A new method for the synthesis of boronate macrocycles

Article abstract of DOI:10.1039/b402510e

The condensation of aryl boronic acids with 2,3-dihydroxypyridine gives boronates, which self-assemble to form tetrameric macrocycles as evidenced by X-ray crystallographic analyses.

Full text of DOI:10.1039/b402510e

A new method for the synthesis of boronate macrocycles
Nicolas Christinat, Rosario Scopelliti and Kay Severin*
Institute of Chemical Sciences and Engineering, Swiss Federal Institute of Technology Lausanne (EPFL),
1015 Lausanne, Switzerland. E-mail: kay.severin@epfl.ch
Received (in Columbia, MO, USA) 18th February 2004, Accepted 2nd April 2004
First published as an Advance Article on the web 21st April 2004
The condensation of aryl boronic acids with 2,3-dihydroxy-
pyridine gives boronates, which self-assemble to form tetra-
meric macrocycles as evidenced by X-ray crystallographic
analyses.
azeotropic benzene–water mixture. Upon cooling of the benzene
solution, the product (1) precipitated as a white powder (isolated
yield: 51%).7†
Compound 1 is well soluble in organic solvents such as
chloroform. This is in contrast to macrocycles prepared by the
method depicted in Scheme 1(a), which were often found to display
a very limited solubility.3 The 1H and 13C NMR spectra of 1
indicated that a highly symmetrical complex had formed because
only one set of signals was found for the phenyl group as well as for
the pyridine ligand. Since NMR spectroscopy is not suited to
determine the association number n, we have investigated the
molecular structure of 1 in the solid state by single crystal X-ray
analysis.‡ It was found that a tetrameric assembly had formed (Fig.
1). As expected, the boronic acid has undergone a condensation
reaction with the two adjacent hydroxy groups to give a five-
membered cyclic ester and the pyridine N-atom forms a dative bond
to the next boron centre. Overall, the macrocycle displays a perfect
S4 symmetry. The planes of the heterocyclic ligands are nearly
orthogonal to each other forming a molecular square. It is
interesting to note that organometallic halfsandwich complexes of
RuII, RhIII, and IrIII, which likewise display a (pseudo)tetrahedral
coordination geometry, form exclusively trimeric macrocycles with
the same bridging heterocycle.5 Apparently, the smaller boron atom
is able to switch the assembly process entirely from n = 3 to n =
4.
The utilization of boron compounds as building blocks for the
construction of macrocyclic two- and three-dimensional assemblies
is an emerging topic in supramolecular chemistry.1 Similar to many
transition metal complexes, boron compounds may form direc-
tional bonds, which are thermodynamically stable but kinetically
labile. This feature allows error correction processes to occur
during assembly reactions, a characteristic that is of central
importance for the synthesis of complex structures. As starting
materials, aryl boronic acids appear to be of special interest since
many derivatives are commercially available. Furthermore, aryl
boronic acids are known to easily undergo condensation reactions
with various diols to form boronic esters,2 a reaction that appears to
be a good entry point for the formation of more complex
structures.
First results in this area have been reported by Höpfl and Farfán.3
They have successfully employed the reaction outlined in Scheme
1(a) for the construction of di-, tri- and tetranuclear boron
macrocycles with ring sizes between 10 and 20 atoms. A different
strategy, which—to the best of our knowledge—has not been
investigated so far, is depicted in Scheme 1(b). Again, an aryl
boronic acid is reacted with a tridentate amino dialcohol. But now,
the ligand has two adjacent hydroxy groups. Consequently, it is
expected that an O–B–OA chelate is formed initially which can then
assembly to give the macrocycle. The main difference between the
two reactions is that for the latter pathway, macrocyclization occurs
via formation of dative B–N bonds4 and not via strong covalent B–
O bonds.
In order to test the flexibility of the synthetic approach, we have
investigated whether substitutions of the bridging ligand and on the
aryl group are tolerated. This is indeed possible as evidenced by the
successful preparation of complexes 2 and 3 (Scheme 2).† For the
synthesis of complex 2, a ligand with a morpholinomethyl group in
In order to test the feasibility of the latter approach, we have first
examined the reaction between phenyl boronic acid and 2,3-dihy-
droxypyridine. This ligand was chosen because it has been
successfully employed in transition metal based self-assembly
reactions using metal fragments with a (pseudo)tetrahedral coor-
dination geometry.5 Furthermore, several derivatives are easily
accessible which would allow to fine-tune the electronic properties
and the solubility of the resulting assemblies.6 In order to efficiently
remove the by-product water, the reaction was performed in
benzene under reflux using a Dean–Stark trap to separate the
Scheme 1 Different strategies to synthesize macrocyclic boronates: (a)
macrocyclization via covalent B–O bonds; (b) macrocyclization via dative
B–N bonds.
Scheme 2 Synthesis of the tetrameric boronates 1–3.
1158
C h e m . C o m m u n . , 2 0 0 4 , 1 1 5 8 – 1 1 5 9
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4

a discussion with Prof. T. Severin and we are thankful for his
comments.
Notes and references
† A suspension of phenyl boronic acid (219 mg, 1.8 mmol) and
2,3-dihydroxypyridine (200 mg, 1.8 mmol) in freshly distilled benzene (60
mL) was heated under reflux using a Dean–Stark trap. After 15 h, the
suspension was filtered hot. Upon cooling, a white precipitate formed which
was isolated and dried under vacuum. (yield: 180 mg, 51%). 1H NMR (400
MHz, CDCl3): d 6.61 (t, 3J = 7 Hz, 4 H, pyridine), 6.67 (d, 3J = 6 Hz, 4
H, pyridine), 6.99 (d, 3J = 7 Hz, 4 H, pyridine), 7.05–7.35 (m, 20 H,
phenyl); 13C NMR (400 MHz, CDCl3): d 114.17, 115.78, 127.69, 128.09,
128.12, 132.04, 151.26, 163.83; 11B NMR (400 MHz, CDCl3): d 11.5;
elemental analysis (%) calc. for C44H32B4N4O8: C 67.07, H 4.09, N 7.11;
obtained: C 67.32, H 4.18, N 6.86. The synthesis of 2 (yield: 84%) and 3
(yield: 84%) was performed analogously using 2,3-dihydroxy-4-morpholi-
nomethyl-pyridine and 2,3,6-C6H2F3B(OH)2. All reactions were carried out
under an inert atmosphere (N2). Crystals were obtained by slow diffusion of
pentane into solutions of the respective complexes in CH2Cl2 (1) or benzene
(2, 3).
Fig. 1 Molecular structure of 1 in the crystal.
position 4 was employed8 and for complex 3, the commercially
available 2,3,6-trifluorophenyl boronic acid was used. The self-
assembly reactions are not affected by these substitutions. This was
confirmed by NMR spectroscopy and single crystal X-ray analy-
ses.‡ For complex 2, the presence of stereogenic boron centres is
manifested by the presence of two diastereotopic methylene
protons, which give rise to two doublets in the 1H NMR spectrum.
The structures of 2 and 3 in the crystal are very similar to that of 1
although a crystallographic S4 symmetry is no longer present. A
comparison of important bond length and angles is given in Table
1.
The B–N and the B–O bonds of the complexes 2 and 3 are
slightly shorter than those found for 1. This leads to an overall
contraction of the macrocycles as reflected by the reduced B…B
distances of 2 and 3. The O–B–N angles found for all complexes are
close to the 109.5° expected for a perfect tetrahedral geometry. The
B–N bond lengths are of special interest because they represent a
key element of the macrocyclic framework. Dative B–N bonds can
range from 1.57 to 2.91 Å.4 The average B–N bond length of the
complexes 1–3 is 1.59 Å. This is shorter than what is typically
found for tetrahedral boronates with N-donor ligands, which show
a value around 1.69 Å. The simple adduct between 4-methylpyr-
idine and 2-phenyl-1,3,2-benzodioxaborole, for example, displays
a B–N bond length of 1.654 Å in the crystal.9
The data described above suggest that the B–N bonds in 1–3 are
thermodynamically rather stable. In order to investigate the kinetic
stability of the assemblies, we have performed scrambling
experiments. Equimolar amounts of complex 1 and 2 were
dissolved in CDCl3. Even after 24 h, the 1H NMR spectrum of the
mixture was unchanged, indicating that no mixed species had
formed. Apparently, the macrocycles are kinetically rather inert.
In summary, we have described a new method for the synthesis
of macrocyclic boronate complexes. It seems likely that this
synthetic concept can be expanded by employing other tridentate
ligands such as 3,4-dihydroxy-2-methylpyridine, 2-hydroxynico-
tinic acid, or 2,3-dihydroxyquinoline, all of which have success-
fully been employed in transition metal-based self-assembly
reactions.10,11 Furthermore, it should be possible to replace aryl
with alkyl boronic acids which would significantly enhance the
structural diversity that is accessible. Preliminary results indicate
that this is indeed possible.12
‡ Crystal data for 1: C44H32B4N4O8, M
= 787.98, tetragonal, a =
16.7128(8), c = 13.9916(11) Å, V = 3908.1(4) Å3, T = 140(2) K, space
group I41/a (no. 88), Z = 4, m(Mo-Ka) = 0.091 mm21, 11 750 reflections
collected, 1729 independent reflections, Rint = 0.0384, R1 [I > 2s(I)] =
0.0347, wR2 (all data) = 0.0943. For 2: C78.5H86B4N8O12, M = 1376.79,
orthorhombic, a = 26.3143(18), b = 28.029(2), c = 20.2248(11) Å, V =
14917.2(17) Å3, T = 140(2) K, space group Iba2 (no. 45), Z = 8, m(Mo-Ka)
= 0.082 mm21, 44 532 reflections collected, 12 697 independent reflec-
tions, Rint = 0.0756, R1 [I > 2s(I)] = 0.0625, wR2 (all data) = 0.1498. For
3: C56H32B4F12N4O8, M
= 1160.10, triclinic, a = 13.538(4), b =
13.595(10), c = 13.813(12) Å, a = 80.64(7), b = 85.41(4), g = 80.58(4)°,
¯
V = 2471(3) Å3, T = 140(2) K, space group P1 (no. 2), Z = 2, m(Mo-Ka)
= 0.133 mm21, 16 104 reflections collected, 8192 independent reflections,
Rint = 0.1172, R1 [I > 2s(I)] = 0.0774, wR2 (all data) = 0.2432. CCDC
233235–233237. See http://www.rsc.org/suppdata/cc/b4/b402510e/ for
crystallographic data in CIF or other electronic format.
1 For reviews, see: (a) H. Höpfl, Struct. Bonding, 2002, 103, 1–56; (b) K.
Ma, M. Scheibitz, S. Scholz and M. Wagner, J. Organomet. Chem.,
2002, 652, 11–19.
2 G. Springsteen and B. Wang, Tetrahedron, 2002, 58, 5291–5300.
3 (a) M. Sánchez, H. Höpfl, M.-E. Ochoa, N. Farfán, R. Santillan and S.
Rojas-Lima, Chem. Eur. J., 2002, 8, 612–621; (b) V. Barba, E.
Gallegos, R. Santillan and N. Farfán, J. Organomet. Chem., 2001, 622,
259–264; (c) V. Barba, R. Luna, D. Castillo, R. Santillan and N. Farfán,
J. Organomet. Chem., 2000, 604, 273–282; (d) V. Barba, D. Cuahutle,
M. E. Ochoa, R. Santillan and N. Farfán, Inorg. Chim. Acta, 2000, 303,
7–11; (e) N. Farfán, H. Höpfl, V. Barba, M. E. Ochoa, R. Santillan, E.
Gómez and A. Gutiérrez, J. Organomet. Chem., 1999, 581, 70–81; (f) H.
Höpfl, M. Sánchez, V. Barba, N. Farfán, S. Rojas and R. Santillan,
Inorg. Chem., 1998, 37, 1679–1692; (g) H. Höpfl and N. Farfán, J.
Organomet. Chem., 1997, 547, 71–77.
4 H. Höpfl, J. Organomet. Chem., 1999, 581, 129–149.
5 (a) Z. Grote, R. Scopelliti and K. Severin, Angew. Chem., Int. Ed., 2003,
42, 3821–3825; (b) M.-L. Lehaire, R. Scopelliti, H. Piotrowski and K.
Severin, Angew. Chem., Int. Ed., 2002, 41, 1419–1422; (c) M.-L.
Lehaire, R. Scopelliti and K. Severin, Chem. Commun., 2002,
2766–2767; (d) M.-L. Lehaire, R. Scopelliti and K. Severin, Inorg.
Chem., 2002, 41, 5466–5474; (e) H. Piotrowski and K. Severin, Proc.
Natl. Acad. Sci. U. S. A., 2002, 99, 4997–5000; (f) H. Piotrowski, K.
Polborn, G. Hilt and K. Severin, J. Am. Chem. Soc., 2001, 123,
2699–2700; (g) H. Piotrowski, G. Hilt, A. Schulz, P. Mayer, K. Polborn
and K. Severin, Chem. Eur. J., 2001, 7, 3196–3208.
6 (a) Z. Grote, M.-L. Lehaire, R. Scopelliti and K. Severin, J. Am. Chem.
Soc., 2003, 125, 13 638–13 639; (b) M.-L. Lehaire, A. Schulz, R.
Scopelliti and K. Severin, Inorg. Chem., 2003, 42, 3576–3581.
7 The total yield of the reaction is > 90% as evidenced by NMR
investigations.
We gratefully acknowledge support of this work by the Swiss
National Science Foundation. The investigations were initiated by
8 The ligand was obtained in a Mannich reaction from 2,3-dihydroxypyr-
idine, see: K.-W. Chi, Y. S. Ahn, T. H. Park, J. S. Ahn, H. A. Kim and
J. Y. Park, J. Korean Chem. Soc., 2001, 45, 51–60.
9 W. Clegg, A. J. Scott, F. E. S. Souza and T. B. Marder, Acta
Crystallogr., Sect. C, 1999, 55, 1885–1888.
Table 1 Selected distances (Å) and angles (°) for the compounds 1–3
B–N
B–O1
B–O2
B…BAa
O2–B–N O1–B–N
1
1.601(2) 1.529(2)
2b 1.587(6) 1.524(5)
3b 1.58(1)
1.506(9)
1.496(2)
1.481(5)
1.487(9)
5.624(2) 110.1(1)
5.318(7) 108.8(3)
105.1(1)
106.0(3)
108.0(6)
10 M.-L. Lehaire, R. Scopelliti, L. Herdeis, K. Polborn, P. Mayer and K.
Severin, Inorg. Chem., 2004, 43, 1609–1617.
11 T. Habereder, M. Warchhold, H. Nöth and K. Severin, Angew. Chem.,
Int. Ed., 1999, 38, 3225–3228.
12 N. Christinat, Diploma Thesis, École Polytechnique Fédérale de
Lausanne, Switzerland, 2004.
5.31(1)
109.0(6)
a The distance between the boron atoms opposite to each other is given.
b Averaged values are given.
C h e m . C o m m u n . , 2 0 0 4 , 1 1 5 8 – 1 1 5 9
1159