This document is the Accepted Manuscript version of a Published Work that appeared
in final form in Crystal Growth & Design, copyright © American Chemical Society after
peer review and technical editing by the publisher. To access the final edited and
published work see http://pubs.acs.org/doi/pdf/10.1021/acs.cgd.6b01732

A Metal-Organic Framework Based on a TetraArylextended Calix[4]pyrrole Ligand: Structure
Control through the Covalent Connectivity in the
Linker.
Jordi Aguilera-Sigalat,a Cristina Sáenz de Pipaón,a Daniel Hernández-Alonso,a Eduardo C.
Escudero-Adán,a José Ramón Galan-Mascarós,*,a,b Pablo Ballester*,a,b
a

Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science and

Technology, Avgda. Països Catalans 16, 43007 Tarragona, Spain.
b

ICREA, Passeig Lluís Companys, 23, 03018 Barcelona, Spain

ABSTRACT. The preparation of isomeric metal-organic frameworks in which the network
topology is controlled by the different covalent connectivity of the organic ligand is an important
step forward in the design of new functional materials. In this context, macrocyclic organic ligands
able to accommodate suitable guests in their own polar internal cavities are appealing candidates
to act as multidentate linkers, which could potentially self-assemble into hierarchical porous

1

structures. Taking this into account, here we report the first successful attempt to incorporate a
tetraaryl-extended calix[4]pyrrole derivative into a transition metal-organic framework (MOF1)
by simply incorporating four terminal carboxylic functional groups at the upper rim of the
macrocyclic scaffold. Remarkably, the structures of the metal organic framework and its
transition-metal carboxylate clusters (secondary building units, SBU) are governed by the position
of the carboxylic substituent in the functionalized meso-aryl units of the linker. Only the tetrameso-arylextended tetracarboxylic calix[4]pyrrole isomer L1, substituted in a single metaposition of their aryl rings, yields a 2-D MOF architecture assembled through complex CuIIcarboxylate clusters (SBU). The clusters have higher nuclearity than the CuII2-(O2CR)4 paddle
wheels produced as SBUs during the assembly of the para-substituted counterpart L2.
Remarkably, the length of the dialkylformamide used as solvent (DMF or DEF) in the synthesis
of the CuII-organic materials derived from L2 played a key role in the structure of the final solid
material. The packing of discrete metal-mediated capsular dimers of L2 switched to that of 1-D
linear coordination polymers when the solvent’s alkyl chains were increased by a methylene unit.
Finally, ligand L3, which featured a longer alkyl spacer between the para-substituted
calix[4]pyrrole core and the terminal carboxylic groups than L2, self-assembled, exclusively, into
discrete capsular coordination dimers also mediated by CuII paddle wheel units.

1. INTRODUCTION
Metal-organic frameworks (MOFs), also referred to as porous coordination polymers, are a class
of hybrid materials built from metals ions (or clusters) and polydentate organic ligands (linkers).1,2
These crystalline materials with high surface areas and ideally permanent porosity have been used
in a wide range of applications, such as gas storage and separation,3 sensing4 or catalysis,5 among

2

others. The vast choice of organic ligands allows infinite possibilities for the preparation of MOFs
where the structure-properties relationship can be modulated accordingly.
The design of the structure’s ligand plays a key role in the preparation of these materials, which
could endow their porosity with a hierarchical configuration, which is with different pore sizes:
large resulting from the MOF framework and small defined by the ligand’s structure. In addition,
the ligand’s structure could also confer a certain degree of conformational flexibility to the MOF
that might be promoted via an external stimuli, such as the response to guests’ addition or light.6,7,8
The effect exerted by the position and length of the ligand’s substituent on the structure of the
resulting metal-organic framework has been demonstrated.9 For instance, Chen et al.10 reported
the influence of the substituent’s length when using an imidazole ligand. Thus, when 2methylimidazole was employed, a sodalite-type MOF structure was obtained, whereas changing
to 2-ethylimidazole led to a zeolitic analcime MOF topology; in other cases, when unsubstituted
imidazole was used, a chain-like structure was assembled.11 Very recently, Dalgarno et al.12
reported the effect of the substituent position on the assembly of metal-organic structures based
on xylyl bis-calix[4]arene ligands. They observed that while ortho- and meta-xylyl biscalix[4]arenes formed clusters with FeIII and LnIII metal centers, the para-xylyl bis-calix[4]arene
derivative yielded a trigonal antiprismatic metal-organic cage that showed good thermal stability
and partial N2 and H2 uptake.
The effect of the solvent in the design and/or functionalization of MOFs is also well known. In
this context, postsynthetic ligand exchange methodologies (PSLE), also referred as solventassisted ligand exchange (SALE), have shown high solvent-dependence during the linker
exchange process of the pre-synthetized metal-organic framework. Several authors have dwelt on
this subject.13,14 On the other hand, the effect produced by the solvent during the crystallization

3

processes involved in the MOFs syntheses is also important. It was shown that solvent polarity
effects modulate the conformation of the resulting polymeric structure, as reported by Chen et al.
15

in the production of “supramolecular isomers”.16 The differences in solvent polarity modulate

the supramolecular interactions that exist between the organic linkers and the solvent molecules,
which in turn endows the metal frameworks with different local coordination geometries.
Calix[4]pyrroles are a class of macrocyclic compounds formed by four pyrrole units linked
through a doubly substituted sp3 hybridized carbon, the so-called meso-carbon.17,18 The structural
versatility and functionality displayed by these compounds makes them attractive candidates to
function as molecular receptors. These molecules show high affinity towards anions and neutral
molecules having electron-rich groups. Calix[4]arenes are structurally related to calix[4]pyrroles.
The former have also played a key role in the development of molecular recognition strategies.
Several studies described the construction of MOFs containing calix[4]arenes as linker units.19 The
first study, reported by Zhang et al.,20 described two 3-D framework constructed from tetra-psulfonatocalix[4]arenes and rare earth cations (i.e. NdIII and EuIII). Subsequently, Dalgarno et al.21
described the preparation of p-carboxylatocalix[4]arene-based MOFs in which the switching of
the assembly process from 1D to 3D scaffolds was governed by steric effects of the ligands.
Thurston et al.22 proposed the synthesis of calix[4]arene containing MOFs in order to build
hierarchical porous structures. However, only theoretical simulations were performed due to lack
of success in the activation of the synthesized assemblies. In this context, Park et al.23 reported the
synthesis and subsequent activation of a [(Pb2@L)  2 DMF]n MOF, where L is a carboxylic acidfunctionalized calix[4]arene, which involves a single-crystal to single-crystal transformation upon
solvent removal.

4

Regarding calix[4]pyrrole ligands, to the best of our knowledge there is only one very recent
report dealing with the preparation of a MOF by combining praseodymium clusters with
conformationally flexible two-wall aryl-extended calix[4]pyrroles. The two-component assembly
resulted in a 3-D structure containing free struts in the 1D channels of the framework. The authors
took advantage of the high affinity of the fluoride anion for the pyrrole NHs of the calixpyrrole.
The binding of the anion induced a change in the conformation of the calix[4]pyrrole scaffold:
from 1,3 alternate to cone-conformation. This conformational change provoked the disassembly
of the MOF network and the consequent release of the entrapped molecules.24 The possibility that
the breakup of the praseodymium MOF might also be mediated or assisted by the formation of
praseodymium fluoride complexes cannot be ruled out.
Hence, owing to the experience of our group on the preparation of functionalized two- and fourwall aryl-extended calix[4]pyrroles and the potential applications of calix[4]pyrrole-based MOFs
for selective molecular recognition and sensing, we focused our efforts on the synthesis and study
of unprecedented MOFs based on four wall aryl-extended calix[4]pyrrole scaffolds. The capability
of metal ions or ion-cluster to direct multiple structures of coordination frameworks is well
established. We selected CuII salts as precursors of the secondary binding units (SBUs) in the
metal-organic structures, owing to their well-known properties to form different CuII-clusters by
coordination with carboxylate groups. Here, we report four novel metal-organic materials, MOF1,
CP2, and the discrete capsular dimers C2 and C3. On the one hand, the tetracarboxylic metasubstituted four wall aryl-extended calix[4]pyrrole L1 produced MOF1. On the other hand, the
para-substituted isomer, L2, afforded either a coordination polymer, CP2, or a discrete capsular
dimer C2 depending on the solvent used in the solvothermal synthesis, DMF or DEF respectively.
The use of the L3 ligand, also featuring para-substitution in the aryl-extended calix[4]pyrrole core

5

but with a longer alkyl spacer spanning the distance between the terminal carboxylic groups and
the meso-phenoxy substituents, exclusively produced a discrete capsular dimer C3. The CuIIcarboxylate clusters acting as secondary binding units (SBUs) in CP2, C2 and C3 had CuII2(O2CR)4 paddle wheel structure. Conversely, the MOF1 featured a more complex metal cluster as
SBU, which comprised four CuII centers bridged by two 3-OH ligands giving rise to two-edge
sharing CuII3(OH) triangles. The CuII ions were 3- and 4- coordinate. Our results demonstrate the
viability of building metal-organic structures based on tetraaryl-extended calix[4]pyrroles.
Additionally, we obtained a variety of topologies in their crystal structures, which are controlled
by the position of the substituent in the meso-aryl rings of the organic linkers, the substituent’s
length, and, in one example, the size of the dialkylformamide used as solvent in the synthesis. Our
findings also showcase the versatility and flexibility featured by calix[4]pyrrole-based metalorganic networks. Indeed, the reported materials demonstrate, for the first time, how significant
changes in topology and morphology of metal-networks can be achieved by employing different
isomeric calix[4]pyrrole ligands as building block.

Scheme 1. Preparation of the different calix[4]pyrrole-based metal-organic assemblies.

6

2. RESULTS
2.1. Synthesis. We explored the formation of metal-organic compounds using three different
meso-tetraaryl meso-tetramethyl calix[4]pyrrole tetracarboxylic acid derivatives: L1: a tetra-calix[4]pyrrole isomer bearing a meta-substituted aryl group geminal to the methyl group in each
one of the four meso-carbons. The meta-substituent is a 2-oxy-acetic acid; L2, a regio-isomer of
L1 with the 2-oxy-acetic acid substituent attached to the para- position of the meso-aryl groups;
and L3, containing the basic tetra- meso-tetraaryl meso-tetramethyl calix[4]pyrrole structure but
featuring a 6-oxy-n-hexanoicacid as substituent in the para-position of the meso-aryl units.
Solvothermal reaction of the three different calix[4]pyrrole derivatives, L1, L2 and L3, was
attempted with metal salts of CuII. Successful reactions yielding unique crystalline products
(Scheme 1) were found to follow an analogous procedure: heating the mixture of the ligand and
the CuII salt in a Teflon-lined autoclave at 100 oC for 24 h and subsequent cooling at room
temperature provided a blue crystalline powder that was isolated by filtration, washed with
dimethyl formamide (DMF) and ethanol and dried at 80 oC for 24 h. We found that the crystallinity

7

of the material increased if a layer of water was placed on the bottom of the reactor and was
carefully covered with the DMF solution containing both the salt and the calix[4]pyrrole ligand
prior to solvothermal treatment. Using this methodology, single crystals suitable for X-Ray
diffraction analysis were obtained, most likely due to the slow diffusion of the DMF into the water
phase. Control experiments carried out in the absence of water or in a homogeneous DMF/H2O
mixture did not afford crystalline materials. The single crystals obtained in layered reaction
conditions correspond to an extended coordination network (MOF1) for L1, and the cage-like
coordination compounds using L2 and L3 (C2 and C3, respectively, vide infra). By replacing
DMF by diethylformamide (DEF), a polymeric structure CP2 was also obtained starting from L2.
Taken together, these results demonstrate the importance of the solvent in the synthesis of these
metal-organic materials. Remarkably, for the L3 ligand the change of solvent did not favor the
formation of any crystalline metal-organic material.
2.2. Crystal structures.
2.2.1 MOF1 and CP2. The crystal structure of the MOF1 is formed by extended coordination
layers on the ab axis, yielding a 2D-like material.25 The layers are built from [Cu4OH2] moieties,
where the CuII atoms form a rhomb, with the two hydroxy groups located at the center of each
triangle (Figure 1a) defined by the sides and the short diagonal. These hydroxy units are not
coplanar, and appear one above and one below the plane defined by the four CuII atoms. The two
CuII atoms at the short diagonal (Cu–Cu distance = 2.868 Å) are connected through this double
hydroxy bridge, and complete their pseudo-square planar coordination geometry with two
carboxylate groups, from two different calix units. These carboxylate groups show a syn-syn
geometry and bridge the CuII atoms in the short diagonal to the CuII atoms in the long diagonal,
displaying a syn-syn coordination mode (Cu–Cu distance = 3.296 Å). The CuII atoms in the long

8

diagonal complete their distorted trigonal bipyramid geometry with one oxygen atom of a terminal
carboxylate group from an additional calix ligand and another oxygen atom of one carboxylic
group of a fourth calix linker. In this way, each [Cu4OH2] is connected to eight different calix
units, four above and four below the plane defined by the short diagonal copper centers (Figure
1b). Each calix unit binds four tetracopper clusters, forming a pseudotetragonal packing in the
layers (Figure 1c). The center of symmetry at the center of the [Cu4] rhomb imposes an eclipsed
conformation for the ligands at both sides of the mean plane, and between all layers (Figure 1d).
The 2D layers are packed though DMF molecules, with one methyl group facing the lower rim of
the calix[4]pyrroles. The four pyrrole NH groups point towards the calix cavity, strongly binding
a solvent DMF molecule and establishing four simultaneous hydrogen bonds with the oxygen
atom. Disordered solvent molecules occupy the empty voids in the crystal.

a)

b)

9

c)

d)

Figure 1. Representation of the crystal structure of MOF1: a) Detail of the Cu4 tetramer with its
coordination geometry. b) Projection on the ac plane of a CuII4 tetramer with all the coordinating
ligands. c) Projection on the ac plane of the layered structure. d) Projection on the ab plane.
Table 1. Crystallographic data and main refinement parameters.
MOF1
Formula

CP2

C2

C3

C128.5H138Cu4N13.5O33.5

C66H76Cu2N6O

C62H62Cu2N6O15.5

C159H195Cu4N13O2

17

9

Mw

2661.68

1352.40

1266.25

3006.43

T (K)

90(2)

100(2)

100(2)

100(2)

Crystal
system

Triclinic

Orthorhombic

Orthorhombic

Triclinic

Space
group

̅
P1

Pbcm

Pmna

̅
P1

a (Å)

16.241(2)

12.3540(6)

21.9220(6)

17.8930(4)

b (Å)

16.868(3)

23.2173(12)

23.5108(8)

20.4413(5)

c (Å)

26.290(2)

21.5694(8)

23.6094(7)

21.0342(5)

 (º)

89.457(10)

90

90

79.281(2)

 (º)

72.01(2)

90

90

84.283(2)

 (º)

88.243(14)

90

90

75.326(2)

10

V (Å3)

6847.0(17)

6186.7(5)

12168.4(6)

7301.4(3)

Z

2

4

8

2

calc

1.291

1.452

1.382

1.367

 (mm–1) 0.690

0.765

0.771

0.654

F (000)

2775

2832

5264

3176

range
(º)

4.056 to 42.236

4.184 to 51.364 4.284 to 50.698

3.652 to 61.290

Index
range

–16 ≤ h ≤ 16

–14 ≤ h ≤ 14

–-25 ≤ h ≤ 26

–25 ≤ h ≤ 25

–17 ≤ k ≤ 17

–25 ≤ k ≤ 28

–28 ≤ k ≤ 28

–29 ≤ k ≤ 29

–26 ≤ l ≤ 26

–26 ≤ l ≤ 26

–27 ≤ l ≤ 28

–27 ≤ l ≤ 27

Reflectio 26238
nstotal

33340

105910

171725

Reflectio 26238
nsunique

6027

11764

40891

Rint

0.0484

0.00528

0.0953

paramete 2485/2769
rs/restrai
ns

472/123

1649/2390

1963/364

Goodnes
s-of-fit
(F2)

1.662

1.441

1.012

R [I > 0.1012
2(I)]

0.0723

0.1788

0.0673

wR2 [I > 0.2343
2(I)]

0.2279

0.3584

0.1374

Rtotal

0.2213

0.0989

0.2287

0.1639

wR2total

0.2630

0.2397

0.3822

0.1716

diff.
peak/hol
e

1.280 / –0.686

1.235 / –1.127

3.269 / –1.308

0.652 / –0.716

(g/cm3)

–*

1.071

11

(e Å–3)
* Multidomain. Refined as a 3

The crystal structure of CP2 is formed by zig-zag 1D chains along the c axis, built from the
repeating unit [Cu2(calix)2] (Figure 2a). The copper atoms form the typical {Cu2(carboxylate)4}
dimer (Cu–Cu distance = 2.66 Å), with the four bridging carboxylates belonging to two different
L2 molecules. These neutral chains form a pseudo-tetragonal packing, with all Cu–Cu axis along
the c direction, and with orthogonal orientation between adjacent chains (Figure 2b). Solvent
molecules, in this case DEF, occupy the axial positions on the Cu dimers. The four pyrrole rings
point their NHs towards the calix cavity in a convergent manner, strongly binding a solvent
molecule (DEF) by establishing four simultaneous hydrogen bonds with the oxygen atom of the
latter.

a)

b)

Figure 2. Representation of the crystal structure of CP2: a) View of the chain structure along the
c axis; b) Projection on the ab plane, highlighting the two orientations of the chains in red and
yellow.

12

2.2.1.1 Effect of the solvent on the assembly of L2 induced by coordination to CuII. As shown
above, the coordination networks in CP2 display the calix[4]pyrrole ligands, L2, in the cone
conformation and with one DEF molecule deep included. Four-wall aryl-extended
calix[4]pyrroles, which are structurally closely related to L2 are known to adopt preferentially an
alternate conformation in solutions of non-polar chlorinated solvents, in which only two NHs of
the pyrrole rings converge. In this conformation, the receptor does not have well-defined aromatic
cavities. Conversely, the same compounds in solution of polar solvents with suitable sizes and
hydrogen-bond acceptor moieties i.e. DMF, acetone or CH3CN adopt the cone conformation, in
which all the NH groups converge into the deeper aromatic cavity defined by this conformation
and establish four simultaneous hydrogen bonds with the oxygen atom of the included solvent
molecule. Single crystals of aryl-extended calix[4]pyrroles grown from polar solvents also show
the receptor in the cone conformation with one included solvent molecule.26
In the case of CP2, the solid-state structure nicely shows that the included DEF molecule in L2
forms four simultaneous hydrogen bonds between its oxygen atom and the four NHs of the pyrrole
rings of the calix core (Figure 3a). We measured an average distance of 3.02 Å for the N···O
interactions supporting the formation of strong hydrogen bonds. Moreover, the inclusion complex
DEF@L2 is stabilized by multiple additional CH- interactions i.e. d(OHC···Arcentroid) = 3.60 Å
and (EtNCH3H2C···Arcentroid) = 3.54 Å. The calix[4]pyrrole L2 as component of the CP2 adopts
the cone conformation and defines a closed polar cavity (Figure 3b). The cavity is closed on the
bottom by the calix[4]pyrrole core and on the top by one molecule of DEF disordered in two
positions, and two water molecules axially coordinated to each one of the two inwardly oriented
Cu atoms of the paddle wheels stabilizing the capsular dimer. Closing the cavity of L2, there is
also a bridging hydrogen bonded water molecule and a pyrrole ring of an adjacent L2 unit. The

13

volume of the cavity was computed to be 167 Å3 (purple surface in Figure 3b) using the SwissPdbViewer 4.1.0 software. In turn, the volume of DEF calculated with the same software is 109
Å3, which allows the calculation of a packing coefficient value (PC) of 65% for the encapsulation
complex of DEF@L2 in the solid state. The calculated PC value is slightly larger than the 55%
value described by Rebek et al. as ideal value for encapsulation complexes in solution and in nonpolar cavities.27 We have observed PC values as large as 75% for encapsulation complexes of Noxides in calix[4]pyrrole-based capsules.28 The existence of strong polar interactions between the
cargo and the container allows PC values larger than the ideal 55% in the resulting encapsulation
complexes.

Figure 3. a) X-ray structure of the inclusion complex DEF@L2 forming the 1D coordination
polymer CP2 through the assembly of two divergent CuII2(O2CR)4 paddle wheels as SBUs. For

14

clarity, the DEF disorder and the non-polar hydrogen atoms of L2 were removed. L2 is shown in
ball-and-stick representation and the included DEF molecule as CPK model; b) Selected section
of the crystal-packing of CP2 that was used to calculate the volume of the polar cavity of L2
including the DEF molecule.
Interestingly, all our attempts to assemble L2 in a metal-organic material with CuII but using
DMF as solvent produced a crystalline solid with completely different morphology. The X-ray
solution of the diffracted data obtained from sub-optimal single crystals displayed the presence of
metal-mediated discrete dimeric cages of L2, C2 (Figure 4). The cage assembly, is neutral and its
covalent structure is based on two square-four connected units CuII2(O2CR)4, a.k.a as CuII paddle
wheels, that bridge two L2 tetra-carboxylate ligands. The capsular container, C2, presents two
polar hemispheres with one included DMF molecule. The two hemispheres are separated by the
two inwardly directed water molecules axially coordinated to the CuII2(O2CR)4 units. Each one of
the two internal water molecules is hydrogen bonded to a DMF molecule closing the polar aromatic
cavities of the two capsular hemispheres. The computed volumes for the capsular hemispheres of
C2 were 136 Å3 (blue) and 134 Å3 (green), respectively. (Figure 4)

15

Figure 4. Solid-state structure of the dimeric capsular assembly DMF2C2. The included DMF
molecules were removed in order to calculate the volumes of the two separate polar cavities of the
capsule.
The encapsulation of one DMF molecule (74 Å3) in each of the two hemispheres provides ideal
PC values of 55% for the resulting encapsulation complex. The putative dimeric capsular assembly
encapsulating DEF, DEF2@C2, would have PC values c.a. to 81% suggesting that the assembly
should not be thermodynamically favored based on the cargo/container volume ratio or
complementarity. For the same token, the putative linear coordination polymer, DMFCP2, which
would result from the inclusion of DMF in L2 instead of DEF, would result in a PC value of 44%.
This estimated value is significantly below the ideal range of 55-70% that is expected for
thermodynamically stable encapsulation complexes. In short, we surmise that the observed
switching in morphology of the metal-assembly of L2 mediated by square connected CuII2(O2CR)4
units, from a zig-zag coordination polymer DEFCP2 to a discrete metallo-capsular assembly

16

DMF2CuC2, is mainly due to volume complementarity. The polar cavity of L2 defined in the
corresponding metallo-assembled solids adapts to the volume of the included solvent molecule by
changing the ligand’s coordination mode.
2.2.2 C2 and C3 dimeric metallocages.
Remarkably, C2 shows an analogous molecular structure to C3, which was assembled under
identical conditions using ligand L3 (Figures 5 and 6). Both solids are composed of discrete neutral
molecules assembled by dimerization of the ligands, L2 or L3, through two CuII2-(O2CR)4 paddle
wheels as SBUs. In C2, the capsules pack in orthogonal orientations with respect to their long axis,
with the calix[4]pyrrole cavities pointing to each other between adjacent units (Figure 5b,c). The
capsules leave empty spaces among them that are occupied by disordered solvent molecules
(DMF).

a)

b)

17

c)
Figure 5. Representation of the crystal structure of C2: a) Representation of the capsule; b)
projection on the [011] plane; c) Detail showing the respective orientation between two adjacent
capsules.
In contrast, the packing of C3 show all capsules packed parallel to each other with respect to
their long axis (Figure 6b,c). A second difference is the lack of an axial water molecule in the
corresponding interior position of the Cu2 dimers, with an inter-dimer Cu–Cu distance of just 3.281
Å. This very short contact precludes the presence of any ligand in these positions. We assign this
to the conformational flexibility of the ligand arms in L3, which allows a closer contact between
the two paddle-wheel units of the capsule and the establishment of Cu···O interactions. The shorter
arms of the L2 ligand do not allow for this close contact. On the outer axial positions of the Cu2 Ipaddle wheels in C3 there is a bound solvent (DMF) molecule. The parallel orientation of all C3
capsules is also due to their higher conformational flexibility. The phenyl rings define the shortest

18

contact between C3 capsules. A C–C distance below 4.2 Å suggests the presence of aromatic
interactions favoring this parallel orientation.

a)

b)

c)
Figure 6. Representation of the crystal structure of C3: a) Representation of the capsule; b)
Projection on the ac plane; c) Detail showing the respective orientation between two adjacent
capsules.

19

2.2.2.1 Effect of the solvent on the on the assembly of L3 induced by coordination with CuII.
The analysis of the volume complementary for the two hemispheres in the dimeric metallocapsular assembly (DMF)2@C3 (Figure 7) produced calculated PC values of 70%, which are near
the maximum of the ideal PC range determined for thermodynamically stable encapsulation
complexes of polar molecules.28 Unfortunately, using identical experimental conditions but
replacing DMF by DEF as solvent in the solvothermal assembly of L3 mediated by CuII did not
produce any crystalline metal-organic solid.

Figure 7. Solid-state structure of the capsular assembly DMF2CuC3. The included DMF
molecules were removed before the calculation of the volumes of its two polar cavities. Four DMF

20

molecules were strategically located in the periphery of the capsule to avoid the escape of the
sphere used to measure the cavity volumes. In the packing of the crystal, the cavities are closed by
adjacent packed capsules.
2.3 Thermal analysis. We studied the thermal stability of the metal-organic compounds MOF1
and CP2. Thermogravimetric analyses (TGA) were carried out in polycrystalline samples under
nitrogen atmosphere (Figure S3 and S4). Both compounds showed a small solvent loss (5-10%)
below 100 ºC, that we assigned to uncoordinated water molecules. No additional solvent loss was
observed until decomposition of the material started, above 200 ºC. According to these results, the
DMF and DEF molecules in the solid structure, particularly those included and hydrogen-bonded
to the NHs of the calix[4]pyrrole cores, were not removed under these conditions. Thus, although
the volume occupied by solvent molecules in the crystal structure is significant (close to 39% of
the crystal volume, 2640 Å3 per unit cell, for MOF1 and 24% of the crystal volume, 1470 Å3 per
unit cell, for CP2),29 the complex removal of the solvent molecules makes difficult to exploit their
intrinsic porosity. BET (Brunauer-Emmett Teller) analyses through nitrogen adsorption isotherms
indicate negligible porosity (Figures S5 and S6 in the SI). We attempted the removal of the solvent
molecules through other methods, including combination of heat and high vacuum, or substitution
with more volatile solvents. However, all our attempts failed to improve DMF/DEF removal.
2.4 Magnetic Characterization. Since the paramagnetic Cu2+ centers (S = 1/2) are connected
through carboxylate bridging ligands in the CuII2(O2CR)4 SBUs, we decided to study their
magnetic properties to determine the superexchange interactions present in these compounds.
Magnetically, CP2, C2 and C3 are identical, both contained the classic and well-studied
CuII2(O2CR)4 paddle wheel.30 In these complexes, a strong antiferromagnetic superexchange
stabilizes the singlet ground state.31 This is at the origin of a broad maximum observed in the

21

magnetic susceptibility vs temperature plot above 150 K (Figure 8). Since the two compounds
showed negligible differences in their magnetic behavior, we present here only the modeled
magnetic data for C2. The magnetic interaction between the two Cu centers (S1 = 1/2, S2 = 1/2)
can be simulated with a simple Hamiltonian;
𝛨 = – 2𝐽(𝑆1 𝑆2 )

(1)

that takes into account isotropic magnetic coupling between magnetic centers. A satisfactory fit of
the experimental data is achieved with the inclusion of a small paramagnetic contribution (para =
C/T) that accounts for the appearance of the Curie tail at very low temperatures (Figure 8). This is
typical of antiferromagnetically coupled compounds, and responds to the presence of crystal
defects or small impurities. The magnetic parameters obtained, g = 2.16 and J = –115 cm–1 (C =
0.04 emu mol–1, ≈ 5% paramagnetic impurity), are in very good agreement with previous data.26

Figure 8. Magnetic susceptibility (m) for C2 as a function of temperature (circles), and the fitting
to a singlet-triplet model.
The magnetic behavior of MOF1 is quite different. It contains a rhomb-like tetracopper unit, with
two different bridges. The Cu–Cu distances are 2.868 Å along the short diagonal, and 3.296 Å

22

along the sides. These are significantly longer than in the Cu2 paddle wheel (2.63 Å). In addition,
the CuO4 conformation is highly distorted from square planar, with one oxygen atom at over 1 Å
from the plane defined by the other three (in contrast, the four O atoms defining a perfect plane in
C2). The molecular bridges are also different, with a double 2-hydroxy square Cu2O2 (Cu–O–Cu
= 96º) in the short diagonal. The side bridges are formed by a double hydroxy/carboxylate
connecting Cu centers in two different geometries (square planar vs distorted trigonal bipyramid).
All these structural features suggested that weaker magnetic coupling was expected for this case,
due to the mismatch between magnetic orbitals. Additionally, the triangular arrangement around
the 3-hydroxy bridges may give rise to spin frustration.
The mT product at room temperature, 1.65 emu K mol–1, matches the expected spin-only value
(1.50 emu K mol–1). When lowering the temperature mT decreases slowly below 150 K, and faster
below 60 K, suggesting the presence of antiferromagnetic coupling between paramagnetic centers.
As a reasonable first approximation, the magnetic superexchange along the long diagonal can be
neglected. Using this approximation, the magnetic data can be modeled with the following
Hamiltonian:
𝛨 = – 2𝐽(𝑆1 𝑆2 + 𝑆1 𝑆3 + 𝑆2 𝑆4 + 𝑆3 𝑆4 )– 2𝐽′(𝑆2 𝑆3 )

(2)

where S1 = S2 = S3 = S4 = 1/2. The best fitting was found for g = 2.05, J = –5.6 cm–1, and J' = –
13.9 cm–1, with the short diagonal superexchange parameter being 2-3 times stronger than the
rhomb-side one. With these parameters, the high temperature data is perfectly reproduced, but the
model deviates at very low temperatures, where the experimental mT product is higher. We assign
this to the presence of paramagnetic impurities (see above), although we did not attempt to add
this contribution to the model to avoid overparametrization. The results are in good agreement
with the expected behavior from the molecular structure: all superexchange interactions are

23

antiferromagnetic although much weaker in this case. Indeed, no maximum is observed in the m
vs T plot (Figure 9).

Figure 9. Temperature dependence of the mT product for MOF1 (circles), and the fitting to a
rhomb-like model (eq. 2).

3. CONCLUDING REMARKS
In summary, we reported the preparation of four calix[4]pyrrole-based metal-organic
compounds. We show that in some of the reported structures, the coordination mode of the organic
likers and the structure of the CuII-carboxylate clusters acting as secondary building unit (SBU)
are dominated by two factors. These are the anchoring position of the carboxylic group attached
to the aryl moiety of the calix[4]pyrrole (i.e. meta- or para-position) and the solvent used in their
solvothermal synthesis. The former was expected, but the latter is much subtler and delicate. The
solvent effect is illustrated by the possibility of switching from discrete calix[4]pyrrole capsules

24

found in C2 to a 1D zig-zag coordination polymer constituting CP2, by just changing the solvent
used in their syntheses from DMF to DEF, respectively. This minor modification promotes a
completely different structure of the solid. We assign this surprising result to the solvent (DMF or
DEF) recognition by the host molecules used as organic linkers, which in turn pre-organizes the
calix[4]pyrrole core of the ligand in the cone conformation but with a substantial different filling
of its polar cavity volume. This modulates the preferred orientation of the aryl-substituted
carboxylic groups in L2, prior to self-assembly, during crystal growth and dictates the preferential
formation of dimers (C2) or the polymer (CP2) observed in the crystals. The obtained results
highlight the strong interplay that exists between molecular recognition, supramolecular
isomerization and solid-state structure. Conformational control of the linker substituents through
binding to solvent molecules or other guests prior to self-assembly can give rise to the expected
and/or unexpected solid-state architectures that may be further developed for sensing and/or
catalysis applications.

4. EXPERIMENTAL SECTION
Characterization. 1H-NMR and 13C-NMR spectra were recorded on a Bruker Avance 400 (400
MHz for 1H-NMR) or on a Bruker Avance 500 (500 MHz for 1H-NMR). Deuterated solvents were
purchased from Aldrich. Chemical shifts are given in ppm. Hight resolution mass spectra were
obtained on a Micro TOF II (Bruker Daltonics) ESI as ionization mode mass spectrometer. Powder
X-Ray (PXRD) data were collected with a D8 Advance Series 2θ/θ powder diffractometer at room
temperature BET: N2 adsorption/desorption isotherms were measured at 77 K using an AutosorbiQ
after the sample was first degassed under vacuum at 120 oC overnight. Surface areas were
determined by the BET method in an appropriate pressure range. Thermogravimetric analyses

25

were performed with a TGA/SDTA851 Mettler Toledo instrument at a heating rate of 5 °C min –1
between 30 and 800 °C under a constant flow of air (50 mL min–1). Magnetic susceptibility
measurements between 2 and 300 K were carried out in a Quantum Design MPMS-XL SQUID
magnetometer using a 1000 Oe field. Pascal’s constants were used to estimate the diamagnetic
corrections for the compounds.
Crystal structure determination. X-Ray diffraction data were collected on a Rigaku XtaLab
P200 Mo Kα rotating anode equipped with a Pilatus 200K hybrid pixel detector equipped with an
Oxford Cryostream 700 plus low temperature device. Data collection were carried out using
CrystalClear-SM Expert 2.1 b29.32 For the data reduction and absorption correction CrysAlisPro
1.171.38.37f was used.33 Structure solution was obtained with the VLD algorithm under
SIR2011.34 Structure refinement was performed with SHELXL-2014.35 For the structure MOF1
the program PLATON SQUEEZE36 was used for treating highly disordered solvent accessible
voids. CCDC 1519366 (C2), 1519369 (C3), 1519368 (MOF1) and 1519367 (CP2) contain the
supplementary crystallographic data for this paper.
Synthesis. All solvent and reagents used in the synthesis of the described compounds were
purchased from commercial sources and were used without further purification.
Calix[4]pyrrole 1 (see SI for the structure): This compound was prepared accordingly to the
procedure described by Ballester et al.37 with slightly modifications: Briefly In a round bottom
flask, 4’-hydroxyacetophenone (10 g, 73.4 mmol), pyrrole (5.35 mL, 77 mmol) and ethyl acetate
(735 mL) were added. HCl 36% (62.5 mL, 734 mmol) was slowly added. The mixture was stirred
under air overnight. To quench the reaction, 150 mL of water were added. Solid KOH was added
until pH 6 and then further neutralized with solid K2CO3 until pH 7. The organic phase was
separated, washed twice with water, dried over Na2SO4, filtered and concentrated under reduced

26

pressure. The residue was dissolved in Et2O (300 mL) and after filtration of insoluble impurities,
the filtrate was concentrate under reduced pressure. The crude was dissolved in dichloromethane
(250 mL), and the solution was left overnight at room temperature to obtain a white precipitate of
1 that was collected and recrystallized from acetonitrile to yield colorless crystals. Yield = 30 %.
1

H-NMR spectrum is in agreement with the previously described by Ballester et al.
Calix[4]pyrrole 2 (see SI for the structure): Compound 1 (3 g, 4.05 mmol) and K2CO3 (4.48 g,

32.4 mmol) were suspended in dry 150 mL DMF. Under stirring, methyl-2-bromoacetate (2.98
mL, 32.4 mmol) was dropwise added. The suspension was heated at 70 oC overnight. Next, the
solution was cooled down, filtered and the solvent evaporated. Distilled water was added to the
solution and the organic product was extracted with dichloromethane, dried over Na2SO4, filtered
and the solvent removed in vacuo. Finally, the brownish precipitated was triturated with
MeOH:hexane (95:5) which gave the compound 2 in 71% yield. 1H-NMR (400 MHz, d-acetone):
 8.92 (br s, 4H , pyrrole-NH), 7.22 (app t, J = 7.58 Hz, 4H, ArH), 6.67-6.53 (m, 12H, ArH), 6.01
(d, J = 2.56 Hz, 8H, -pyrrole), 4.67 (s, 8H, CH2), 3.76 (s, 12H, CH3), 1.89 (s, 12H, CH3). 13CNMR

(400

MHz,

d-acetone)

HRMS
(ESI) m/z 1051.4106 (M + Na)+ calcd for C60H60N4NaO12, found 1051.4118.
Calix[4]pyrrole L1: Calix[4]pyrrole 2, (3 mmol) was dispersed in 80 mL THF:H2O (1:1) and
stirred vigorously while lithium hydroxide (18 mmol) was added. After 1 h of stirring, the
tetrahydrofurane was evaporated under reduced pressure and the aqueous solution was acidified
with HCl 1N. A precipitate appears which was extracted with ethyl acetate (120 mL). The organic
layers were dried over Na2SO4 and evaporated under reduced pressure to yield the pure product.
Yield = 81 %. 1H-NMR (400 MHz, d-acetone):  8.91 (br s, 4H , pyrrole-NH), 7.25 (app t, J =

27

8.29 Hz, 4H, ArH), 6.69-6.55 (m, 12H, ArH), 6.01 (d, J = 2.55 Hz, 8H, -pyrrole), 4.65 (s, 8H,
CH2), 3.76 (s, 12H, CH3).
Calix[4]pyrrole 3 (see SI for the structure): This compound was prepared accordingly to the
procedure described by Sessler et al.38 with slightly modifications: Briefly, in a round bottom flask
equipped with a magnetic bar and under N2 atmosphere, 4’-hydroxyacetophenone (10 g, 72.7
mmol) and pyrrole (5.08 mL, 72.7 mmol) were dissolved in methanol (150 mL). Methanesulfonic
acid (1.3 mL, 20 mmol) was slowly added via syringe under stirring. The reaction was stirred at
room temperature overnight. To quench the reaction, 200 mL of water were added forming a black
tar that was recovered by filtration, dissolved in 100 mL of diethyl ether (helped by ultrasounds).
After filtering through a pleated filter paper, the mixture was concentrated under vacuum to yield
a red foam-like material. The desired isomer is recrystallized from glacial acetic acid as colorless
crystals in 45 % yield (13.4 g). 1H-NMR spectrum is in agreement with the previously described
by Sessler et al.
Calix[4]pyrrole 4 (see SI for the structure): Compound 3 (3 g, 4.05 mmol) and K2CO3 (4.48 g,
32.4 mmol) were suspended in dry 150 mL DMF. Under stirring, methyl-2-bromoacetate (2.98
mL, 32.4 mmol) was dropwise added. The suspension was heated at 70 oC overnight. Next, the
solution was cooled down, filtered and the solvent evaporated. Water was added to the crude and
the organic product was extracted with dichloromethane, dried over Na2SO4, filtered and the
solvent removed in vacuo. Finally, the brownish precipitated was triturated with MeOH which
gave the compound 4 in 86 % yield. 1H-NMR (400 MHz, d-acetone):  8.85 (br s, 4H , pyrroleNH), 6.92-6.80 (m, 16H, ArH), 6.00 (d, J = 2.58 Hz, 8H, -pyrrole), 4.70 (s, 8H, CH2), 3.78 (s,
12H,

CH3),

1.83

(s,

12H,

CH3).

13

C-NMR

(400

MHz,

d-acetone)

28

HRMS

(ESI)

m/z

1051.4106 (M + Na)+ calcd for C60H60N4NaO12, found 1051.4107.
Calix[4]pyrrole 5 (see SI for the structure): Compound 3 (0.3 g, 0.405 mmol) and K2CO3 (448
mg, 3.24 mmol) were suspended in dry 15 mL DMF. Under stirring, ethyl-6-bromohexanoate
(0.576 mL, 3.24 mmol) was added. The suspension was heated at 70 oC overnight. Next, the
solution was cooled down, filtered and the solvent evaporated. Water was added to the solution
and the organic product was extracted with dichloromethane, dried over Na2SO4, filtered and the
solvent removed in vacuo. Finally, the brownish precipitated was recrystallized from EtOH which
gave the compound 5 in 60 % yield. 1H-NMR (400 MHz, d-acetone):  8.81 (br s, 4H , pyrroleNH), 6.91-6.81 (m, 16H, ArH), 6.00 (d, J = 2.40 Hz, 8H, -pyrrole), 4.11 (q, J = 7.08 Hz, 8H,
CH2CH3), 3.95 (t, J = 6.00 Hz, 8H, CH2CH2), 2.35 (t, J = 7.27 Hz, 8H, CH2CH2), 1.86 (s, 12H,
CH3), 1.82-1.74 (m, 8H, CH2CH2CH2), 1.73-1.64 (m, 8H, CH2CH2CH2), 1.55-1.48 (m, 8H,
CH2CH2CH2), 1.25 (t, J = 7.43 Hz, 12H, CH2CH3),

13

C-NMR (400 MHz, d-acetone)


HRMS (ESI) m/z 1331.7235 (M + Na)+ calcd for C80H100N4NaO12, found 1331.7519
Calix[4]pyrrole L2: Calix[4]pyrrole 4, (3.35 mmol) was dispersed in 100 mL THF:H2O (1:1)
and stirred vigorously while lithium hydroxide (20.11 mmol) was added. After 1 h of stirring, the
tetrahydrofurane was evaporated under reduced pressure and the aqueous solution was acidified
with HCl 1N. A precipitate appears which was extracted with ethyl acetate (150 mL). The organic
layers were dried over Na2SO4 and evaporated under reduced pressure to yield the pure product.
Yield L2 = 90 %, 1H-NMR (400 MHz, d-acetone):  8.84 (br s, 4H , pyrrole-NH), 6.92-6.81 (m,
16H, ArH), 6.00 (d, J = 2.66 Hz, 8H, -pyrrole), 4.68 (s, 8H, CH2), 1.83 (s, 12H, CH3).

29

Calix[4]pyrrole L3 was obtained by using the same procedure than L2 but starting from
calix[4]pyrrole 5. Yield L3 = 91 % 1H-NMR (400 MHz, d-acetone):  8.77 (br s, 4H , pyrroleNH), 6.91-6.80 (m, 16H, ArH), 5.97 (d, J = 2.57 Hz, 8H, -pyrrole), 3.94 (t, J = 6.35 Hz, 8H,
CH2CH2), 2.35 (t, J = 7.21 Hz, 8H, CH2CH2), 1.85 (s, 12H, CH3), 1.82-1.74 (m, 8H, CH2CH2CH2),
1.73-1.62 (m, 8H, CH2CH2CH2), 1.57-1.48 (m, 8H, CH2CH2CH2).
MOF-1: 400 mg of L1 (0.411 mmol) and 45 mg of Cu(NO3)2 x H2O (0.205 mmol, ratio
ligand:metal 2:1) were dissolved in 45 mL of DMF and carefully layered on 21 mL of deionized
H2O. The mixture is heated in a Teflon-lined autoclave at 100 oC for 24 h. The resulting blue
precipitate was filtered, washed with DMF and EtOH, and finally dried at 80 oC for 24 h. Yield =
63 mg
CP-2: 300 mg of L2 (0.307 mmol) and 37 mg of Cu(NO3)2 x H2O (0.154 mmol) were dissolved
in 35 mL of DMF and carefully layered on 17.5 mL of deionized H2O. The mixture is heated in a
Teflon-lined autoclave at 100 oC for 24 h. The resulting blue precipitate was filtered, washed with
DMF and EtOH, and finally dried at 80 oC for 24 h. Yield = 58 mg
Capsule C2: 800 mg of L1 (0.822 mmol) and 99 mg of Cu(NO3)2 x H2O (0.411 mmol, ratio
ligand:metal 2:1) were dissolved in 90 mL of DMF and carefully layered on 45 mL of deionized
H2O. The mixture is heated in a Teflon-lined autoclave at 100 oC for 24 h. The resulting blue
precipitate was filtered, washed with DMF and EtOH, and finally dried at 80 oC for 24 h. Yield =
155 mg
Capsule C3: 23.5 mg of L3 (0.021 mmol) and 2.5 mg of Cu(NO3)2 x H2O (0.001 mmol, ratio
ligand:metal 2:1) were dissolved in 2 mL of DMF and carefully layered on 1 mL of deionized
H2O. The mixture is heated in a Teflon-lined autoclave at 100 oC for 24 h. The resulting blue

30

precipitate was filtered, washed with DMF and EtOH, and finally dried at 80 oC for 24 h. Yield =
9 mg
ASSOCIATED CONTENT
Supporting Information. The following files are available free of charge. PXRD, TGA, BET,
1

H-NMRs, details on the disorder models for structures C2 and MOF1 and additional magnetic

data.
AUTHOR INFORMATION
Corresponding Author
jrgalan@iciq.es; pballester@iciq.es.
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
ACKNOWLEDGMENT
We are thankful for the financial support of the EU (ERC Stg Grant 279313, CHEMCOMP); the
Spanish Ministerio de Economía y Competitividad (MINECO) through projects CTQ2015-71287R, CTQ2014-56295-R and the Severo Ochoa Excellence Accreditation 2014-2018 SEV-20130319, FEDER funds (project CTQ2014-56295-R), and the Generalitat de Catalunya (2014-SGR797, and CERCA Programme).. JAS gratefully thanks the Marie Curie COFUND Action from the
European Commission for cofinancing his postdoctoral fellowship. We also thank one of the
reviewers for noticing us the topological type of MOF1. The corresponding information is included
in reference 25.

31

REFERENCES

32

1

Kitagawa, S.; Kitaura, R.; Noro, S.-i., Angew. Chem. Int. Ed. 2004, 43, 2334-2375.

2

Cheetham, A. K.; Rao, C. N. R.; Feller, R. K., Chem. Commun. 2006, 4780-4795.

3

Li, J.-R.; Kuppler, R. J.; Zhou, H.-C., Chem. Soc. Rev. 2009, 38, 1477-1504.

4

Kreno, L. E.; Leong, K.; Farha, O. K.; Allendorf, M.; Van Duyne, R. P.; Hupp, J. T., Chem.

Rev. 2012, 112, 1105-1125.
5

Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen, S. T.; Hupp, J. T., Chem. Soc. Rev.

2009, 38, 1450-1459.
6

Schneemann, A.; Bon, V.; Schwedler, I.; Senkovska, I.; Kaskel, S.; Fischer, R. A., Chem. Soc.

Rev. 2014, 43, 6062-6096.
7

Li, X.; Chen, X.; Jiang, F.; Chen, L.; Lu, S.; Chen, Q.; Wu, M.; Yuan, D.; Hong, M., Chem.

Commun. 2016, 52, 2277-2280.
8

Walton, I. M.; Cox, J. M.; Coppin, J. A.; Linderman, C. M.; Patel, D. G.; Benedict, J. B., Chem.

Commun. 2013, 49, 8012-8014.
9 Martin, A. D.; Easun, T. L.; Argent, S. P.; Lewis,

W; Blake, A.J.; Schroder, M.; CrystEngComm

2017. DOI: 10.1039/C6CE01965J
10

Huang, X.-C.; Lin, Y.-Y.; Zhang, J.-P.; Chen, X.-M., Angew. Chem. Int. Ed. 2006, 118, 1587-

1589.
11

Tian, Y.-Q.; Xu, H.-J.; Weng, L.-H.; Chen, Z.-X.; Zhao, D.-Y.; You, X.-Z., Eur. J. Inorg.

Chem. 2004, 2004, 1813-1816.

33

12

Coletta, M.; McLellan, R.; Murphy, P.; Leube, B. T.; Sanz, S.; Clowes, R.; Gagnon, K. J.;

Teat, S. J.; Cooper, A. I.; Paterson, M. J.; Brechin, E. K.; Dalgarno, S. J., Chem.--Eur. J. 2016, 22,
8791-8795.
13

Karagiaridi, O.; Bury, W.; Mondloch, J. E.; Hupp, J. T.; Farha, O. K., Angew. Chem. Int. Ed.

2014, 53, 4530-4540.
14

Kim, M.; Cahill, J. F.; Su, Y.; Prather, K. A.; Cohen, S. M., Chem. Sci. 2012, 3, 126-130.

15

Huang, X.-C.; Zhang, J.-P.; Lin, Y.-Y.; Chen, X.-M., Chem. Commun. 2005, 2232-2234.

16

Hennigar, T. L.; MacQuarrie, D. C.; Losier, P.; Rogers, R. D.; Zaworotko, M. J., Angew.

Chem. Int. Ed. 1997, 36, 972-973.
17

Kim, D. S.; Sessler, J. L., Chem. Soc. Rev. 2015, 44, 532-546.

18

Saha, I.; Lee, J. T.; Lee, C.-H., Eur. J. Org. Chem. 2015, 2015, 3859-3885.

19

Quinlan, E.; Matthews, S. E.; Gunnlaugsson, T., J. Org. Chem. 2007, 72, 7497-7503.

20

Liao, W.; Liu, C.; Wang, X.; Zhu, G.; Zhao, X.; Zhang, H., CrystEngComm 2009, 11, 2282-

2284.
21

Cholewa, P. P.; Beavers, C. M.; Teat, S. J.; Dalgarno, S. J., Cryst. Growth Des. 2013, 13,

5165-5168.
22

Bew, S. P.; Burrows, A. D.; Duren, T.; Mahon, M. F.; Moghadam, P. Z.; Sebestyen, V. M.;

Thurston, S., Chem. Commun. 2012, 48, 4824-4826.
23

Lee, E.; Kim, Y.; Heo, J.; Park, K.-M., Cryst. Growth Des. 2015, 15, 3556-3560.

34

24

Lee, J.; Waggoner, N. W.; Polanco, L.; You, G. R.; Lynch, V.; Kim, S. K.; Humphrey, S. M.;

Sessler, J. L., Chem. Commun. 2016, 52, 8514-8517.
25

The topological type of MOF1 is a 4,8L15 net (the 15th 4,8-c layer net by TOPOS) with

{420.8}{46}2 point symbol. See http://topospro.com/.
26

Gil-Ramírez, G.; Escudero-Adán, E. C.; Benet-Buchholz, J.; Ballester, P., Angew. Chem. Int.

Ed. 2008, 47, 4114-4118.
27

Mecozzi, S.; Rebek, J. J.; Chem.--Eur. J. 1998, 4, 1016-1022.

28

Galan, A.; Espelt, M.; Ballester, P.; Supramol. Chem. 2016, 28, 455-463.

29

A sphere of 1.2 Å of radius was calculated to be trapped in the porous.

30

De Loth, P.; Cassoux, P.; Daudey, J. P.; Malrieu, J. P., J. Am. Chem. Soc. 1981, 103, 4007-

4016.
31

Hay, P. J.; Thibeault, J. C.; Hoffmann, R., J. Am. Chem. Soc. 1975, 97, 4884-4899.

32

Rigaku, 2013, CrystalClear-SM Expert 2.1 b29.

33

Rigaku Oxford Diffraction, 2015, CrysAlisPro Software system, version 1.171.38.37f, Rigaku

Corporation, Oxford, UK.
34

Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.; Giacovazzo, C.;

Mallamo, M.; Mazzone, A.; Polidori, G.; Spagna, R. J. Appl. Cryst. 2012, 45, 357-361.
35

Sheldrick, G. M. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3-8.

36

Spek, A. L. Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 9-18.

35

37

Guinovart, T.; Hernández-Alonso, D.; Adriaenssens, L.; Blondeau, P.; Martínez-Belmonte,

M.; Rius, F. X.; Andrade, F. J.; Ballester, P., Angew. Chem. Int. Ed. 2016, 55, 2435-2440.
38

Anzenbacher, P.; Jursíková, K.; Lynch, V. M.; Gale, P. A.; Sessler, J. L., J. Am. Chem. Soc.

1999, 121, 11020-11021.

36

37

ToC: Tetraaryl-extended calix[4]pyrrole derivatives with carboxylic functional groups yield
capsules, 1D chains or 2D layered structures as determined by the reaction conditions with Cu2+
salts.

38