Three-dimensional six-connecting organic building blocks based on polychlorotriphenylmethyl units - Synthesis, self-assembly, and magnetic properties

Article abstract of DOI:10.1002/chem.200600447

The synthesis of a three-dimensional, six-connecting, organic building block based on a robust, rigid, and open-shell polychlorotriphenylmethyl (PTM) unit (radical 1) is reported, and its self-assembly properties are described in detail. The tendencies of

Full text of DOI:10.1002/chem.200600447

DOI: 10.1002/chem.200600447
Three-Dimensional Six-Connecting Organic Building Blocks Based on
Polychlorotriphenylmethyl Units—Synthesis, Self-Assembly,
and Magnetic Properties
Nans Roques,[a] Daniel Maspoch,[a] Klaus Wurst,[b] Daniel Ruiz-Molina,[a]
Concepció Rovira,[a] and Jaume Veciana*[a]
Abstract: The synthesis of a three-di-
mensional,six-connecting,organic
lar interactions at low temperatures,
whereas when the radicals are isolated
by THF molecules these interactions
are antiferromagnetic and very weak.
The role played by the carboxylic
groups not only in the self-assembly
properties but also in the transmission
of the magnetic interactions has been
illustrated by determination of the
crystal structure and measurement of
the magnetic properties of the corre-
sponding hexaester radical 6,in which
the close packing of molecular units
gives rise to weak antiferromagnetic in-
termolecular interactions. Attempts to
avoid solvation of the molecules in the
solid state and to increase the structur-
al and magnetic dimensionality were
pursued by recrystallization of both
compounds 1 and 4 from concentrated
nitric acid,affording two three-dimen-
sional (3D) robust hydrogen-bonded
structures. While the structure obtained
with compound 4 is characterized by
the presence of polar channels and
boxes containing water guest molecules
along the c axis,radical 1 was oxidized
to the corresponding fuchsone 10,
which presented a completely different
close-packed,guest-free structure.
building block based on a robust,rigid,
and open-shell polychlorotriphenyl-
methyl (PTM) unit (radical 1) is re-
ported,and its self-assembly properties
are described in detail. The tendencies
of this highly polar molecule and its
hydrogenated precursor,compound 4,
to form hydrogen bonds with oxygenat-
ed solvents ([1·THF6] and [4·THF6])
were reduced by replacing THF with
diethyl ether in the crystallization proc-
ess to yield two-dimensional (2D) hy-
drogen-bonded structures ([1·
A
Keywords: free
radicals
·
and [4·(Et2O)3]). The presence of direct
ACHTREUNG
hexacarboxylic acids
·
magnetic
hydrogen bonds between the radicals
in the latter phase of 1 gives rise to
very weak ferromagnetic intermolecu-
properties
·
self-assembly
·
structure dimensionality
Introduction
Control of molecular organization is an essential aspect of
the development of purely organic solids with remarkable
physical or chemical properties and structural characteris-
tics.[1] Whereas supramolecular chemistry has allowed access
to numerous examples of molecular organizations in one
and two dimensions,[2,3] those organized in three dimensions
are still challenging propositions.[4] The fairly limited
number of purely organic three-dimensional assemblies re-
ported so far may be explained by the difficulties involved
in synthesizing well-programmed molecular building
blocks,[5] with regard to the positions of the noncovalent in-
teractive sites,molecular shape,rigidity,and amphiphilici-
ty,[6,7] and also by the complexity inherent in obtaining the
required fine sophisticated self-assemblies through noncova-
lent interactions.[8]
[a] Dr. N. Roques,Dr. D. Maspoch,Dr. D. Ruiz-Molina,Prof. C. Rovira,
Prof. J. Veciana
Institut de Ci›nca de Materials de Barcelona (CSIC)
Campus Universitari,08193 Bellaterra,Barcelona,Catalonia (Spain)
Fax : (+34)93-580-5729
E-mail: vecianaj@icmab.es
[b] Dr. K. Wurst
Institut für Allgemeine Anorganische und Theoretische Chemie
Universitat Innsbruck,6020,Innrain 52a
Innsbruck (Austria)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author. It includes two
figures illustrating the supramolecular motifs involved in the self-as-
sembly of molecule 10.
9238
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

FULL PAPER
Such synthetic and structural considerations are even
more difficult to fulfill in the case of purely organic magnet-
ic materials based on the self-assembly of open-shell tec-
tons.[1d,9] Different families of open-shell tectons have been
used with this as a goal and it has,for instance,been shown
that the relatively small size and flatness of verdazyl radicals
enables them to p-stack,resulting in very strong magnetic
exchange interactions despite the radical centers being quite
far apart.[10] Other examples include the series of dithiadia-
zolyl and dithiazolyl radicals,which exhibit bulk magnetic
order and room temperature bistability.[11] Nevertheless,the
family of open-shell tectons most widely used up to now is
that of a-nitronyl nitroxide, a-imino nitroxide,or tert-butyl
nitroxide derivatives,because of their high stabilities and
the abilities of their nitroxide groups to act as hydrogen-
bond acceptors,so the self-assembly of such nitroxide-based
To circumvent such inconveniences,our group has initiat-
ed a strategy based on the synthesis and study of polychloro-
triphenylmethyl (PTM) radicals functionalized with carbox-
ylic groups. Our interest in these radicals is threefold. Firstly,
they exhibit high thermal and chemical stability. Secondly,
from a structural point of view,the rigidity and bulkiness of
this family of open-shell molecules[22] would be expected to
prevent close packing to form robust networks[23] and to
minimize undesired through-space magnetic interactions in
the solid state. Last but not least,among the numerous func-
tional groups that have been used to build self-assembled
networks,the choice of carboxylic acid groups is dictated by
the wide range of supramolecular synthons potentially of-
fered by their presence,[24] their well known capacity to form
strong and directional hydrogen bonds,and also their ability
to promote magnetic interactions between radical species.[20]
We have thus previously described the synthesis,self-assem-
bly,and magnetic properties of two PTM radicals substitut-
ed with two (PTMDC) and three (PTMTC) carboxylic acid
functions in para positions by this approach. Consistently
with their rigidity and their structural analogy with the iso-
phthalic and trimesic acids,the self-assembly of these two
radicals created noninterpenetrated open-framework struc-
tures consisting of 2D hydrogen-bonded layers.[23] Indeed,
while an example of a purely organic open-framework
(POROF-1) with dominant antiferromagnetic interactions
was obtained with the diacid radical,[25] the self-assembly of
the triacid radical produced a robust,porous,extended net-
work (POROF-2),giving rise to highly polar nanotubular
channels for the first time,together with ferromagnetic or-
dering at low temperatures.[26]
building blocks substituted with one or several hydrogen-
[12]
bond donors,such as phenol,
boronic acid,[13] imidazole,[14]
benzimidazole,[15] triazole,[16] uracil,[17] pyrazole,[18] phenyla-
cetylene,[19] or benzoic acid[20],has been extensively studied.
The resulting synthons enable both structural control and
transmission of magnetic interactions between the radical
À
molecules through strong (O H···O) or weak (CH3···O) hy-
drogen bonds.[12,21] This approach shows that the structural
dimensionality of the material—and hence the propagation
of the magnetic interactions through the supramolecular
structure—may to some extent be controlled through the
design of radical molecules bearing hydrogen-bond donor
and acceptor groups. However,the difficulties lie in deter-
mining the exact strength and the nature of the magnetic in-
teractions through the supramolecular pathways,as addi-
tional undesired intermolecular through-space interactions
are often present in the solid state and very frequently ob-
struct proper design of magnetic exchange pathways.[12e,17,18]
As a natural evolution due to the interest engendered by
the previously described work,we report here the synthesis,
supramolecular organization,and magnetic properties of
radical 1,which is substituted
with six carboxylic acid groups
at the meta positions of the
phenyl rings. In addition to the
synthetic challenge implied in
its synthesis and characteriza-
tion,[27,28] the changes both in
the number (six) and in the lo-
cation (meta positions) of the
carboxylic groups around the
PTM moiety would be expected
to afford new 3D assemblies,as
recently demonstrated in the
case of triphenylmethane-based
polyphenols.[29] Moreover,the
presence of six carboxylic
groups in radical 1 might be ex-
pected to enhance the number
of magnetic exchange pathways
between PTM radicals by
means of the hydrogen-bonding
network.[30]
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9239

J. Veciana et al.
Results and Discussion
work for the hexacarboxylic species. Indeed,whilst in the
cases of the di- and triacid PTM compounds the formation
of the anionic species had been complete within three days,
treatment of the hexa-acid 4 under the same conditions only
afforded the corresponding hexa-anion through deprotona-
tion of the acidic groups,as confirmed by UV/visible meas-
urements. Increasing of the reaction time and/or modifica-
tion of the reaction conditions (change of base,solvent,or
reaction temperature) either gave the same result or,when
reaction times were longer than one month,resulted in the
transformation of the hexacarboxylate derivative into un-
known compounds. It seems reasonable to think that the
negative shell produced by the presence of the six carboxy-
late groups located in meta positions impedes easy removal
of the hydrogen atom located in the central carbon atom of
the PTM unit.
To overcome these difficulties,compound 4 was first con-
verted into the hexaester 5 with diazomethane.[32] After-
wards,treatment of 5 with tetrabutylammonium hydroxide
yielded the intermediate carbanion,which presented the ab-
sorption band characteristic of an anionic PTM species at
520 nm.[33] This intermediate was then oxidized to the corre-
sponding radical 6 by addition of an excess of para-chloranil.
Synthesis: Radical 1 was prepared by the original six-step
procedure detailed in Scheme 1. A Friedel–Crafts reaction
performed with aluminum trichloride as a catalyst and with
a large excess of 1,3,5-trichlorobenzene over chloroform
(9:1) enabled the formation of compound 2,possessing a
PTM skeleton with six hydrogen atoms,all in meta positions.
Dichloromethyl groups were then introduced at the meta
positions by treatment of 2 with an excess of chloroform in
the presence of aluminum trichloride to afford compound 3
as a white solid.[31] Subsequently,the dichloromethyl groups
were converted to provide the corresponding hexacarboxylic
acid 4 through hydrolysis/oxidation by treatment of 3 with
20% oleum.[25,26] To the best of our knowledge,this is the
first time that conditions of this kind have been used to pro-
vide access to a hexacarboxylic acid molecule. The next step
was the formation of the radical species by the methodology
previously used to prepare the para-di- and -tricarboxylic
acid radicals PTMDC and PTMTC. This one-pot reaction
consists of two steps,starting with the formation of the cor-
responding polyanion and continuing with the oxidation of
the central carbanion. Surprisingly,this reaction did not
Scheme 1. i) AlCl3,CHCl 3, excess of 1,3,5-trichlorobenzene, 1108C (yield: 70%); ii) AlCl3,excess of CHCl 3,90 8C (70%). iii) Oleum (20%),150 8C
(75%); iv) Excess of NaOH,DMSO,RT; v) diazomethane,Et
2O,25 8C (80%); vi) n-Bu4NOH,THF,25 8C,vii) p-Chloranil,THF,25 8C (80%);
viii) H2SO4,100 8C (90%).
9240
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
The methyl ester groups were finally removed by heating
compound 6 in concentrated sulfuric acid to afford radical 1
in an overall yield of 21% after the six steps. Remarkable is
the solubility of molecules 1 and 4 in solvents,such as water,
ethers,and alcohols.
Molecular and crystalline structures of 1 and 4: Several at-
tempts to recrystallize radical 1 and its hydrogenated precur-
sor 4 were made with various solvent mixtures,concentra-
tions,and crystallization techniques. Most of these failed,
giving amorphous solids when mixtures involving nonoxy-
genated solvents,such as chloroform or dichloromethane,
were used to dissolve the hexa-acid derivatives. In contrast,
two different crystalline solvate phases were obtained by
slow diffusion of n-hexane over either THF or Et2O solu-
tions in the cases both of compound 1 and of compound 4.
It is important to emphasize that in spite of the change in
the hybridization state of the central methyl carbon atom
between radical 1 and its hydrogenated precursor 4,associ-
ated with the transformation from a planar to a tetrahedral
configuration,the resulting crystallographic phases of the
two compounds are very similar. Crystallographic data for
the different crystalline phases obtained are shown in
Table 1.
THF solvates: Crystals of 4 were obtained by layering a
THF solution of the hexa-acid derivative with n-hexane. Al-
though the crystals were found to lose crystallinity once
they had been removed from the mother liquor,the X-ray
structure was determined by handling the crystals with care
at low temperatures. A representative ORTEP drawing for
the PTM unit is shown in Figure 1. This solvate crystallizes
in the R-3 trigonal space group,with six molecules of hexa-
acid 4 and 36 molecules of THF in the unit cell.
Figure 1. Representative ORTEP views of the molecule common to
[4·THF6] (a) and [4·(Et2O)3] (b). Thermal ellipsoids are set at 20% prob-
N
ability and symmetry-related solvent molecules have been omitted for
clarity (representative atoms labeled).
The molecule possesses a threefold rotation axis that
passes through the central carbon atom (C3). As a conse-
Table 1. Crystallographic data and structure refinement of solvates of 1 and 4 and of compounds 6 and 10.
A
[4·
(Et2O)3]
[4·
N
]
6
N
[1·
U
10
formula
Mr
lattice type
space group
a [Š]
b [Š]
c [Š]
a [8]
C49H55Cl9O18
1250.98
trigonal
R-3
24.8944(6)
24.8944(5)
15.3111(3)
90
C37H37Cl9O15
1040.72
trigonal
R-3
15.6119(5)
15.6119(5)
33.7497(5)
90
C25H10Cl9O13.5
845.38
tetragonal
P-421c
20.7352(3)
20.7352(4)
15.2669(5)
90
C31H18Cl9O12
901.50
monoclinic
C2/c
23.0403(4)
13.6539(5)
12.7045(6)
90
C49H54Cl9O18
1249.97
trigonal
R-3
24.9133(6)
24.9133(8)
15.0874(8)
90
C37H36Cl9O15
1039.71
trigonal
R-3
15.6488(8)
15.6448(6)
33.8402(7)
90
C25H6Cl8O13
797.90
orthorhombic
Ccca
14.3641(3)
21.6196(3)
19.5798(5)
90
b [8]
90
120
8217.5(3)
6
1.517
213(2)
3207
2842
0.0517
0.0451
0.1133
0.1096
90
120
7123.8(3)
6
1.456
233(2)
2069
1777
0.0543
0.0464
0.1301
0.1232
90
90
105.588(2)
90
3849.7(2)
4
1.555
233(2)
3377
90
120
8109.7(5)
6
1.536
233(2)
2347
1761
0.0903
0.0678
0.2055
0.1884
90
120
7176.7(5)
6
1.443
233(2)
2211
1880
0.0679
0.0588
0.1703
0.1624
90
90
g [8]
V [Š3]
Z
6564.0(3)
8
1.711
233(2)
4563
4150
0.0535
0.0472
0.1203
0.1167
6080.4(2)
8
1.743
233(2)
2391
2009
0.0578
0.0478
0.1408
0.1326
1calcd [gcmÀ3
T [K]
]
unique reflections
reflections observed [I>2s(I)]
R1 (all data)
R1 [I>2s(I)]
wR2 (all data)
2730
0.0517
0.0377
0.0978
0.0920
wR2 [I>2s(I)]
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9241

J. Veciana et al.
quence,the dihedral angles between the mean planes of the
three polychlorinated aromatic rings and that of the three
bridgehead carbon atoms (the reference plane) are equal to
588,causing the molecule to adopt a three-bladed,propeller-
ties in their three-bladed,propellerlike conformations (Fig-
ure 2c–d),the alternating distances between C3 atoms of
different molecules being 6.86 and 8.45 Š. Additional weak
hydrogen bonds between THF molecules of [4·THF6] units
À
like conformation with a given P (plus) or M (minus) helici-
belonging to different columns (six by unit: C16 H···O5=
[34]
À
ty,as is usual for these compounds.
Because of the steric
2.7 Š; C15 H···O5=1298) enable their close packing,result-
hindrance from the chlorine atoms in the positions ortho to
each carboxylate group,the carboxylates are twisted by 80
(C9-C4-C1-O1) and 888 (C9-C8-C2-O3) with respect to the
ring to which they are bonded,with the two acidic hydrogen
atoms lying above and below the plane of the phenyl ring.
Each carboxylic acid group forms one strong hydrogen
bond with a THF molecule,with bond lengths of 1.75(2)
ing in a honeycomblike arrangement of the THF molecules
around the molecules of hexa-acid when the structure is
viewed along the c axis (Figure 2e).
Slow diffusion of n-hexane into a THF solution of radical
1 allowed small crystals,suitable for single-crystal X-ray
analysis,of the radical solvate [ 1·THF6] to be obtained. This
radical solvate also crystallizes in the R-3 trigonal space
group with six molecules of 1 and 36 molecules of THF in
the unit cell. The change in the hybridization of the central
carbon atom (sp2 in 1 versus sp3 in 4) is translated into a de-
crease in the value of the angles between the phenyl rings
and the reference plane of the molecule (558 versus 588),in
which the central carbon (C3) atom and the three aromatic
bridgehead carbon atoms (C6) are now fixed. This change in
the central carbon atom hybridization state is also accompa-
À
À
and 1.79(4) Š (O2 H···O5 and O4 H···O6) and bond angles
À
À
of 167(5) and 147(6)8 (O2 H···O5 and O4 H···O6),respec-
tively (Figure 2a–b),so [ 4·THF6] units are formed in the
crystal. Their self-assembly is assumed to take place through
the formation of weak C=O···HC
A
bonds in [4·THF6] units) with bond lengths of 2.67,2.70,and
À
À
À
2.72 Š (C12 H···O3,C15 H···O2,C14 H···O4) and bond
À
À
angles of 139,135,and 159 8 (C12 H···O4,C15 H···O2,
À
À
C14 H···O4). The resulting supramolecular columns are
made up of molecules of 4 with alternating P and M helici-
nied by a shortening of the C3 Cbridgehead bond length,
1.478(5) Š in 1 versus 1.548(3) Š in 4,and an increase in
À
À
the C C3 C angles from
114.8(2) in 4 to 119.96(3) in 1.
As in [4·THF6] solvate,the
three equivalent phenyl rings
adopt a propellerlike arrange-
ment around the central
carbon atom,the carboxylic
acid groups making angles of
878 with them (C9-C4-C1-O1ꢀ
C9-C8-C2-O3),with acid hy-
drogen atoms lying above and
below the phenyl ring to which
they are attached. The carbox-
ylic hydrogen atoms are in-
volved in hydrogen bonds with
highly disordered THF mole-
cules. Hydrogen atoms of the
THF molecules were not found
and not calculated,but the
analogy between 1 and 4 ena-
bles us to predict quite similar
supramolecular arrangements
of the two solvates. However,
small structural changes in 1
have given rise to alternating
intermolecular C3–C3 distan-
ces of 7.51 and 7.57 Š and to
slight changes in the cell pa-
rameters.
Diethyl ether solvates: Crystals
of a different solvate of 4 were
grown by diffusing n-hexane
into an Et2O solution of 4. As
Figure 2. Crystal structure of [4·THF6]. a) Hydrogen bonds (dashed lines) with THF molecules. b) View of the
supramolecular [4·THF6] cluster (THF molecules are represented as spheres for clarity). c) Packing of the clus-
ters along the c axis and d) top view of the resulting column. e) Self-assembly of the columns yield a honey-
comblike arrangement of the THF molecules around the molecules of 4.
9242
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
in the case of THF solvates,the crystals were unstable when
they were removed from the mother liquor,but the solvateꢁs
X-ray structure could be determined by handling the crystals
ecule of hexa-acid is hydrogen-bonded with two neighboring
hexa-acid units through one carboxylic group,with a bond
À
length of 1.78(3) Š (O4 H···O1) and a bond angle of
À
with care at low temperatures. This new solvate,[ 4·
A
167(6)8 (O4 H···O1). The remaining carboxylic group locat-
crystallizes in the R-3 trigonal space group with six mole-
cules of hexa-acid and eighteen molecules of Et2O in the
unit cell. The molecule possesses a threefold rotation axis
that passes through the central carbon atom (C3),and the
phenyl rings are twisted by angles of 558 with respect to the
reference plane of the molecule. Because of the presence of
the chlorine atoms at the ortho positions,the angles between
the carboxylic acid groups and the phenyl rings to which
they are bonded have values of 84 (C9-C4-C1-O1) and 838
(C9-C8-C2-O3). In this conformation,the acidic hydrogen
atoms of each of the carboxylic groups lie above and below
the mean plane of the phenyl ring to which they are
bonded.
ed in the same phenyl ring is involved in the formation of
one hydrogen bond with an Et2O molecule,with a bond
À
length of 1.72(3) Š (O2 H···O5) and a bond angle of
À
167(6)8 (O2 H···O5). This carboxylic acid–Et2O interaction
is reinforced by a supplementary weak hydrogen bond be-
tween the carbonyl group and one of the methyl groups of
À
À
the solvent molecule (O1···H C11=2.62 Š and O1···H
C11=1358).[35] The same kind of weak hydrogen bond is
also found between this Et2O molecule and a molecule of
À
hexa-acid presenting the same helicity in the layer (C11
À
H···O2=2.6 Š and C11 H···O2=1658).
As every molecule has six carboxylic acid groups—half of
them involved in hydrogen bonds with neighboring mole-
cules of hexa-acid and half with solvent molecules—each
molecule of 4 participates in the formation of three identical
hexameric units,resulting in polar windows that propagate
along the ab plane (Figure 3d). Three Et2O molecules are
localized on both sides of each polar window and contribute
to isolating one layer from each other,avoiding the forma-
tion of hydrogen bonds or Cl···Cl and/or Cl···O short con-
tacts between molecules of hexa-acid. Interdigitation of sol-
The most striking difference with the THF solvate is the
presence of several direct hydrogen bonds between mole-
cules of hexa-acid,resulting in the formation of 2D hydro-
gen-bonded corrugated layers running in the ab plane. The
6
main repetitive unit consists of an R6 (48) hydrogen-bonded
hexamer formed by molecules of hexa-acid with alternation
of their P and M helicities in their three-bladed,propeller-
like conformations (Figure 3a–c). In this hexamer,each mol-
Figure 3. Crystal structure of [4·
A
G
6
(Et2O molecules are represented as spheres for clarity). c) R6 (48) hexamer and d) extended hydrogen-bonded corrugated layer (clear spheres represent
ether molecules below the layer). e) ABC arrangement of the layers,isolated from one another by solvent molecules.
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9243

J. Veciana et al.
vent molecules belonging to neighboring layers yields an
ABC packing of the layers along the c axis (Figure 3e). This
packing is achieved through two types of weak hydrogen
bonds involving carboxylic groups of the PTM units and
The presence of the ester groups completely modifies the
solubility properties of radical 6,a fact that enabled the
growth of crystals suitable for X-ray crystal analysis from a
mixture of CH2Cl2 and n-hexane.[37] The hexaester radical
crystallizes in the C2/c monoclinic space group with four
molecules of 6 packed in the unit cell. The molecule pres-
ents a twofold rotation axis that passes through Cl5 and the
sp2 central carbon atom C10 (Figure 4). As a result,two of
À
CH2 or CH3 groups of the Et2O molecules (O1···H C10 and
À
O3···H C13 with bond lengths of 2.66 and 2.72 Š and bond
angles of 168 and 143.58,respectively). This produces an in-
terlayer shortest distance of 6.42 Š between C3 central
carbon atoms.
Slow diffusion of n-hexane into an Et2O solution of radi-
cal 1 afforded small crystals of the radical solvate [1·
(Et2O)3]. As already also observed in the case of the THF
solvates of 1 and 4,the self-assembled motifs of the diethyl
ether solvates are identical,with only small variations in the
bond lengths,bond angles,and,by extension,the cell pa-
rameters. The main values of dihedral angles,bond lengths,
and bond angles are reported in Table 2,in which the values
Table 2. Main bond lengths [Š] and bond and dihedral angles [8] found
in [4·
N
N
[4·
G
[1·(Et2O)3]
N
F
[a] [8]
53.9(1)
53.0(1)
À
C3 Cbridgehead [Š]
1.540(3)
115.5(2)
83.3(3)
1.483(4)
119.2(1)
83.1(6)
À
À
C C3 C [8]
C9-C8-C2-O3 [8]
Figure 4. ORTEP representation of radical 6. Thermal ellipsoids are set
at 20% probability (representative atoms labeled).
C9-C4-C1-O1 [8]
84.3(5)
85.1(6)
[b]
À
O4 H···O1 [Š]
1.78(3) [167(6)]
1.72(3) (167(6))
2.62 (135)
2.60 (165)
2.66 (168)
2.70 (143)
1.86(3) [163(5)]
1.82(5) (153(8))
2.64 (134)
2.68 (165)
2.71 (166)
2.74 (144)
[b]
À
O2 H···O5 [Š]
[b]
À
O1···H C11 [Š]
O2···H C11 [Š]
O1···H C10 [Š]
O3···H C13 [Š]
[b]
[b]
[b]
À
the aromatic rings are equivalent and the phenyl rings make
dihedral angles of 42 and 448 with the reference plane of the
molecule. Torsion angles between the methyl ester groups
and the phenyl rings to which they are bonded are 88 (C14-
C13-C3-O5),86 (C6-C5-C1-O1),and 72 8 (C6-C7-C2-O3) in
the case of the two aromatic rings that are identical. Four
À
À
[a] F corresponds to the dihedral angle made by the phenyl rings and the
reference plane of the molecule. [b] Angles are indicated in parentheses
[8].
À
À
weak hydrogen bonds (C15 H···O3=2.35 Š and C15
H···O3=1528) per molecule result in the formation of layers
along the ab plane that have the particularity of being made
up only of PTM radicals presenting P or M helicities
(Figure 5). Packing between P and M layers (AB packing)
along the c axis is assumed,through four Cl···Cl (Cl4···Cl1 =
3.45 Š) and twelve Cl···O contacts (Cl3···O2=3.00,
Cl4···O4=3.19,and Cl1···O6 =3.23 Š) per molecule. The
shortest distance between central carbon atoms is encoun-
tered between P and M molecules along the c axis,with a
value of 6.35 Š.
for [4·(Et2O)3] have also been included for purposes of com-
G
parison. Because of the sp2 hybridization of the central
carbon (C3),this atom and the three bridgehead carbon
atoms (C6) are fixed in the same plane. While the dihedral
angles between carboxylic groups and phenyl rings are quite
similar,the strongest hydrogen bonds observed in the 4 sol-
À
vate are weakened. Indeed,the bond length of the O2
H···O5 interaction increases by values of up to 0.1 Š,where-
as the bond angle decreases by 148.[36] Smaller variations are
observed in the weak hydrogen bonds involving the CH2 or
CH3 groups of the Et2O molecules and the carbonyl groups
of 1. The shortest distance between central carbon atoms is
also slightly increased in relation to that observed in the di-
ethyl ether solvate of 4 (6.89 versus 6.42 Š,respectively).
Magnetic properties: X-band ESR spectra of 10À4 m solu-
tions of radicals 1 and 6 in Et2O were obtained. The ESR
spectra of the two radicals at room temperature each show
one central main line surrounded by four weak satellite
lines (Figure 6). These satellite lines originate from the hy-
perfine coupling between the unpaired electron and the
magnetically active 13C nuclei in natural abundance at the a,
bridgehead,and ortho positions. Experimental spectra can
be simulated by using the isotropic hyperfine coupling con-
stants,as reported in Table 3. For purposes of comparison
Molecular and crystal structures of 6: To provide more in-
sight into the correspondence between hydrogen bonding
through carboxylic groups and the resulting magnetic prop-
erties in radical 1,the hexaester radical 6 was crystallized
and its structure was also analyzed in detail. Crystallograph-
ic data are given in Table 1.
9244
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
ylic groups,together with the
increase in their number,do
not produce any noticeable
change in the spin-density dis-
tribution around the PTM
core,which in both radicals is
mostly localized about the cen-
tral carbon atoms.[20d,33]
The static magnetic suscepti-
bilities of polycrystalline sam-
ples of both the THF and the
diethyl ether solvates of 1 and
of the hexaester radical 6 were
measured between 2 and 300 K
on
a SQUID susceptometer
with an external field of
1000 G. It is important to em-
phasize that the polycrystalline
samples of both solvates of 1
were measured in the presence
of a few drops of the mother
liquor to avoid fragmentation
and loss of crystallinity,which
may have disrupted the hydro-
gen-bonded
network
and
Figure 5. Crystal structure of hexaester 6. Cl···Cl and Cl···O contacts give rise to the AB packing of weakly hy-
drogen-bonded P and M hexaester layers.
changed the magnetic ex-
change interactions. At room
temperature,the cT product
values for all three samples are in good agreement with the
theoretical value of 0.375 emuKmolÀ1 expected for uncorre-
lated S=1/2 systems with g=2.0. With decreasing tempera-
ture the cT product values remained constant for all three
samples down to very low temperatures,whereupon differ-
ent behavior patterns were observed. The cT versus T plot
for [1·THF6],for instance,exhibits paramagnetic behavior in
the 10–200 K temperature range,but with the onset of very
weak antiferromagnetic interactions below 10 K. Similar re-
sults were obtained for the hexaester radical 6,which shows
paramagnetic behavior from 300 down to 15 K,below which
the cT value smoothly decreases,indicating the presence of
very weak antiferromagnetic interactions. Both curves were
fitted to the Curie–Weiss law,and Weiss constant ( q) values
of À0.30 and À0.50 K were obtained for radicals 1 and 6,re-
spectively. In contrast,the cT versus T plot for [1·(Et2O)3]
G
exhibits an increase in the cT product value at low tempera-
tures,consistent with the presence of intermolecular ferro-
Figure 6. Experimentally measured ESR spectrum (bottom) of a dilute
solution of 1 in Et2O at room temperature ([C]ꢀ10À4 m). Experimentally
measured spectrum (middle) in which the central line has been enlarged
in order to observe the satellite lines,together with simulated spectrum
(top). Simulation parameters are reported in Table 3.
Table 3. Hyperfine coupling constants obtained from simulations of ESR
spectra of diluted solutions of radicals 1 and 6. Values obtained for PTM
radicals substituted with one (PTMMC),two (PTMDC),and three
(PTMTC) carboxylic acid groups are given for comparison.
PTM radicals
aC-methyl [Gauss]
aC-ortho [Gauss]
aC-bridgehead [Gauss]
we also include our previously reported ESR data for car-
boxylic acid substitution at para positions in PTM radi-
cals.[20d,25,26] No significant differences between the meta-
and the para-substituted radicals are observed,which indi-
cates that the changes in the relative positions of the carbox-
PTMMC
PTMDC
PTMTC
1
6
29.8
29.8
29.8
29.3
29.3
12.7
12.5
12.6
13.0
13.0
10.3
10.3
10.2
10.5
10.5
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9245

J. Veciana et al.
magnetic interactions between different radical units. The q
value obtained from the best fit of experimentally measured
data to the Curie–Weiss law yielded a q value of 0.30 K.
Differences observed for the magnetic interactions in the
different solvates of radical 1 are directly related to the
structural differences observed in these solvates. The THF
solvate of 1 represents the formation of a supramolecular
cluster composed of one radical and six THF molecules. Ac-
cordingly,the presence of the THF molecules within the su-
pramolecular structure tends to disrupt the formation of a
PTM hydrogen-bonded network and consequently the trans-
mission of magnetic interactions mediated by carboxylic
groups between the radical units. As a consequence,the
presence of weak antiferromagnetic interactions in this sol-
vate is attributable to through-space magnetic interactions,
consistently with the lack of direct hydrogen bonds between
the PTMHC units and the distance separating the radical
centers (7.51 Š).
pounds is a general trend,as was confirmed after a thorough
crystallographic search on the Cambridge Crystallographic
Database and other different bibliographic sources. This
quest revealed the existence of only a limited number of
tetra- (or more) carboxylic acid-substituted scaffolds[7a–e,27,28]
that crystallize without disruption of the direct hydrogen
bonds between carboxylic acid groups by solvent molecules.
In the cases of compounds 7 and 8,for instance,the forma-
2
tion of R2 (8) dimers between carboxylic groups created in-
terpenetrated,hydrogen-bonded,diamondlike lattices. Crys-
tals of these molecules were grown by a gas-phase neutrali-
zation technique,that is,the diffusion of HCl vapor into sa-
turated solutions of the tetracarboxylates of 7 and 8.[7a,b]
With compound 9 a hydrogen-bonded layer was obtained
from a crystallization from concentrated nitric acid.[27a]
With compounds 1 and 4,HCl gas diffusion failed to pro-
vide crystals of sufficient quality to solve the X-ray crystal
structure. However,crystallization of compound 4 from con-
centrated nitric acid (65%) at 1008C afforded robust,color-
In the diethyl ether solvate of radical 1,the structural
analysis reveals PTM hydrogen-bonded layers made up of
6
the unusual R6 (48) repetitive hexameric units. The exis-
tence of direct hydrogen bonds between the PTMHC units
in this solvate gives rise to weak ferromagnetic interactions,
which are mediated and propagated through the repetitive
hydrogen-bonded hexameric units in the layers.
The role played here by the hydrogen-bonded carboxylic
groups in the formation of hydrogen-bonded layers and in
the transmission of magnetic interactions is highlighted by
comparing the structural and magnetic properties of [1·-
less prisms of solvate [4·(H2O)1.5]. This solvate crystallizes in
G
the P-421c tetragonal space group with eight molecules of
hexa-acid and twelve molecules of water in the unit cell.
The three aromatic rings here are different from those in
the THF and diethyl ether solvates of the same compound,
being twisted by angles of 48,48,and 59 8 with respect to the
reference plane of the molecule (Figure 7). Moreover,the
carboxylic acid groups are also twisted by angles of between
82 and 888,owing to the presence of the bulky chlorine
atoms at the ortho positions. Main distances and bond
(Et2O)3] with those of methyl ester 6. The latter compound
exhibits a close-packed,guest-free structure,most probably
due to the presence of methyl ester groups that do not have
the crucial ability to form strong directional intermolecular
interactions. The presence of short Cl···Cl contacts,which
are well known to yield antiferromagnetic interactions in
PTM derivatives,may be argued to explain the magnetic be-
havior found for this radical.
Attempts to obtain solvent-free structures—crystallizations
from nitric acid: The presence of weak but ferromagnetic in-
teractions propagating along the layers of the diethyl ether
solvate of radical 1 motivated us to increase the structural
dimensionality of the hydrogen-bonded network to a 3D hy-
drogen-bonded form,which may potentially exhibit magnet-
ic ordering. The main limitation to overcome in achieving
this target was the use of polar solvents,otherwise necessary
to dissolve the sample,but with the potential to disrupt the
hydrogen-bonded supramolecular network because of their
competing interactions for hydrogen bonds. Initially,to
avoid the use of solvents that might become involved in the
formation of the hydrogen bonds,different solvents,such as
diethyl ether or diisopropyl ether,and several crystallization
trials (such as slow diffusion or slow evaporation) were car-
ried out. All these trials failed,in most cases resulting in the
recovery of amorphous solids.
Figure 7. ORTEP representation of the PTM molecule of 4 found in [4·
(H2O)1.5]. Thermal ellipsoids are set at 20% probability and symmetry-
related solvent molecules have been omitted for clarity (representative
atoms labeled).
The difficulty involved in obtaining pure carboxylic acid
hydrogen-bonded networks with polycarboxylic acid com-
9246
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
angles for [4·
A
axis (Figure 8d). This kind of assembly generates polar
channels that are filled by isolated water molecules,separat-
ed from one another by 7.63 Š (Figure 8e). They are held in
the network by four hydrogen bonds with the carboxylic
gen bonds involving water molecules being described by
O···O distances.
As far as the supramolecular packing is concerned,four
hexa-acid molecules of alternating P and M helicities are as-
À
À
acid groups (two O8 H···O13 and two O13 H···O9). All
these motifs are reinforced by Cl···Cl and Cl···O short con-
tacts (see Table 4). Platon calculations revealed that the
structure of this solvate possesses an accessible volume close
to 8.5% when the water molecules are omitted,correspond-
Table 4. Main bond lengths [Š] and bond angles [8] encountered for hy-
drogen bonds and short contacts in [4·(H2O)1.5].
3
ing to a solvent-accessible volume of 568.1 Š with respect
À
O2 H···O14
to the volume of the total unit cell (6564.0 Š3).[38]
À
O4 H···O1
The similarities already found for the THF and diethyl
ether solvates of 1 and 4,together with the obtaining of a
porous material by crystallization of 4 from nitric acid,
prompted us to crystallize radical 1 from the same acid: if
successful,this could have led us to the obtaining of a pure
organic porous material with a 3D hydrogen-bonded struc-
ture similar to the water solvate of 4,with foreseeable po-
tential magnetic properties. However,and in spite of the
fact that the fully chlorinated PTM radical had already been
shown to be stable in concentrated nitric acid over several
days,[39] 1 was oxidized with nitric acid to give the corre-
sponding diamagnetic fuchsone 10 (see Scheme 2),probably
À
O6 H···O3
À
O8 H···O13
À
O10 H···O7
À
O12 H···O5
À
O13 H···O8
À
O13 H···O9
À
O14 H···O11
À
O14 H···O14
Cl1···Cl9
Cl8···Cl9
Cl5···O6
Cl6···O9
Cl9···O8
À
sociated through two O10
À
H···O7 and two O6 H···O3 hy-
4
drogen bonds to form R4 (48)
tetramers in the ab plane (Fig-
ure 8a). The resulting polar
nest is filled by two water mole-
À
cules that form two O2
À
H···O14 and two O14 H···O11
hydrogen bonds with PTM mol-
ecules. In addition,these two
water molecules are linked
À
through an O14 H···O14 hy-
drogen bond. Packing of a simi-
lar supramolecular unit takes
place along the c axis,through
À
four O12 H···O5 hydrogen
bonds (Figure 8b). The distance
between water molecules be-
longing to different dimers in
the resulting polar box is
3.38 Š. Packing of the boxes
À
takes place through O4 H···O1
hydrogen bonds,generating in-
finite supramolecular square
towers running along the c axis
(Figure 8c). Each molecule of
4,located in the corner of the
square,is involved in the for-
mation of similar motifs,result-
ing in a chessboardlike arrange-
ment of the towers when the
4
Figure 8. Crystal structure of [4·
A
box containing water resulting from the packing of two tetramers. c) Supramolecular tower running along the
c axis. d) Chessboardlike arrangement of the supramolecular towers (squares),producing polar channels (sur-
structure is viewed along the c rounded) filled by isolated water guest molecules (e). Water molecules are represented as spheres for clarity.
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9247

J. Veciana et al.
low in relation to those found
in 1 and its precursors (be-
tween 80 and 888),despite the
presence of the chlorine atoms
at ortho positions. A 1:1 disor-
der is observed for each car-
boxylic acid group,with appa-
À
rent equivalent C O bond
lengths (around 1.25 Š). These
bond lengths are slightly short-
ened in the carboxylic groups
on the quinone ring (around
Scheme 2.
1.19 Š),probably because of
better conjugation with the
as a result of the good solubility of this compound in nitric
acid at high temperatures,which contrasts with the low solu-
bility of the fully chlorinated PTM radical.[40]
Compound 10 was obtained as dark orange blocks that
crystallize in the orthorhombic space group with eight mole-
cules of fuchsone packed in the unit cell. Cell parameters
are reported in Table 1 and an ORTEP drawing for com-
pound 10 is shown in Figure 9. Fuchsone 10 exhibits a two-
six-membered ring arising from the lower dihedral angles
encountered for these groups. As a consequence of this dis-
order,the hydrogen atoms are also disordered,showing a
0.5 occupancy for those bounded to the oxygen atoms of
each carboxylic group. Because of such disorder,the hydro-
gen bonds are described below only in terms of the O···O
distances.
The three different types of carboxylic acid groups in-
volved in the three different supramolecular motives (I,II,
and III) are shown in Figures S1 and S2 in the Supporting
2
Information. The R2 (8) dimers involving the carboxylic
groups that contain C2 (pattern I,O3···O4 =2.64 Š) give
rise to the formation of helical chains running in the ac
plane (Figure 10).
Two different helicities are obtained,depending on the
conformation adopted by 10,giving rise to M and P chains
formed by (M)-10 and (P)-10 molecules,respectively. These
chains associate to give M- and P-layers in a peculiar zigzag
supramolecular pattern (II,Figure S1) that involves the car-
boxylic acid groups containing C2 and the ketone oxygen
atom (O7). These supramolecular polymeric chains involve
O7 and two O5 oxygen atoms as potential hydrogen bond
acceptors (O5···O5=2.64 Š and O7···O5=2.80 Š).
The 1:1 disorder observed for the two carboxylic acid
groups and the hydrogen atoms connected to them,together
with the two O···O distances encountered (which are short
enough to make possible hydrogen bonds),did not enable
us to determine the nature of the supramolecular motif.[42]
“Interpenetration” of the M and P layers (Figure 11) occurs
through Cl···O short contacts (two per molecule,Cl1···O3 =
3.05 Š) and p–p interactions between C=O groups and aro-
matic or quinone rings (O3···C14=3.15 Š and O7···C9=
3.04 Š). P/M interpenetrated layers are connected through
hydrogen-bonded polymeric chains running along the ab
plane involving both (M)- and (P)-10 molecules (III,Fig-
ure S2). The O···O distances (O1···O1=2.63 Š,O2···O2 =
2.70 Š),together with geometrical factors (C1-O1-O1-C1 =
1198 and C1-O2-O2-C1=1808),[43] enable this chain to be
described as one more example of the very rare syn-anti cat-
emer-type motif.[44]
Figure 9. ORTEP representation of the fuchsone 10 with occupancy of
0.5 for each hydrogen atom (positional disorder). Thermal ellipsoids are
set at 20% probability and symmetry-related solvent molecules have
been omitted for clarity (representative atoms labeled).
fold rotation axis that passes through the ketone oxygen
atom (O7) and the central sp2 carbon atom (C10). As a con-
sequence,the two aromatic rings substituted by three chlor-
ine atoms are equivalent. In comparison with the THF and
diethyl solvates of 1,in which the hybridization state of the
central carbon atom is also sp2,the presence of a double
bond between C10 and C11 is translated into a very low di-
hedral angle for the quinone ring,which is twisted by 28 8
with respect to the reference plane of the molecule,whereas
the two other aromatic rings present dihedral angles of 578
with the reference plane.[41] The carboxylic acid groups on
the trichlorinated rings are twisted by angles of 81 (C7-C6-
C1-O2) and 708 (C7-C8-C2-O3). The latter value is quite
Although diamagnetic and nonporous,the structure ob-
tained for compound 10 does provide an illustration of how
subtle modifications in the geometry of the PTM-based
9248
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
tecton could give rise to dra-
matic changes in the self-as-
sembly of the molecules and in
the resulting final structures.
Indeed, 10 possesses a rigid
core with a well-defined,three-
bladed,helix geometry sur-
rounded by six meta-carboxylic
acid groups,similar to that re-
ported for compounds 1 and 4.
The most important changes in
the tecton relate to the forma-
tion of the fuchsone ring
during the recrystallization,
which is mainly reflected in a
modification of the helicity
angle,the replacement of one
bulky para chlorine atom by a
less voluminous and potential
hydrogen-bond
acceptor
oxygen atom—of the ketone
group—and by the dihedral
angle modifications for the car-
boxylic groups close to this
oxygen atom. Noteworthy is
the presence of the unusual
moiety involving three hydro-
gen-bond acceptors and two
hydrogen-bond donors that
gives rise both to the forma-
tion of layers and to the forma-
tion of the unusual syn-anti
catemer-type motif that con-
nects the P/M interpenetrated
layers to create the 3D,sol-
vent-free,hydrogen-bonded
structure (Figure 12).
If the hydrogen bonds link-
ing the P and M molecules are
neglected in this motif,the
close-packed,guest-free struc-
ture may be described as the
result of the interpenetration
of
a nanoporous framework
made of P molecules of 10
with the corresponding nano-
porous framework made of M
molecules of 10. Even though
the structure is a nonporous
structure,as a result of better
shape complementarities than
Figure 10. Self-assembly of 10. Helical P and M chains,re-
2
sulting from R2 (8) dimer associations of (P)- and (M)-10,
are associated to form P and M layers. Chlorine–oxygen
short contacts enable the “interpenetration” of the two
layers in the ac plane.
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9249

J. Veciana et al.
yields by an original six-step
methodology. Study of the self-
assembly of this 3D,six-con-
necting organic building block
and of its precursor 4 revealed
that it is possible,through a
slight tuning of the crystalliza-
tion conditions,to control the
structural dimensionality and
to go from 0 to 2D structures,
in which the association of rad-
icals through direct hydrogen
bonds to form layers gives rise
to the presence of weak ferro-
magnetic intermolecular inter-
actions between radicals. At-
tempts to produce 3D hydro-
gen-bonded structures were
performed by recrystallizion of
4 and 1 in concentrated nitric
acid. While a 3D hydrogen-
bonded structure presenting
polar boxes and channels was
obtained with 4,a completely
different structure was ob-
tained with radical 1,as it was
oxidized in situ to the corre-
sponding fuchsone 10 during
the recrystallization process.
The structural analysis we per-
formed on this byproduct
clearly illustrates how subtle
changes in the molecular struc-
ture of a tecton may induce
dramatic changes both in the
synthons that are formed and
also in the resulting final struc-
ture. Crystallization attempts
Figure 11. Self-assembly of (P/M)-interpenetrated layers.
to find appropriate conditions
that will enable us to avoid the
in the corresponding (P)- and (M)-PTM units,it clearly
shows that the tectons based on the hexacarboxylic acid
PTM are promising candidates for the design of porous 3D
hydrogen-bonded networks. From a supramolecular and a
magnetic point of view,the challenge is now to find appro-
priate conditions to avoid the disruption of the network by
solvent molecules,interpenetration,or the chemical altera-
tion of the radical.
disruption of the network by solvent molecules,interpene-
tration,and the chemical alteration of the radical are cur-
rently in progress in our laboratory. Our efforts are also
being addressed towards the use of molecule 1 as an open-
shell organic ligand of paramagnetic metal ions,with the
idea of building robust and 3D magnetic metal-organic
frameworks.[45]
Experimental Section
Conclusion
X-ray measurement and structure determination: Data collection was
performed on a Nonius Kappa CCD with graphite-monochromatized
MoKa radiation (l=0.71073 Š) and a nominal crystal to area detector
distance of 36 mm. Intensities were integrated by using DENZO and
scaled with SCALEPACK. Several scans in the f and w directions were
made to increase the number of redundant reflections,which were aver-
aged in the refinement cycles. This procedure replaces an empirical ab-
In this article we have reported the detailed synthesis of the
first example of a PTM radical substituted with six carboxyl-
ic acid groups at meta positions,compound 1,which to the
best of our knowledge is the first example of a paramagnetic
hexacarboxylic acid. This molecule was obtained in good
9250
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
5.6:1 and the minor part is not included in the structural discussion. In 1
the first THF molecule is only slightly disordered and the disorder was
À
À
not solved,whereas the second one (C14 C17 O6) was completely disor-
dered in a 1:1 ratio with partially overlying carbon atoms. This THF mol-
ecule could only be refined with bond restraints and isotropic displace-
ment parameters. Finally,all the oxygen atoms of the carboxylic acid
groups of 10 are 1:1 positionally disordered,with the hydroxy oxygen
atoms overlaying exactly with the carbonyl oxygen atoms of the other
part,so hydrogen atoms are connected to each oxygen atom with occu-
pancies of 0.5. The hydrogen atoms at O2 and O5 were found in these
positions and were refined with bond restraints; the other hydrogen
atoms at O1,O3,O4,and O6 were also found,but were refined in calcu-
lated positions.
CCDC 602916–602920,276958,and 276959 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.
Materials and methods: Solvents were distilled before use. In particular,
Et2O was dried over sodium/benzophenone and distilled under argon.
All the commercial reagents were purchased from Aldrich and used as
received. Chromatography was performed on SDS silica gel (60F254). 1H
and 13C NMR spectra were recorded on a Bruker ARX 300 spectrometer
(1H NMR recorded at 300 MHz, 13C NMR recorded at 300 MHz),FTIR
spectra on a Perkin–Elmer Spectrum One spectrometer,and the UV/visi-
ble spectra on a VARIAN Cary 5 instrument. Direct current (dc) mag-
netic susceptibility measurements were carried out on a Quantum Design
MPMS SQUID susceptometer with a 55 kG magnet operating in the
range of 4–320 K. All measurements were collected in a field of 10 kG.
Background correction data were collected from magnetic susceptibility
measurements of the holder capsules. Diamagnetic corrections estimated
from the Pascal contents were applied to all data for determining the
molar paramagnetic susceptibilities of the compounds. ESR spectra were
recorded on a Bruker ESP-300E spectrometer operating in the X-band
(9.3 GHz). The signal-to-noise ratio was improved by accumulation of
scans with use of the F/F lock accessory to guarantee high-field reprodu-
cibility. Precautions were taken to avoid undesirable spectral line-broad-
ening,such as that arising from microwave power saturations and mag-
netic field over-modulation. To avoid dipolar broadening,the solutions
were carefully degassed three times by using vacuum cycles with pure
Ar. The g values were determined against the DPPH standard (g=
2.0030).
Figure 12. P and M nanoporous frameworks giving rise to the close-
packed,guest-free structure of 10.
sorption correction to a good approximation. The structures were solved
by direct methods (SHELXS86) and refined against F2 (SHELX97). The
R value of [1·THF6] is slightly higher,the result of low diffraction from
small crystal size (the volume was around 10 times smaller than that of
[4·THF6]).
Compound 2:[31a]
A
mixture of 1,3,5-trichlorobenzene (10.14 g,
55.9 mmol,9 equiv),anhydrous chloroform (0.5 mL,6.2 mmol,1 equiv),
and aluminum chloride (0.91 g,6.8 mmol,1.1 equiv) was heated at 80 8C
for 2.5 h in a glass pressure vessel. The mixture was then poured onto
ice/hydrochloric acid (1n) and extracted several times with chloroform.
The organic phase was dried over anhydrous sodium sulfate and the sol-
vent was removed. The excess of 1,3,5-trichlorobenzene was removed
over silica gel by using pure hexane as the eluent to give 2 as a white
powder in 70% yield (2.4 g,4.34 mmol). M.p. 246–248 8C (lit.[31a] m.p.
246–2488C); 1H NMR (CCl4/D2O) d=6.9 (s,1H),7.1 (d, J=2 Hz,3H;
Harom),7.2 ppm (d, J=2 Hz,3H; H arom); IR (KBr): n˜ =3120,3080,2905,
All non-hydrogen atoms were refined with anisotropic displacement pa-
rameters,except in disordered THF in [ 1·THF6]. Hydrogen atoms attach-
ed to carbon atoms were calculated and refined with isotropic displace-
ment parameters 1.2 or 1.5 times higher than the value of their carbon
atoms. Hydrogen atoms were found at most carboxylic acid groups and
were refined isotropically with bond restraints (dOH =0.85 Š),except in
the case of [4·(H2O)1.5],in which hydrogen atoms were not exactly local-
G
1570,1540,1430,1420,1365,1245,1185,1170,1140,1130,895,854,815,
ized in the same positions and were therefore refined in calculated posi-
tions,whilst hydrogen atoms of water molecules were omitted. Because
of the disorder of hydrogen atoms in 10 the treatment of their refinement
was mixed,with refinement in found and calculated positions (see
below).
À1
800,670,570,440 cm
.
Compound 3:[31b] A mixture of compound 2 (1.8 g,3.25 mmol,1 equiv),
anhydrous chloroform (140 mL),and aluminum chloride (0.66 g,
4.95 mmol,1.5 equiv) was heated at 80 8C for 24 h. The mixture was then
poured onto ice/hydrochloric acid (1n) and extracted several times with
chloroform. The organic phase was dried over anhydrous sodium sulfate
and the solvent was removed. The crude product was purified over silica
gel by using CHCl3/hexane 1:3 as the eluent to give 3 as a white powder
in 70% yield (2.39 g,2.27 mmol). M.p. 250 8C (decomp); 1H NMR
There are different types of disorder in the structures. In the THF and
water solvates of 4 the central carbon atoms are positionally disordered
in the opposite direction to the plane of the three attached phenyl
carbon atoms in ratios of 2:1 and 3:2,respectively. Further positional dis-
order occurs in the water molecule of [4·(H2O)1.5],where O14 and O14A
N
(CD2Cl2): d=6.9–7.2 (m,1H),7.6–7.8 ppm (m,6H); IR (KBr): n˜ =3044,
have a ratio of 3:1 and the position of O14A in the structure is not dis-
cussed. The THF molecules in the THF solvates of 1 and 4 are also posi-
À1
1532,1373,1274,1223,1043,1007,788,721,685,582,437 cm
.
À
À
tionally disordered. In 4 the first THF molecule (C10 C13 O5) is slightly
disordered,and only the oxygen atom was refined in two positions,
Compound 4: Compound 3 (500 mg,0.47 mmol) was mixed with oleum
(20%) and the mixture was heated at 1508C for 24 h. The final solution
was cooled and poured onto cracked ice,the aqueous phase was saturat-
ed with brine and extracted with Et2O,and the organic phase was con-
À
À
whereas the second THF molecule (C14 C17 O6) was completely disor-
dered into two positions. The ratio of the two THF molecules is around
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9251

J. Veciana et al.
centrated and extracted with aqueous sodium carbonate. The resulting
aqueous phase was acidified,saturated with brine,and extracted several
times with Et2O. The organic phase was dried over anhydrous sodium
sulfate and the solvent was removed. The crude product was dissolved in
Et2O (5 mL) and precipitated several times in hexane to give 4 as a white
powder in 75% yield (280 mg,0.36 mmol). 1H NMR ([D6]DMSO): d=
[1] a) G. R. Desiraju, Crystal Engineering: The Design of Organic
Solids,Elsevier,Amsterdam, 1989; b) P. Langley,J. Hulliger, Chem.
Soc. Rev. 1999, 28,279–291; c) M. Hollingsworth, Science 2002, 295,
2410–2413; L. Perez-Garcia,D. Amabilino, Chem. Soc. Rev. 2002,
31,342–356; d) J. Veciana,H. Iwamura, MRS Bull. 2000, 25,41–51;
e) S. Yagai,T. Karatsu,A. Kitamura, Chem. Eur. J. 2005, 11,4054–
4063.
5.5 (brs,6H),6.8 ppm (s,1H);
13C NMR ([D6]DMSO): d=55.3,130.8,
136.6,137.4,138.6,140.3,141.5,168.9 ppm; IR (KBr):
n˜ =3500–2500,
[2] a) J. MacDonald,G. Whitesides, Chem. Rev. 1994, 94,2383–2420;
b) D. Lawrence,T. Jiang,M. Levett, Chem. Rev. 1995, 95,2229–
2260.
1710,1540,1355,1299,1230,1203,1119,1036,968,816,731,605,
471 cmÀ1; elemental analysis (%) calcd for C25H7Cl9O12: C 36.69,H 0.86;
found: C 36.65,H 0.89.
[3] a) T. Tanaka,T. Tasaki,Y. Aoyama, J. Am. Chem. Soc. 2002, 124,
12453–12462; b) R. Melendez,A. Hamilton, Top. Curr. Chem.
1998, 198,97–129.
Compound 5: An Et2O solution of freshly prepared diazomethane (0.4m,
16 mL,6.43 mmol,7 equiv) was slowly added at room temperature to a
solution of hexa-acid 4 (750 mg,0.92 mmol,1 equiv) in freshly distilled
Et2O (100 mL). The mixture was stirred for 8 h and the excess of diazo-
methane was removed by bubbling nitrogen. The organic phase was
washed with aqueous sodium carbonate,dried over anhydrous sodium
sulfate,and the solvent was evaporated. The remaining solid was purified
over silica gel by using pure dichloromethane as an eluent to give 5 as a
white powder in 80% yield (670 mg,0.74 mmol). M.p. 192 8C (decomp);
1H NMR ([D6]DMSO): d=3.9 (s,9H),4.0 (s,9H),6.9 ppm (s,1H);
13C NMR ([D6]DMSO): d=50.6,53.1,53.2,128.5,133.7,133.8,134,134.9,
135,163.4,163.5 ppm; IR (KBr): n˜ =3006,2955,2919,2850,1744,1555,
1436,1378,1327,1253,1182,1112,1042,991,935,891,833,822,807,785,
[4] a) B. Moulton,M. Zaworotko, Chem. Rev. 2001, 101,1629–1658;
b) B. Gong,C. Zheng,E. Skrzypczak-Jankun,Y. Yan,J. Zhang,
J.
Am. Chem. Soc. 1998, 120,11194–11195; c) N. Malek,T. Maris,M.-
E. Perron,J. D. Wuest, Angew. Chem. 2005, 117,4089–4093; Angew.
Chem. Int. Ed. 2005, 44,4021–4025; d) C. Krishnamohan Sharma,
A. Clearfield, J. Am. Chem. Soc. 2000, 122,4394–4402.
[5] D. Philp,J. Fraser Stoddart, Angew. Chem. 1996, 108,1242–1286;
Angew. Chem. Int. Ed. Engl. 1996, 35,1154–1196.
[6] G. R. Desiraju, Angew. Chem. 1995, 107,2541–2558; Angew. Chem.
Int. Ed. Engl. 1995, 34,2311–2327.
765,671,645,620,560,543,490 cm
C31H19Cl9O12: C 41.25,H 2.12; found: C 41.25,H 2.18.
Radical 6: tetrabutylammonium hydroxide solution in methanol
À1; elemental analysis (%) calcd for
[7] To the best of our knowledge,most of the purely organic 3D assem-
blies reported so far are based on four-connecting molecular build-
ing blocks with Td symmetry,such as adamantane,tetraalkyl- or -ar-
ylmethane,silane,or stannane,and twisted bipyrazole or biphenyl
derivatives. See: a) O. Ermer,A. Eling, Angew. Chem. 1988, 100,
856–860; Angew. Chem. Int. Ed. Engl. 1988, 27,829–833; b) O.
Ermer, J. Am. Chem. Soc. 1988, 110,3747–3754; c) P. Holy,J.
Zavada,I. Cisarova,J. Podlaha, Angew. Chem. 1999, 111,393–395;
Angew. Chem. Int. Ed. 1999, 38,381–383; d) P. Holy,J. Zavada,I.
Cisarova,J. Podlaha, Tetrahedron: Asymmetry 2001, 12,3035–3045;
e) J. Moorthy,R. Natarajan,P. Venugopalan, J. Org. Chem. 2005, 70,
8568–8571; f) X. Wang,M. Simard,J. D. Wuest, J. Am. Chem. Soc.
1994, 116,12119–12120; g) D. S. Reddy,D. Craig,G. R. Desiraju, J.
Am. Chem. Soc. 1996, 118,4090–4093; h) P. Brunet,M. Simard,
J. D. Wuest, J. Am. Chem. Soc. 1997, 119,2737–2738; i) D. S. Reddy,
A
(1.75 mL,0.26 mmol,1.2 equiv) was added to a solution of 5 (200 mg,
0.22 mmol,1 equiv) in freshly distilled THF (50 mL). The mixture was
stirred for 1 h,and p-chloranil (220 mg,0.88 mmol,4 equiv) was added as
a solid. After 2 h of stirring,the solvent was removed to yield a purple
solid that was purified over a preparative silica-gel plate by using pure
CH2Cl2 as an eluent. The pure radical 6 was recovered as a red solid,in
80% yield (160 mg,0.18 mmol). M.p. 186 8C (decomp); IR (KBr): n˜ =
2955,2924,2853,1747,1556,1454,1435,1379,1328,1237,1177,1116,
987,932,888,824,804,763,740,641,615,547,484 cm
À1; UV/Vis (THF):
lmax (e)=378 (30050),545 nm (775 mol À1 dm3 cmÀ1); elemental analysis
(%) calcd for C31H18Cl9O12: C 41.30,H 2.01; found: C 41.36,H 1.94.
Radical 1: Compound 5 (160 mg,0.18 mmol) was mixed with concentrat-
ed sulfuric acid and the mixture was heated at 908C for 12 h. The final
solution was cooled and poured onto cracked ice,the aqueous phase was
saturated with brine and extracted with Et2O. The organic phase was con-
centrated and extracted with aqueous sodium carbonate. The resulting
aqueous phase was acidified,saturated with brine,and extracted several
times with Et2O. The organic phase was dried over anhydrous sodium
sulfate,the solvent was removed,and the crude product was dissolved in
Et2O (5 mL) and precipitated several times in hexane to give 1 as a red
powder in 90% yield (130 mg,0.16 mmol). IR (KBr): n˜ =3500–2500,
1715,1631,1538,1439,1340,1246,1120,1064,974,928,900,830,812,
T. Dewa,K. Endo,Y. Aoyama,
4438; Angew. Chem. Int. Ed. 2000, 39,4266–4268; j) R. Thaimattam,
C. K. Sharma,A. Clearfield,G. R. Desiraju, Cryst. Growth Des.
2001, 1,103–106; k) J.-H. Fournier,T. Maris,M. Simard,J. D.
Wuest, Cryst. Growth Des. 2003, 3,535–540; l) I. Boldog,E. Rusa-
Angew. Chem. 2000, 112,4436–
nov,J. Sieler,S. Blaurock,K. Domasevitch,
Chem. Commun. 2003,
740–741; m) J.-H. Fournier,T. Maris,J. D. Wuest,W. Guo,E. Ga-
loppini, J. Am. Chem. Soc. 2003, 125,1002–1006; n) H. Sauriat-Dor-
izon,T. Maris,J. D. Wuest, J. Org. Chem. 2003, 68,240–246; o) J.-H.
Fournier,T. Maris,J. D. Wuest, J. Org. Chem. 2004, 69,1762–1775;
741,681,622,583,487 cm
À1; UV/Vis (THF): lmax (e)=378 (30150),
p) O. Saied,T. Maris,X. Wang,M. Simard,J. D. Wuest,
J. Am.
545 nm (820 molÀ1 dm3 cmÀ1); elemental analysis (%) calcd for
C25H6Cl9O12: C 36.74,H 0.74; found: C 36.70,0.77.
Chem. Soc. 2005, 127,10008–10009.
[8] a) J. D. Dunitz, Chem. Commun. 2003,545–548; b) G. R. Desiraju,
Nat. Mater. 2002, 1,77–79; c) A. Gavezzotti, Acc. Chem. Res. 1994,
27,309–314; d) J. Maddox, Nature 1988, 335,201.
[9] a) Magnetic Properties of Organic Materials (Ed.: P. Lahti),Marcel
Dekker,New York, 1999; b) Magnetism, Vols. 1–3: Molecules to Ma-
terials (Eds.: J. Miller,M. Drillon),Wiley-VCH,Weinheim, 2001.
[10] R. G. Hicks,M. T. Lemaire,L. Öhrstrçm,J. F. Richardson,L. K.
Thompson,Z. Xu, J. Am. Chem. Soc. 2001, 123,7154–7159.
Compound 10: IR (KBr): n˜ =3500–2500,1716,1666,1596,1550,1478,
1384,1350,1250,1123,929,885,711,622 cm
À1; UV/Vis (DMSO): lmax
(e)=444 nm (20090 molÀ1 dm3 cmÀ1); elemental analysis (%) calcd for
C25H6Cl8O13: C 36.74,H 0.74; found: C 36.67,H 0.80.
[11] A. Alberola,R. J. Less,C. Pask,J. M. Rawson,F. Palacio,P. Oliete,
C. Paulsen,A. Yamaguchi,R. D. Farley,D. M. Murphy,
Chem. 2003, 113,4930–4933; Angew. Chem. Int. Ed. 2003, 42,4782–
Angew.
Acknowledgements
4785.
This work was supported by the EU under a Marie Curie Research
Training Network contract (“QuEMolNa” number MCRTN-CT-2003-
504880) and NOE MAGMANET (contract 515767-2),as well as by
MEC (DGI,Spain) under project MAT2003-04699. N.R. thanks the
MCRTN for its postdoctoral contract and D.M. is grateful to the Gener-
alitat de Catalunya for a postdoctoral grant. We would also like to thank
Dr. Jose Vidal-Gancedo for his help with the EPR spectra.
[12] a) E. Hernandez,M. Mas,E. Molins,C. Rovira,J. Veciana,
Angew.
Chem. 1993, 105,919–921; Angew. Chem. Int. Ed. Engl. 1993, 32,
882–884; b) J. Cirujeda,L. Ochando,J. Amigo,C. Rovira,J. Veci-
ana, Angew. Chem. 1995, 107,99–102; Angew. Chem. Int. Ed. Engl.
1995, 34,55–57; c) J. Cirujeda,M. Mas,E. Molins,F. Lanfranc de
Panthou,J. Laugier,J. Park,C. Paulsen,P. Rey,C. Rovira,J. Veci-
9252
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
Chem. Eur. J. 2006, 12,9238 – 9253

Six-Connecting Organic Building Blocks
FULL PAPER
ana, J. Chem. Soc. Chem. Commun. 1995,709–710; d) J. Cirujeda,
E. Hernandez-Gasio,C. Rovira,J.-L. Stranger,P. Turek,J. Veciana,
J. Mater. Chem. 1995, 5,243–252; e) J. Veciana,J. Cirujeda,C.
Rovira,E. Molins,J. Novoa, J. Phys. I 1996,1967–1986; f) T. Suga-
wara,M. Matsushita,A. Isuoka,N. Wada,N. Takeda,M. Ishikawa,
J. Am. Chem. Soc. 1997, 119,4369–4379.
[31] a) M. Ballester,J. Riera,J. CastaÇer,C. Rovira,O. Armet,
1986,64–66; b) D. Ruiz-Molina,J. Veciana,F. Palacio,C. Rovira,
Org. Chem., 1997, 62,9009–9017.
[32] Diazomethane was prepared from Diazald (Aldrich) by a well-docu-
mented procedure. See: Technical Bulletin AL113,Aldrich,Milwau-
kee,WI, 2000.
Synthesis
J.
[13] T. Akita,K. Kobayashi, Adv. Mater. 1997, 9,346–349.
[14] a) N. Yoshioka,M. Irisawa,Y. Mochizuki,K. Tano,H. Inoue,S.
Ohba, Chem. Lett. 1997, 26,251–252; b) N. Yoshioka,M. Irisawa,Y.
Mochizuki,T. Aoki,H. Inoue, Mol. Cryst. Liq. Cryst. Sci. Technol.
Sect. A 1997, 306,403–408.
[33] Radicals 1 and 6 have very similar UV/Visible spectra,with absorp-
tion bands characteristic of PTM radicals at 380 and 550 nm. See:
M. Ballester,J. Riera,J. CastaÇer,A. Rodrꢂguez,C. Rovira,J. Veci-
ana, J. Org. Chem. 1982, 47,4498–4505.
[34] O. Armet,J. Veciana,C. Rovira,J. Riera,J. Castaner,E. Molins,J.
Rius,C. Miratvilles,S. Olivilla,J. Brichfeus, J. Phys. Chem. 1987, 91,
5608.
[15] J. Ferrer,P. D. Lahti,C. George,G. Antonerra,F. Palacio,
Chem.
Mater. 1999, 11,2205–2210.
[16] A. Lang,Y. Pei,L. Ouahab,O. Khan,
[17] R. Feher,D. Amabilino,K. Wurst,J. Veciana,
Cryst. Sci. Technol. Sect. A 1999, 334,333–345.
[18] L. Catala,R. Feher,D. Amabilino,K. Wurst,J. Veciana,
Adv. Mater. 1996, 8,60–62.
[35] a) S. Dale,M. Elsegood,A. Coombs, Cryst. Eng. Comm., 2004, 6,
328; b) S. Dale,M. Elsegood, Acta Crystallogr. Sect. E 2003, 59,
o1205.
[36] T. Steiner, Angew. Chem. 2002, 114,50–80; Angew. Chem. Int. Ed.
2002, 41,48.
[37] Poor quality crystals were always obtained on diffusion of n-hexane
into THF or Et2O solutions of 6.
[38] Determined by using A. M. C. T. PLATON,Utrecht University,
Utrecht,The Netherlands,A. L. Spek, 1998.
Mol. Cryst. Liq.
Polyhedron
NewJ.
2001, 20,1563–1569.
[19] a) F. Romero,R. A. Ziessel,A. De Cian,J. Fischer,P. Turek,
Chem. 1996, 20,919–924; b) F. Romero,R. Ziessel,M. Drillon,J.-L.
Tholence,C. Paulsen,N. Kyritsakas,J. Fischer, Adv. Mater. 1996, 8,
826–829; c) F. Romero,R. Ziessel,M. Bonnet,Y. Potillon,E. Re-
ssouche,J. Schweitzer,B. Delley,A. Grand,C. Paulsen,
Chem. Soc. 2000, 122,1298–1309.
J. Am.
[39] M. Ballester,J. Riera,J. CastaÇer,C. Badia,J. Monso,
J. Am. Chem.
Soc. 1971, 93,2215–2225.
[20] a) K. Inoue,H. Iwamura, Chem. Phys. Lett. 1993, 207,551–554;
b) O. FØlix,M. W. Hosseini,A. De Cian,J. Fischer,L. Catala,P.
Turek, Tetrahedron Lett. 1999, 40,2943–2946; c) C. Stroh,F.
[40] Similar oxidations have been reported to occur slowly on treatment
of PTMMC and PTMMI (PTM substituted with an iodine atom in
the para position) with concentrated nitric acid at room tempera-
Romero,N. Kyritsakas,L. Catala,P. Turek,R. Ziessel,
J. Mater.
ture. See: M. Ballester,J. Riera,J. CastaÇer,A. IbaÇez,J. Pujadas,
J.
Chem. 1999, 9,875–882; d) D. Maspoch,L. Catala,P. Gerbier,D.
Ruiz-Molina,J. Vidal-Gancedo,K. Wurst,C. Rovira,J. Veciana,
Chem. Eur. J. 2002, 8,3635–3645; e) M. Minguet,D. Amabilino,J.
Org. Chem. 1982, 47,259–264.
[41] Such dihedral angles and the bond lengths exhibited by 10 are in
agreement with those previously reported for the fully chlorinated
fuchsone. See: E. Molins,J. Rius,C. Miratvilles, Z. Kristallogr. New
Cryst. Struct. 1984, 169,149–158.
[42] The dihedral angles in the motif (O7-C14-C3-O5=548),together
with bond lengths and IR data,enable us to discount the formation
of any cyclic hydrogen bonding involving the ketone function similar
to that encountered in some b-diketone derivatives. See: G. Gilli,F.
Cirujeda,K. Wurst,I. Mata,E. Molins,J. Novoa,J. Veciana,
Eur. J. 2000, 6,2350–3361; f) M. Baskett,P. D. Lahti, Polyhedron
2005, 24,2645–2652.
Chem.
[21] T. Otsuka,T. Okuno,K. Awaga,T. Inabe,
1157–1163.
J. Mater. Chem. 1998, 8,
[22] M. Ballester, Acc. Chem. Res. 1985, 18,380–387.
[23] K. Kobayashi,T. Shirasaka,A. Sato,E. Horn,N. Furukawa,
Angew.
Bellucci,F. Feretti,G. Gilli,
1028.
[43] Similar bond lengths and dihedral angles have been encountered in
the case of cubanecarboxylic acid derivatives presenting the syn,an-
ti-catemer motive. See: S. Kuduva,D. Craig,A. Nangia,G. R. Desir-
aju, J. Am. Chem. Soc. 1999, 121,1936–1944.
[44] To the best of our knowledge, syn,anti-catemer motifs have been
only reported in reference [43],in phenylpropiolic acid derivatives
and in monocarboxylic acid derivatives RCOOH with small R sub-
J. Am. Chem. Soc. 1989, 111,1023–
Chem. 1999, 111,3692–3694; Angew. Chem. Int. Ed. 1999, 38,3483–
3486.
[24] a) G. R. Desiraju, Crystal Engineering: The Crystal as a Supramolec-
ular Entity, Vol. 2: Perspectives in Supramolecular Chemistry,Wiley,
New York, 1996; b) J. Bernstein,M. Etter,L. Leiserowitz, Structure
Correlation, Vol. 2,Wiley,New York, 1994,Chapter 11.
[25] D. Maspoch,N. Domingo,D. Ruiz-Molina,K. Wurst,J. Tejada,C.
Rovira,J. Veciana, J. Am. Chem. Soc. 2004, 126,730–731.
[26] D. Maspoch,N. Domingo,D. Ruiz-Molina,K. Wurst,G. Vaughan,J.
Tejada,C. Rovira,J. Veciana, Angew. Chem. 2004, 116,1864–1868;
Angew. Chem. Int. Ed. 2004, 43,1828–1832.
stituents, a-oxalic,formic,acetic,and
b-terolic acids. See: a) G. R.
Desiraju,B. Murty,K. Kishan, Chem. Mater. 1990, 2,447–449; b) D.
Das,R. Jetti,R. Boese,G. R. Desiraju, Cryst. Growth Des. 2003, 3,
675–681; c) D. Das,G. R. Desiraju,R. Jetti,R. Boese, Acta Crystal-
logr. Sect. E 2005, 61,O1588–O1589; d) J. Derissen,P. Smit, Acta
Crystallogr. Sect. B 1974, 30,2240–2242; e) I. Nahringbauer, Acta
Crystallogr. Sect. B 1978, 34,315–318; f) V. Benghiat,L. Leissero-
witz, J. Chem. Soc. Perkin Trans. 2 1972,1763–1972.
[27] Only a few examples of hexa-acid molecules have been reported in
the literature. See: a) S. Darlow, Acta Crystallogr. 1961, 14,159–
166; b) J. Podlaha,I. Cisarova,P. Holy,J. Zavada,
Z. Kristallogr.
NewCryst. Struct. 1999, 214,185–186; c) J. Beeson,L. Fitzgerald,J.
Gallucci,R. Gerkin,J. Rademacher,A. Czarnick, J. Am. Chem. Soc.
1994, 116,4621–4622.
[45] a) D. Maspoch,D. Ruiz-Molina,K. Wurst,N. Domingo,M. Cavalli-
[28] To the best of our knowledge,only one example of an octa-acid has
been reported. See: P. Holy,P. Sehnal,M. Tichy,J. Zavada,I. Cisar-
ova, Tetrahedron: Asymmetry 2004, 15,3805–3810.
ni,F. Biscarini,J. Tejada,C. Rovira,J. Veciana,
Nat. Mater. 2003, 2,
190–195; b) D. Maspoch,D. Ruiz-Molina,K. Wurst,C. Rovira,J.
Veciana, Chem. Commun. 2004,1164–1165; c) D. Maspoch,N.
Domingo,D. Ruiz-Molina,K. Wurst,J.-M. Hernꢃndez,G. Vaughan,
[29] B. Venkataramanan,W. L. G. James,J. J. Vittal,V. Suresh,
Cryst.
Growth Des. 2004, 4,553–561.
C. Rovira,F. Lloret,J. Tejada,J. Veciana,
5035–5037.
Chem. Commun. 2005,
[30] A preliminary communication of this work has already been pub-
lished. See: N. Roques,D. Maspoch,N. Domingo,K. Wurst,J.
Tejada,C. Rovira,J. Veciana, Chem. Commun. 2005,4801–4803.
Received: March 31,2006
Published online: September 12,2006
Chem. Eur. J. 2006, 12,9238 – 9253
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA,Weinheim
www.chemeurj.org
9253