatic cavity. It is worth noting that cavitands 2
leased into the solution during hydrogenation.
are chiral molecules possessing two elements of asymmetry.
The nonsymmetric substitution of one of the four phenyl-
dioxo bridges creates asymmetry in the covalent structure
of a cavitand 2, resulting in chiral stereogenic carbon atoms
(Figure 3 side views). In this sense, enantiomers of 2 can
easily be differentiated by the reverse absolute configura-
tions of the methine carbon atom (R or S) embedded in the
nine-membered ring encompassing the phenyldioxo bridge
with the single amide.
Figure 2. Line drawing structures of norbornene (6), nortricyclene
(7), dimer 8, and dimer 10.
Here we disclose the synthesis of a series of regioisomeric
heptaamide self-folding cavitands 2a–c, which are structur-
ally related to 1 but each possess a single pyridyl residue
installed at the upper rim.
In addition, we report the results obtained for the RhI-
catalyzed hydrogenation of NBD in the presence of these
heptaamide cavitands 2 as ligands. The design of the uni-
molecular pyridyl containers 2 for the inclusion of the RhI
metal center is based on the combination of two supra-
molecular strategies: (a) ligand-templated assembly, and
(b) use of an organometallic cationic complex as the guest
in the molecular container. Our expectations were that the
pyridyl residue would be able to act in a cooperative man-
ner with the electron-rich interior of the cavitand (cation–
π), thus increasing the thermodynamic stabilities of the RhI
caviplexes derived from 2 relative to the previously reported
system based on the unfunctionalized octaamide cavitand
1. If this were the case, the release of 3+ or (NBD)Rh+
complexes in solution during the hydrogenation should be
reduced, and the formation of dimer 8 might even be elim-
inated.
Figure 3. Side and top views of the X-ray structures of the racemic
pair of diastereoisomers present in the crystal packing of 2b:
(a) (M,R)-2b and b) (P,S)-2b. Nonpolar hydrogen atoms of 2b have
been omitted for clarity. The included hexane (from the solvent)
molecule is shown as a CPK model.
The other element of asymmetry is of supramolecular
origin.[24] It originates from the unidirectional orientation
of the carbonyl groups of the amides at the upper rim (Fig-
ure 3, top views). The amides’ directionality derives from
the formation of an intramolecular belt of hydrogen
Results and Discussion
Synthesis of the Cavitands and Assignment of
Conformation
The regioisomeric pyridyl cavitands 2a–c and the phenyl bonds.[25] Because of this second element of asymmetry, ca-
heptaamide cavitand 2d were prepared by acylation of the vitands 2 can exist as mixtures of two diastereoisomeric
known monoamino hexaamide cavitand 2e[23] with the cor- pairs of mutual enantiomers (four diastereomers).
responding acylpyridyl or benzoyl chlorides, respectively.
Interestingly, only one of two possible racemic pairs is
We used Schotten–Baumann conditions to perform the re- detected in the packings of single crystals obtained for 2b
actions. The yields ranged from 70–85%. Cavitand 2d, with and 2d [(M,R)-2 and (P,S)-2]. The preferred sense of orien-
a phenyl group instead of a pyridyl residue, was prepared tation (P/M) of the belt of hydrogen-bonded amides is con-
as a reference model system. By analogous synthetic pro- trolled by the favoured role of the single amide as hydrogen-
cedures, a series of N-phenylpyridinecarboxamides 5a–c bond acceptor.
Eur. J. Org. Chem. 2014, 4276–4282
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4277

S. Korom, P. Ballester
FULL PAPER
In the particular case of 2a, the crystal packing shows
that the aromatic cavity is filled with one acetone molecule
and also with the ethyl group of an adjacent cavitand mole-
cule, forming a dimeric aggregate (Figure 4).
Figure 4. Solid-state structure of the dimer of 2a. Bifurcated hydro-
gen bonds are depicted as dotted lines. The molecule of acetone
(from the solvent) included in 2a is shown as a CPK model. Non-
polar hydrogen atoms in 2a have been omitted for clarity.
The dimer is held together by four bifurcated intermo-
lecular hydrogen bonds established between the carbonyl
oxygen atom of the single-amide group of each monomer
and the NH groups of two amide groups of the adjacent 2a
molecule. The intermolecular interactions disrupt the unidi-
rectional orientation of the amide groups at the upper rim
of 2a. Two enantiomers of 2a are present in each dimer.
Figure 5. Energy-minimized structures of the different metallo
complexes resulting from the interaction of 2b with 3+: (a) 3+ʚ2b,
(b) (NBD)Rh+ʚ2b, (c) endo-(NBD)Rh+·2b, and (d) exo-(NBD)
Rh+·2b. The RhI metal and the NBD ligands are shown as CPK
models; 2b is displayed in stick representation. The nitrogen atom
of the pyridyl residue is highlighted as a larger ball. Nonpolar hy-
drogen atoms in 2b have been omitted for clarity.
1
The H NMR spectra of CD2Cl2 solutions of cavitands
2a–d each feature a single set of proton signals consistent
with C1 symmetry. The signals of the amide protons are
downfield-shifted, suggesting their involvement in hydrogen
bonding. The four chemically non-equivalent methine pro-
tons resonate at δ ≈ 5.6 ppm. Taken together, these results
suggest that cavitands 2 exist as enantiomeric pairs of single
diastereoisomers in solution, adopting the vase conforma-
tion.[26] These results are in agreement with the solid-state
data and with reported findings for related heptaamide cav-
itands.[23]
of 3+ (Figure 5, a) or (NBD)Rh+ (Figure 5, b) in their aro-
matic cavities without assistance of the coordination bond
with the pyridyl nitrogen atom.
The calculated energies (MM3 as implemented in SCIG-
RESS v3.0) for the complexes of 2a/2b with (NBD)Rh+ in
exo and endo coordination geometries indicate that the for-
mer is energetically preferred for caviplex (NBD)Rh+·2a
whereas the latter is favored in the case of (NBD)Rh+·2b.
Because the endo-(NBD)Rh+·2b complex features a co-
operative binding mode, which combines the formation of
a coordination N–RhI bond with cation–π interactions, we
expected that its thermodynamic and kinetic properties
Binding Studies of Cavitands 2 and Model Systems 5 with
3·BF4
Simple molecular modeling studies suggested two dif- would be improved in relation to those of 2a or the parent
ferent geometries (endo and exo) for the coordination com- 3+ʚ1 complex. This improvement should translate into re-
plexes formed through the interaction of the pyridyl nitro- duced leaking of (NBD)Rh+ into the bulk solution under
gen atom in 2a and 2b and the rhodium metal of (NBD) hydrogenation conditions. A recent computational study
Rh+. The formation of the N–Rh bond induces the dissoci- showed that the reductive dimerization of NBD to afford
ation of one NBD ligand in 3+ (vide infra). In the exo bind- dimer 8 requires the coordination of three NBD molecules
ing geometry the (NBD)Rh+ unit would point away from to the RhI center.[28] For steric reasons, three molecules of
the aromatic cavity (Figure 5, d).
NBD cannot coordinate simultaneously to RhI if the metal
Conversely, in the endo conformer the (NBD)Rh+ unit is center is included in the cavity of 1 or 2. In short, cavitand
directed towards the aromatic cavity (Figure 5, c). Com- 2b stands out as an ideal candidate to suppress the forma-
pound 2c, with the nitrogen atom in the para position of tion of dimer 8 observed in the rhodium-catalyzed hydro-
the pyridyl substituent, would be expected to direct the genation of NBD.
(NBD)Rh+ unit away from the aromatic cavity in any of its
conformers.
With the aim of studying the binding properties (towards
3+) of the pyridyl carboxamide residues installed at the up-
For the three pyridyl cavitands 2a–c, owing to the weak per rim of each of cavitands 2a–c in isolation from those of
coordination interaction that exists between RhI and nitro- the close-by aromatic cavity, we synthesized pyridyl carbox-
gen atoms,[27] it is sensible to consider the direct inclusion amides 5a–d as model systems and studied their binding
4278
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 4276–4282

Cavitands as Ligands in Rh-Catalyzed Hydrogenation
properties. The interactions between 5a–d and 3·BF4 were discussed above, we assigned endo and exo geometries to
probed by 1H NMR spectroscopy. The addition of 0.8 equiv. the two coordination complexes of 2a·Rh(NBD)+. The two
of 3·BF4 to separate dichloromethane solutions of the three complexes are engaged in a chemical exchange process. The
pyridine carboxamides 5a–c (4.5 mm) induced significant complexes’ interconversion is slow on the NMR timescale
changes in the chemical shifts of the signals of the amide because it must occur through a sizeable conformational
NH and pyridyl protons relative to those in the free ligands. change of the cavitand (vase-to-kite) rather than a simple
The results of the titration experiments indicate that the co- C–C bond rotation. The small integral measured for the
ordination of carboxamides 5a–c to 3+ induces different signal that resonates furthest upfield (δ = –2.64 ppm) and
levels of dissociation of one of the NBD ligands, yielding is assigned to the Hc protons of the Rh(NBD)+ included in
free NBD and the 1:1 heteroleptic complexes 5a–c·Rh- the aromatic cavity of 2a indicates that the exo conformer
(NBD)+. The complexes are each stabilized by the formation is the major species (Ͼ90%) in solution.
of a N–RhI coordination bond. The protons of the free and
The use of cavitand 2b (3-carboxypyridine) produced a
bound NBDs are observed as separate signals, indicating that more complex distribution of species. From the integrals of
they are involved in a chemical exchange process that is slow selected proton signals, we calculated the percentages of the
on the NMR timescale. Conversely, a single set of proton different species in solution to be: 62% of the exo-
signals is observed for each of the N-phenyl carboxamides 2b·Rh(NBD)+ conformer, 21% of the endo-2b·Rh(NBD)+
5a–c. Moreover, the extents to which the 1:1 complexes 5a– conformer, and 17% of the inclusion complex 3+ʚ2b not
c·Rh(NBD)+ are formed in equimolar solutions depend on assisted by the formation of a coordination bond. Contrary
the position of the nitrogen atom in the pyridyl substituent to our expectations, the endo-2b·Rh(NBD)+ complex is not
of the ligand. In the case of 5a, in which the pyridyl nitrogen the major species in solution. Interestingly, the protons of
atom is ortho to the amide function, the complex 5a·Rh- the bound NBD in endo-2b·Rh(NBD)+ resonate as dia-
(NBD)+ is formed quantitatively in an equimolar mixture of stereotopic signals, due to cavity’s chirality imparted by the
partners. Most likely, this is due to the existence of a ditopic unidirectional orientation of the amide groups at the upper
interaction between the ligand 5a and the RhI metal center. rim. By using competitive pairwise experiments we also de-
Additional evidence for RhI coordination to the carbonyl termined that the complexes of cavitand 2b had a minimal
oxygen in 5a·Rh(NBD)+ was provided by the downfield shift thermodynamic advantage with respect to 3+ʚ1. This result
experienced by the amide NH. In striking contrast, ligand was again completely unexpected on the basis of our initial
5d, lacking the pyridyl residue, induces the dissociation of hypothesis.
3+ into NBD and the formation of 5d·Rh(NBD)+ to a mini-
The third pyridyl cavitand 2c (4-carboxypyridine), with
mum extent. Therefore, ligand 5d must coordinate to the its N atom pointing away from the aromatic cavity, pro-
RhI through its carbonyl oxygen atom. It is worth men- duced a very different distribution of complexes. The major
tioning that in this latter case the chemical exchange pro- species detected in solution is the inclusion complex (NBD)
cesses both of free and bound NBD and of free and bound Rh+ʚ2c (68%), which does not involve a coordination
ligand 5d are slow on the NMR timescale.
bond, probably due to geometric constraints. The coordina-
Next, we investigated the binding properties of the cavit- tion complex 2c·Rh(NBD)+ (23%) and the inclusion com-
and series 2 towards 3+. Using 1D and 2D H NMR spec- plex 3+ʚ2c (9%) are also present in the mixture. The pro-
1
troscopy, we analyzed the compositions of separate CD2Cl2 tons of the external NBD ligands in the last complex also
solutions containing 1 equiv. of 3+ and 1.2 equiv. of each resonate as diastereotopic signals.
cavitand 2a–d. In all cases, the proton signals corresponding
Finally, cavitand 2d (carboxyphenyl) produced the in-
to the free NBD (4) were easily identified (Ha, Hb, and Hc; clusion complexes (NBD)Rh+ʚ2d (50%) and 3+ʚ2d (45%)
see the Supporting Information). This observation parallels in similar quantities, with a small amount of free 3+ (5%)
the findings made with the model systems 5a–c. In short, remaining in solution.
the coordination of the pyridyl substituents in cavitands 2a–
c with 3+ induces the dissociation of one of their NBD li-
gands to different extents. The presence of cross-peaks in
2D EXSY experiments revealed that the protons of free
NBD were involved in chemical exchange processes with
Rhodium-Catalyzed Hydrogenations of NBD with the
Model Systems and the Cavitands as Ligands
bound NBD molecules in more than one magnetic environ-
The coordination complexes of 5a–d and 3·BF4 were
ment. The olefinic proton (Ha) of the free NBD was the tested as catalysts in the hydrogenation of norbornadiene 4
best signal to analyze the multi-site exchange systems. In (1 mol-% of 3·BF4, 1.2 mol-% of 5a–d, room temp., 1 atm
the case of 2a (2-carboxypyridine cavitand) interacting with H2). The GC/FID analysis of the hydrogenation crude
3+, 1 equiv. of the free NBD is released and found to ex- products after 45 min reaction time showed significant dif-
change with NBD units in two different bound states. The ferences in composition. The results obtained are summa-
quantitative dissociation of 3+ is in agreement with the re- rized in Table 1.
sult obtained for the model 5a. Most likely, the ditopic
Several conclusions can be drawn from the data: (1) the
(N,O) metal-ligand interaction present in the 2a·Rh- monotopic coordination of the RhI metal center to the pyr-
(NBD)+ complex is responsible for its high thermodynamic idyl nitrogen atom in ligands 5b and 5c produces highly
stability. On the basis of the molecular modeling studies active species for the hydrogenation of NBD that exclusively
Eur. J. Org. Chem. 2014, 4276–4282
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4279

S. Korom, P. Ballester
FULL PAPER
Table 1. Percentages of NBD (4) consumed (“con”) after 45 min Probably the inclusion complexes of 2c are competent in
reaction time and the resulting product distributions (%) for the
norbornene production but they are less active than the N-
laboratory-scale catalytic hydrogenations performed in this study
coordinated counterpart 2c·Rh(NBD)+ that mainly yields
(see the Supporting Information for experimental details).
nortricyclene (7).
Entry System
% 4 con Product distribution in %[a]
Finally, the results obtained for the system 2d+3+ also
support the competence of the inclusion complexes (NBD)
Rh+ʚ2d and 3+ʚ2d in the production of norbornene (6).
However, the percentages of produced norbornene are
lower than for the system 1+3+. We explain this result by
considering that the elimination in 2d of one of the amide
groups at the upper rim, with respect to 1, distorts the aro-
matic cavity and makes its inclusion complexes with
Rh(NBD)+ thermodynamically less stable. In turn, the re-
duced thermodynamic stability of the inclusion complexes
of 2d relative to 1 translates into the presence of significant
amounts of free 3+ in solution, as already observed in the
binding experiment. By the same token, the release of
Rh(NBD)+ or 3+ into solution during hydrogenation must
be favored for 2d in relation to 1. The low levels of nortricy-
6
7
8
10
other
1
2
3
4
5
6
7
8
3+[b]
80
30
55
30
35
35
30
60
50
60
75
0.9
2.9
1.0
0
73.4
0
15.6
0
0
0
0
0
0
35.4
72.9
24.7
92.2
2.0
0
0
0
0
0
0
0
0
4.9
0
0
0
0
0
0
0
3+[c]
1 + 3+
2a + 3+
5a + 3+
2b + 3+
5b + 3+
2c + 3+
5c + 3+
2d + 3+
5d + 3+
81.7
60.1
47.6
42.4
0
18.3
0
64.6
1.1
0.7
39.9
52.4
57.6
100
81.7
100
0
9
10
11
0
1.2
1.3
23.5
[a] Values (%) are corrected under the assumption that the response
factor of the dimer is twice that of the monomer. [b] Without pre-
activation of catalyst. [c] No hydrogen.
yield nortricyclene (7, Entries 7 and 9), (2) the ditopic N,O clene (7) present in the crude products of Entries 1, 2, 3, 10,
coordination of the RhI center exerted by ligand 5a hydro- and 11 attest that the exo species with N coordination to
genates NBD to produce a close to 50:50 mixture of nor- the RhI metal center must be responsible for its production.
bornene (6) and nortricyclene (7, Entry 5), and (3) ligand
5d, lacking the pyridyl group, does not have a significant
effect on the catalytic properties exhibited by 3+ alone (En-
tries 1 and 11).
Conclusions
The structural modifications of cavitand 1 represented by
the cavitand series 2a–c are not successful in allowing their
application as ligands for the exclusive rhodium-catalyzed
It is worth noting that dimer 10 is produced in significant
amounts in Entries 1 and 11. On the other hand, catalyst
3+ in the absence of H2 induces the dimerization of NBD
to yield dimer 10 (Entry 2).[29]
hydrogenation of NBD (4) to norbornene (6). The emer-
gence of exo geometries in the complexes formed between
2a–c and 3+ provides a new reaction pathway that was not
important in the case of 1. The exo complexes feature a
coordination bond between the nitrogen atom of the pyridyl
residue at the upper rim of cavitands 2a–c and the metal
center of (NBD)Rh+. In agreement with the results ob-
tained for the model systems 5a–c, the exo coordination
complexes are competent in producing nortricyclene (7) as
the main hydrogenation product of NBD. In addition, the
significant levels of norbornene (6) produced in the hydro-
genation of NBD when the supramolecular systems 2a–d
and 3+ are employed as catalysts is attributed to a reaction
pathway involving the inclusion of the RhI metal center in
the deep cavities of the hosts. In the endo complexes of 2a
and 2b the cavitand may act as a first- or second-sphere
ligand depending on the type of interaction established with
the included rhodium center. The obtained results are not
conclusive in supporting the hypothesis that endo complexes
assisted by coordination with the pyridyl residue can also
modulate the conversion of NBD into nortricyclene (7).
We learned during the course of this study that the reac-
tion conditions used to perform the hydrogenation experi-
ments result in a slow diffusion of hydrogen[30] into the re-
action mixture. For this reason, and for the particular case
of catalyst 3+, we must conclude that the reactions produc-
ing the hydrogenated dimer 8 and non-hydrogenated 10 be-
come competitive. Next, we evaluated the catalytic perform-
ances of the caviplex systems in the hydrogenation of NBD
under reaction and analytical conditions identical to those
above (1 mol-% of 3·BF4 and 1.2 mol-% of 2a–d at room
temp. and under 1 atm H2).
The 2a+3+ system (Entry 4) produces norbornene (6)
and nortricyclene (7) in a 3:2 ratio. This result corresponds
to a moderate increase in the amount of norbornene pro-
duced relative to the model system (Entry 5). We ascribe
this increase in norbornene production to a high catalytic
activity of the endo-2a·Rh(NBD)+ complex present in solu-
tion in low amount (Ͻ10%). At first sight, the system
2b+3+ seems to be less efficient in the production of nor-
bornene than 2a+3+. However, if the product distribution
obtained for 2b+3+ is compared with that of the corre-
sponding model systems 5b+3+ then the opposite conclu-
sion is reached. As we did for the 2a+3+ system, we also
ascribe the norbornene production for 2b+3+ to the endo-
2b·Rh(NBD)+ complex, now present in solution in larger
amounts (21%). Consistently with this explanation, the sys-
tem 2c+3+, lacking an endo geometry complex assisted by
Experimental Section
General Methods: All reagents and solvents were obtained from
commercial suppliers and used without further purification unless
otherwise stated. Cavitands 1 and 2f were synthesized by the experi-
mental procedure published by Rebek.[21,23] Dimer molecules 8 and
N-coordination, is less effective in norbornene production. 10 were synthesized by the procedure published by Katz.[22,29] 1D
4280
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 4276–4282

Cavitands as Ligands in Rh-Catalyzed Hydrogenation
and 2D NMR spectra were recorded with a Bruker Avance 400 or
a Bruker Avance 500 spectrometer. All deuterated solvents (Sigma–
Aldrich) were used without any further purification. Chemical
shifts are given in ppm and peaks are referenced relative to the
solvent residual peak (δCHDCl = 5.32 ppm, δCHCl = 7.26 ppm,
δDMSO = 2.49 ppm). All NMR J values are given in Hz. All MS
were recorded with a Bruker MALDI-TOF Autoflex instrument.
Product distributions were analyzed by use of an Agilent gas chro-
matograph 6890N with a FID detector. Melting points were deter-
mined with a Mettler Toledo MP70 Melting Point System.
9.23 (s, 1 H), 9.01 (s, 1 H), 8.91 (s, 1 H), 8.76 (s, 1 H), 8.64 (dd, J
= 4.8, 1.7 Hz, 1 H), 8.49 (s, 1 H), 8.10 (s, 1 H), 7.78 (dt, J = 7.9,
1.9 Hz, 1 H), 7.70 (s, 1 H), 7.60 (s, 1 H), 7.58 (s, 1 H), 7.56 (s, 1
H), 7.53 (s, 1 H), 7.51 (s, 2 H), 7.50 (s, 1 H), 7.36 (s, 2 H), 7.34 (s,
2 H), 7.29 (s, 2 H), 7.27 (s, 2 H), 7.27 (s, 1 H), 7.23 (s, 1 H), 5.74
(t, J = 8.3 Hz, 1 H), 5.67–5.59 (m, 2 H), 5.59–5.52 (m, 1 H), 2.62–
2.18 (m, 18 H), 1.70 (hept, J = 7.9 Hz, 2 H), 1.37–0.89 (m, 27 H),
2
3
0.77 (t, J = 7.6 Hz, 3 H) ppm. FTIR: ν = ν(C–N) 1403; ν(Ar) 1597,
˜
892; ν(C=O) 1664; ν(alkyl) 2967, 2934, 2872; ν(–NH···O=)
3251 cm–1. MS (MALDI, +): calcd for C84H82N8O15 + H+ 1443.6;
found 1443.6.
General Procedure for the Synthesis of Cavitands 2a–d: The nitro
group in 2f was reduced to the corresponding amine 2e followed
by acylation under Schotten–Baumann conditions to afford the de-
sired cavitands 2a–d in good yields after purification by column
chromatography.
Cavitand 2c: M.p. Ͼ250 °C with decomposition. 1H NMR
(500 MHz, [D2]dichloromethane): δ = 9.59 (s, 1 H), 9.45 (s, 1 H),
9.32 (s, 1 H), 8.98 (s, 1 H), 8.77–8.72 (m, 1 H), 8.61–8.55 (m, 2 H),
8.14 (s, 1 H), 7.85–7.81 (m, 1 H), 7.71 (s, 1 H), 7.58 (s, 1 H), 7.56
(s, 1 H), 7.55 (s, 1 H), 7.54 (s, 1 H), 7.52 (s, 1 H), 7.50 (s, 1 H),
7.37 (s, 1 H), 7.35 (s, 2 H), 7.34 (s, 1 H), 7.32 (s, 1 H), 7.30 (s, 1
H), 7.29 (s, 1 H), 7.28 (s, 1 H), 7.27 (s, 1 H), 7.24 (s, 2 H), 7.23 (s,
1 H), 5.73 (t, J = 8.4 Hz, 1 H), 5.63 (dt, J = 12.5, 8.3 Hz, 2 H),
5.57 (t, J = 8.3 Hz, 1 H), 2.59–2.20 (m, 20 H), 1.33–1.17 (m, 18
H), 1.11–1.03 (m, 3 H), 1.04–0.95 (m, 6 H), 0.77 (t, J = 7.6 Hz, 3
A typical synthetic procedure is described. A solution of 2f
(428 mg, 0.313 mmol, 1 equiv.) in THF (25 mL) was placed in a
50 mL two-necked flask. A catalytic amount of Raney-Ni (washed
four times with THF, followed by separation with a magnet and
decantation) was added to this solution. The mixture was stirred
with a magnetic bar and purged with hydrogen (by passing hydro-
gen over the solution for a few seconds). The hydrogenation reac-
tion was performed under hydrogen (1 atm) and at 40 °C for 2 h.
The resulting solution was filtered through a 0.45 μm syringe filter
to remove the catalyst, and the solvent was removed under reduced
pressure, affording 2e (410 mg, 0.306 mmol, 98%).
H) ppm. FTIR: ν = ν(C–N) 1401; ν(Ar) 1599, 892; ν(C=O)
˜
1658; ν(alkyl) 2963, 2935, 2874; ν(–NH···O=) 3253 cm–1. MS
(MALDI, +): calcd for C84H82N8O15 + H+ 1443.6; found 1443.6.
Cavitand 2d: Crystals suitable for X-ray crystal structure analysis
were obtained.[31] M.p. Ͼ250 °C with decomposition. 1H NMR
(400 MHz, [D2]dichloromethane): δ = 9.43 (s, 1 H), 9.38 (s, 1 H),
9.35 (s, 1 H), 9.11 (s, 1 H), 8.36 (s, 1 H), 8.05 (dd, J = 8.3, 1.2 Hz,
1 H), 7.92 (s, 1 H), 7.73 (s, 1 H), 7.63 (s, 1 H), 7.62 (s, 1 H), 7.60
(s, 1 H), 7.57 (s, 1 H), 7.50 (s, 1 H), 7.48 (s, 1 H), 7.47 (s, 2 H),
7.45 (s, 2 H), 7.43 (s, 1 H), 7.37 (s, 1 H), 7.37 (s, 1 H), 7.36 (s, 1
H), 7.35 (s, 2 H), 7.34 (s, 2 H), 7.28 (s, 1 H), 7.27 (s, 1 H), 7.25 (s,
1 H), 5.72 (t, J = 8.3 Hz, 1 H), 5.68–5.59 (m, 2 H), 5.64–5.53 (m,
1 H), 2.55–2.18 (m, 16 H), 1.86–1.57 (m, 4 H), 1.31–1.16 (m, 12
H), 1.04 (ddq, J = 30.3, 15.6, 7.5 Hz, 15 H), 0.76 (t, J = 7.5 Hz, 3
The amino cavitand 2e was immediately used in the next reaction
step – acylation of the amine group under Schotten–Baumann con-
ditions to produce cavitands 2a–d. To obtain 2b, for example, the
experimental procedure consisted of adding a solution of potas-
sium carbonate (1.2 g, 8.68 mmol, 28 equiv.) in water (18 mL) to a
solution of 2e (410 mg, 0.306 mmol, 1 equiv.) in ethyl acetate
(25 mL). Next, solid nicotinoyl chloride (142 mg, 0.96 mmol,
3 equiv.) was added to the above solution in small portions with
vigorous stirring. The mixture was stirred at room temperature for
16 h under nitrogen. After this time, the organic phase was sepa-
rated, dried with anhydrous sodium sulfate, and filtered, and the
solvent was evaporated in vacuo. The solid residue obtained was
purified by flash column chromatography on silica with a gradient
of methanol (0–5%) in dichloromethane to afford 2b (390 mg,
0.27 mmol, 86%) as a solid.
H) ppm. FTIR: ν = ν(C–N) 1402; ν(Ar) 1598, 892; ν(C=O)
˜
1664; ν(alkyl) 2965, 2933, 2872; ν(–NH···O=) 3249 cm–1. MS
(MALDI, +): calcd for C84H82N8O15 + Na+ 1464.6; found 1464.6.
General Procedure for the Synthesis of Reference Ligands 5a–d: Un-
der Schotten–Baumann conditions analogous to those described
for the synthesis of cavitands 2a–d, aniline was acylated to produce
reference ligands 5a–d. Crystallization from water/acetone mixtures
afforded the pure compounds, in 75% yield for 5a, 64% yield for
5b, 43% yield for 5c, and 89% yield for 5d.
By use of similar experimental procedures the cavitand 2a was iso-
lated in 77% yield, 2c in 69% yield and 2d in 79% yield.
Cavitand 2a: Crystals suitable for X-ray crystal structure analysis
were obtained.[31] M.p. Ͼ250 °C with decomposition. 1H NMR
(400 MHz, [D2]dichloromethane): δ = 9.79 (s, 1 H), 9.56 (s, 1 H),
9.54 (s, 1 H), 9.13 (s, 1 H), 8.99 (s, 1 H), 8.72 (s, 1 H), 8.59 (d, J =
4.2 Hz, 1 H), 8.54 (s, 1 H), 8.18 (d, J = 2.6 Hz, 1 H), 8.11 (d, J =
7.8 Hz, 1 H), 7.97 (s, 1 H), 7.89 (td, J = 7.7, 1.7 Hz, 1 H), 7.79 (s,
1 H), 7.77 (s, 1 H), 7.75 (s, 1 H), 7.73 (s, 1 H), 7.50 (ddd, J = 7.6,
4.8, 1.1 Hz, 1 H), 7.42 (s, 1 H), 7.40 (s, 1 H), 7.37 (s, 1 H), 7.36 (s,
1 H), 7.34 (s, 1 H), 7.31 (s, 1 H), 7.27 (s, 1 H), 7.25 (s, 1 H), 7.25
(s, 1 H), 7.24 (s, 1 H), 6.84 (dd, J = 8.7, 2.2 Hz, 1 H), 5.75–5.57
(m, 4 H), 2.52–2.21 (m, 20 H), 1.22 (ddd, J = 16.2, 13.0, 7.6 Hz,
15 H), 1.05 (td, J = 7.1, 4.3 Hz, 9 H), 0.99 (td, J = 7.2, 3.5 Hz, 6
N-Phenylpyridine-2-carboxamide (5a): 1H NMR (500 MHz,
CDCl3): δ = 10.03 (s, 1 H), 8.62 (ddd, J = 4.7, 1.6, 0.9 Hz, 1 H),
8.31 (dt, J = 7.8, 1.0 Hz, 1 H), 7.91 (td, J = 7.7, 1.7 Hz, 1 H), 7.79
(dt, J = 8.7, 1.6 Hz, 2 H), 7.49 (ddd, J = 7.6, 4.8, 1.2 Hz, 1 H),
7.43–7.33 (m, 2 H), 7.19–7.10 (m, 1 H) ppm.[32].
N-Phenylpyridine-3-carboxamide (5b): 1H NMR (500 MHz,
CDCl3): δ = 9.08 (d, J = 1.8 Hz, 1 H), 8.76 (dd, J = 4.8, 1.6 Hz, 1
H), 8.20 (dt, J = 7.9, 2.0 Hz, 1 H), 8.04 (s, 1 H), 7.64 (d, J = 7.8 Hz,
2 H), 7.43 (ddd, J = 7.9, 4.8, 0.8 Hz, 1 H), 7.41–7.35 (m, 2 H),
7.21–7.16 (m, 1 H) ppm.[33]
N-Phenylpyridine-4-carboxamide (5c): 1H NMR (500 MHz, [D6]-
DMSO): δ = 10.48 (s, 1 H), 9.06–8.56 (m, 2 H), 7.97–7.79 (m, 2
H), 7.76 (d, J = 7.7 Hz, 2 H), 7.37 (t, J = 7.9 Hz, 2 H), 7.13 (t, J
= 7.4 Hz, 1 H) ppm.[34]
H) ppm. FTIR: ν = ν(C–N) 1401; ν(Ar) 1599, 892; ν(C=O)
˜
1662; ν(alkyl) 2964, 2935, 2875; ν(–NH···O=) 3261 cm–1. MS
(MALDI, +): calcd for C84H82N8O15 + H+ 1443.6; found 1443.6.
Cavitand 2b: Crystals suitable for X-ray crystal structure analysis
were obtained.[31] M.p. Ͼ250 °C with decomposition. 1H NMR
(400 MHz, [D2]dichloromethane): δ = 9.53 (s, 1 H), 9.41 (s, 1 H),
N-Phenylbenzamide (5d): 1H NMR (500 MHz, CDCl3): δ = 7.88 (s,
1 H), 7.86 (d, J = 1.5 Hz, 2 H), 7.65 (d, J = 7.8 Hz, 2 H), 7.59–
Eur. J. Org. Chem. 2014, 4276–4282
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4281

S. Korom, P. Ballester
FULL PAPER
[12] D. Fiedler, R. G. Bergman, K. N. Raymond, Angew. Chem. Int.
Ed. 2006, 45, 745–748; Angew. Chem. 2006, 118, 759.
[13] Z. J. Wang, C. J. Brown, R. G. Bergman, K. N. Raymond, F. D.
Toste, J. Am. Chem. Soc. 2011, 133, 7358–7360.
[14] A. Cavarzan, A. Scarso, P. Sgarbossa, G. Strukul, J. N. H.
Reek, J. Am. Chem. Soc. 2011, 133, 2848–2851.
[15] V. F. Slagt, J. N. H. Reek, P. C. J. Kamer, P. W. N. M.
van Leeuwen, Angew. Chem. Int. Ed. 2001, 40, 4271–4274; An-
gew. Chem. 2001, 113, 4401.
[16] A. W. Kleij, J. N. H. Reek, Chem. Eur. J. 2006, 12, 4219–4227.
[17] V. F. Slagt, P. C. J. Kamer, P. W. N. M. van Leeuwen, J. N. H.
Reek, J. Am. Chem. Soc. 2004, 126, 1526–1536.
7.51 (m, 1 H), 7.48 (t, J = 7.5 Hz, 2 H), 7.41–7.34 (m, 2 H), 7.16
(t, J = 7.4 Hz, 1 H) ppm.[32]
Catalytic Hydrogenation Reactions: A typical laboratory-scale
hydrogenation experiment consisted of dissolving 3·BF4 (20 μmol)
and the appropriate cavitand or model ligand (1, 2a–d, 5a–d,
25 μmol) in dichloromethane (HPLC grade, 6.8 mL). The resulting
solution was transferred to a 25 mL three-necked flask fitted with
a hydrogen inlet. The catalyst was preactivated by purging the flask
with hydrogen and stirring the mixture under 1 bar of hydrogen for
5 min. Next, neat norbornadiene (4, 2.12 mmol) was added in one
shot, and the reaction evolution was immediately sampled by tak-
ing small aliquots of the reaction mixture (100 μL). Aliquots were
taken every 15 min. Each aliquot was diluted with dichloromethane
(100 μL), filtered through a silica gel plug, and analyzed by GC/
FID.
[18] B. Kang, J. W. Kurutz, K. T. Youm, R. K. Totten, J. T. Hupp,
S. T. Nguyen, Chem. Sci. 2012, 3, 1938–1944.
[19] C. J. Brown, G. M. Miller, M. W. Johnson, R. G. Bergman,
K. N. Raymond, J. Am. Chem. Soc. 2011, 133, 11964–11966.
[20]
D. M. Rudkevich, G. Hilmersson, J. Rebek, J. Am. Chem. Soc.
1997, 119, 9911–9912.
Supporting Information (see footnote on the first page of this arti-
[21] D. M. Rudkevich, G. Hilmersson, J. Rebek, J. Am. Chem. Soc.
1998, 120, 12216–12225.
[22] R. J. Roth, T. J. Katz, Tetrahedron Lett. 1972, 13, 2503–2504.
[23] O. B. Berryman, A. C. Sather, A. Lledó, J. Rebek, Angew.
Chem. Int. Ed. 2011, 50, 9400–9403; Angew. Chem. 2011, 123,
9572.
1
cle): H NMR spectra and MALDI-MS isotopic distributions for
synthesized cavitands, 2D 1H NMR spectra of the encapsulation
1
complexes, H NMR spectra for competitive titration studies, GC/
FID traces for catalytic experiments.
[24] A. Szumna, Chem. Soc. Rev. 2010, 39, 4274–4285.
[25] The stereochemical designations are described in the following
sequence: axial chirality (P or M), defined by the unidirectional
(clockwise or counterclockwise) sense of rotation of the seven
amide carbonyl groups viewed from above the cavity, followed
by the absolute configuration, by application of the CIP rules,
of the methyne (R or S) in the nine-membered ring containing
the asymmetric phenyldioxo bridge.
Acknowledgments
The authors are grateful for funding from the Ministerio de Econ-
omía y Competitividad (MINECO) (CTQ2011-23014), the Genera-
litat de Catalunya DIUE (2009SGR6868), and the ICIQ Founda-
tion.
[26] J. R. Moran, J. L. Ericson, E. Dalcanale, J. A. Bryant, C. B.
Knobler, D. J. Cram, J. Am. Chem. Soc. 1991, 113, 5707–5714.
[27] A. J. Pardey, C. Longo, Coord. Chem. Rev. 2010, 254, 254–272.
[1] P. W. N. M. van Leeuwen (Ed.), Supramolecular Catalysis, 1st
ed., Wiley-VCH, Weinheim, Germany, 2008.
[2] M. Raynal, P. Ballester, A. Vidal-Ferran, P. W. N. M. [28] A. Hamilton, M. Gicquel, P. Ballester, C. Bo, Curr. Org. Chem.
van Leeuwen, Chem. Soc. Rev. 2014, 43, 1660–1733. 2013, 17, 1499–1506.
[3] M. Raynal, P. Ballester, A. Vidal-Ferran, P. W. N. M. [29] J. J. Mrowca, T. J. Katz, J. Am. Chem. Soc. 1966, 88, 4012–
van Leeuwen, Chem. Soc. Rev. 2014, 43, 1734–1787.
[4] A. Cavarzan, J. N. H. Reek, F. Trentin, A. Scarso, G. Strukul,
Catal. Sci. Technol. 2013, 3, 2898–2901.
4015.
[30] M. B. Krüger, C. Selle, D. Heller, W. Baumann, J. Chem. Eng.
Data 2012, 57, 1737–1744.
[31] CCDC-990079 (for 2a), -990080 (for 2b), and -990081 (for 2d)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[5] B. W. Purse, A. Gissot, J. Rebek, J. Am. Chem. Soc. 2005, 127,
11222–11223.
[6] T. Murase, Y. Nishijima, M. Fujita, J. Am. Chem. Soc. 2012,
134, 162–164.
[7] C. J. Hastings, M. P. Backlund, R. G. Bergman, K. N. Ray-
mond, Angew. Chem. Int. Ed. 2011, 50, 10570–10573; Angew. [32] L. Zhang, W. Wang, A. Wang, Y. Cui, X. Yang, Y. Huang, X.
Chem. 2011, 123, 10758.
Liu, W. Liu, J.-Y. Son, H. Oji, T. Zhang, Green Chem. 2013,
[8] J. Meeuwissen, J. N. H. Reek, Nature Chem. 2010, 2, 615–621.
[9] H. J. Hwang, J. R. Carey, E. T. Brower, A. J. Gengenbach, J. A.
Abramite, Y. Lu, J. Am. Chem. Soc. 2005, 127, 15356–15357.
[10] T. S. Koblenz, J. Wassenaar, J. N. H. Reek, Chem. Soc. Rev.
2008, 37, 247–262.
15, 2680–2684.
[33] W. Fang, Q. Deng, M. Xu, T. Tu, Org. Lett. 2013, 15, 3678–
3681.
[34] K. Motoshima, Y. Hiwasa, M. Yoshikawa, K. Fujimoto, A.
Tai, H. Kakuta, K. Sasaki, ChemMedChem 2007, 2, 1527–
1532.
[11] M. A. Sarmentero, H. Fernández-Pérez, E. Zuidema, C. Bo, A.
Vidal-Ferran, P. Ballester, Angew. Chem. Int. Ed. 2010, 49,
7489–7492; Angew. Chem. 2010, 122, 7651.
Received: April 10, 2014
Published Online: June 3, 2014
4282
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 4276–4282