Aluminium(III) porphyrins as supramolecular building blocks

Article abstract of DOI:10.1039/b605435h

Aluminium(III) porphyrin-carboxylate complexes, including a porphyrin pentamer, have been characterised by NMR spectroscopy, MALDI spectrometry and single crystal X-ray diffraction; these complexes can also be coordinated by a sixth, nitrogenous, ligand to the aluminium(III) centre. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b605435h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Aluminium(III) porphyrins as supramolecular building blocks{
Gregory J. E. Davidson,a Lok H. Tong,a Paul R. Raithbybc and Jeremy K. M. Sanders*a
Received (in Cambridge, UK) 19th April 2006, Accepted 2nd June 2006
First published as an Advance Article on the web 19th June 2006
DOI: 10.1039/b605435h
Aluminium(III) porphyrin–carboxylate complexes, including a
porphyrin pentamer, have been characterised by NMR
spectroscopy, MALDI spectrometry and single crystal X-ray
diffraction; these complexes can also be coordinated by a sixth,
nitrogenous, ligand to the aluminium(III) centre.
The design of multiporphyrin arrays has been an area of intense
research. Their use in areas such as molecular recognition, sensing,
molecular-level electronics and natural system mimics has been the
ESI{).{ An ethanol molecule is coordinated to the sixth
major driving force. These arrays have been assembled by covalent
bond formation or through non-covalent interactions such as
hydrogen bonding.1 However, designing and building arrays
utilising axial coordination has proven to be a useful approach
to the development of these assemblies with well defined
geometries.2 Knowledge of the preferred coordination number,
geometry and ligand preference of a metalloporphyrin allows for
the rational design of such arrays. To date, much of the attention
has been paid to metalloporphyrins containing zinc(II),
ruthenium(II), tin(IV), platinum(II) and rhodium(III) metal centres.3
For many years the use of aluminium(III) porphyrins has mostly
been limited to polymerisation reactions and activation of organic
substrates.4 There currently exist only a few examples in which an
aluminium(III) porphyrin has been used for the construction of
arrays which employ phenolates as the axial ligand.5 Surprisingly,
the use of benzoates as the axial ligand, for the development of
arrays, has never been investigated. There have been isolated
reports of aluminium(III)–benzoate complexes, but no systematic
studies.6 Herein, we present a systematic study of axially bound
benzoates to aluminium(III) porphyrins and their potential use as
supramolecular building blocks.
coordination site of the aluminium(III) centre in the solid state.
The axially bound benzoate ligand is not displaced by water but
exchange does occur with excess competitive carboxylic acids. This
is slow on the chemical shift timescale, but rapid on the relaxation
timescale, giving intense EXSY cross peaks between free and
coordinated acids. A Hammett plot was constructed to gain some
insight into the thermodynamics of the exchange process by
making a series of 1:1 solutions containing 3 and a 4-substituted
benzoic acid (R = –NO2, –F, –Me, –tert-butyl and –OMe). In all
cases, the exchange was slow on the chemical shift timescale and
the ratio of association constants, K/KH, was determined from
the integration values of the two bound species at equilibrium. The
absolute association constants could not be determined since the
aluminium(III) centre is always carboxylate-bound even at low
concentrations. Fig. 1 shows the resulting linear Hammett plot.
The +0.76 r value is consistent with the development of a negative
charge during the exchange process, the equilibrium being shifted
to the right by electron-withdrawing groups. While the exact
mechanism of exchange is unknown, we speculate that it is similar
to that previously described for the tin(IV)–acetate exchange.7
During these studies a six-coordinate aluminium(III) porphyrin
complex was never detected: the complexes always maintained
The synthesis of 3 proceeded smoothly upon the addition of
benzoic acid to a solution of the reactive intermediate 2, generated
by the addition of trimethylaluminium to tetraphenylporphyrin
1
(TPP), 1. The H NMR spectrum confirmed the coordination of
the axially bound benzoate by the upfield shift of the benzoate
proton resonances due to the shielding effect of the porphyrin. The
magnitude of the upfield shifts for the o-, m- and p-protons of the
benzoate are 2.96 ppm, 1.07 ppm and 0.90 ppm, respectively.
Single crystal X-ray diffraction also confirmed the structure (see
aUniversity of Cambridge, Department of Chemistry, Lensfield Road,
Cambridge, Cambridgeshire, UK CB2 1EW. E-mail: jkms@cam.ac.uk;
Fax: +44 1223 336411; Tel: +44 1223 336017
bDepartment of Chemistry, University of Bath, Claverton Down, Bath,
UK BA2 7AY
cCCLRC Daresbury Laboratory, Daresbury, Warrington, UK
WA4 4AD
{ Electronic supplementary information (ESI) available: X-Ray data,
spectroscopic data, experimental procedures, Hammett plot data and
competition data. See DOI: 10.1039/b605435h
Fig. 1 Hammett plot.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 3087–3089 | 3087

View Article Online
their 1:1 stoichiometry. This is confirmed by the fact that there are
no significant shifts in the proton resonances of any excess ‘free’
acid.
It has been shown previously that neutral nitrogen donor
ligands can be used to fill the vacant sixth coordination site.8 Using
UV-vis spectroscopy the binding of 4-tert-butylpyridine or
N-methylimidazole to 3 was confirmed by a bathochromic shift
of the Soret band from 414 nm to 426 nm. Comparable shifts have
been reported for the addition of pyridines to zinc(II) porphyrin.3a
The UV-vis titrations show an isosbestic point confirming the
presence of only two species in solution: the association constants
were determined to be 3 6 103 M21 for the 4-tert-butylpyridine
and 4 6 105 M21 for the N-methylimidazole.9 In a competition
experiment for N-methylimidazole between complex 3 and the
para-nitro substituted version, there was a preference for the para-
nitro complex by a factor of 4, due to the more electron deficient
aluminium(III) metal centre (see ESI{).
Fig. 3 1H NMR comparison of complex 3 (top) and the coordination of
4-tert-butylpyridine to complex 3 (bottom), emphasising the upfield shift
of the benzoate resonances. Scale in ppm.
Single crystal X-ray analysis (Fig. 2) confirms the coordination
of 4-tert-butylpyridine to the vacant site of 3.{ The aluminium is
bound to the four nitrogens of the porphyrin, with bond lengths
other than the very reactive complex 2, is required. Fortunately,
complex 4, a hydrolysed version of 2, can also be used to generate
3 quantitatively by the addition of one equivalent of benzoic acid.
This reaction also generates one equivalent of water which has no
effect on the highly stable complex 3.
˚
˚
ranging from 1.995(4) A to 2.017(4) A. The bond lengths of the
benzoate oxygen, O(2), and the nitrogen of the 4-tert-butylpyr-
idine, N(5), to the aluminium(III) metal centre, Al(1), are 1.843(3)
˚
and 2.215(4) A, respectively. It has been suggested that five-
coordinate aluminium(III) porphyrin in solution bearing an axial
oxygen donor exists with the metal centre pulled up out of the
plane of the porphyrin, although very little convincing evidence
has been shown.10 To date, there is no known X-ray structure of a
five-coordinate aluminium(III)–benzoate porphyrin complex. The
benzoate bound to complex 3 will be an effective reporter if the
aluminium(III) centre is pulled back into the plane, since this
should result in an additional upfield shift of the benzoate proton
resonances. The 1H NMR spectrum, see Fig. 3, shows the effect of
the addition of 2 equivalents of 4-tert-butylpyridine to 3 on the
proton resonances of the benzoate. As expected, an additional
upfield shift of the coordinated benzoate is seen, consistent with
the benzoate being pulled towards the porphyrin plane. These new
chemical shifts correspond nicely with those of the benzoates
bound to a six-coordinate tin(IV) centre.7
As a first test a porphyrin pentamer was sought. Scheme 1
details the approach used for the construction of a pentamer
consisting of a central tetra(4-carboxyphenyl) porphyrin, 5,
surrounded by four tetraphenylaluminium(III) porphyrins. The
synthesis was carried by the addition of 5 to six equivalents of 4 in
chloroform.11 The formation of the desired structure, 6, was
confirmed by 1H NMR spectroscopy and MALDI mass spectro-
metry. The MALDI spectrum displays the [6]?+ peak at m/z
3343.01, close to the calculated value of 3343.03, with a matching
isotopic pattern. The stepwise formation of 6 can be studied by 1H
NMR spectroscopy as shown in Fig. 4, where the effect of the
addition of two and four equivalents of 4 upon the central NH
protons of the tetra(4-carboxyphenyl) porphyrin core can be seen.
The addition of two equivalents shows the presence of mono-,
di- (both cis and trans), tri- and tetrasubstituted complexes. The
addition of two more equivalents yielded the expected tetrasub-
stituted complex, pentamer 6, indicated by the presence of only
one signal for the central NH protons. This additive upfield shift of
the central protons has been seen previously.12
In order for these aluminium(III) porphyrins to be useful for the
construction of large arrays, a more suitable starting material,
Scheme 1 The synthesis of the porphyrin pentamer. (a) 6 equiv.
Fig. 2 Ball and stick representation of 4-tert-butylpyridine coordinated
to complex 3.
AlTPP(OH), CHCl3.
3088 | Chem. Commun., 2006, 3087–3089
This journal is ß The Royal Society of Chemistry 2006

View Article Online
capable of binding both an oxygen donor and nitrogen donor,
simultaneously, in the axial positions. This robustness and ability
to coordinatively differentiate the two sides of the porphyrin make
them an attractive alternative to tin(IV) porphyrins. The usefulness
of complex 4 as a starting material for large arrays has been
demonstrated through quantitative formation of the stable
porphyrin pentamer, 6. Currently, we are taking advantage of
the coordinative properties of the outer faces, using imidazole-
based ligands, of complex 6 to generate mixed metal arrays.
We thank Dr John Warren (SRS Daresbury) and Dr John
Davies (University of Cambridge) for crystallographic support and
the EPSRC and NSERC (G. J. E. D.) for financial support.
Notes and references
{ CCDC 604449 (3), 604450 (6) and 604451 (3?4-tert-butylpyridine). For
crystallographic data in CIF or other electronic format see DOI: 10.1039/
b605435h. Crystal data for 3: C55H45AlN4O4, M = 852.93, monoclinic,
Fig. 4 1H NMR comparison showing the effects of adding 2 equiv.
(bottom) and 4 equiv. (top) of 4 on the central NH proton resonances in
the core of the pentamer, 6.
˚
P21/n (No. 14), a = 13.326(2), b = 16.495(3), c = 21.114(4) A,
3
21
˚
b = 108.394(2)u, V = 4404.2(13) A , Z = 4, m(Mo-Ka) = 0.100 mm
,
T = 150(2) K, dark red plates; 6652 independent measured reflections
(Rint = 0.0895), F2 refinement, R1 = 0.0501, wR2 = 0.112 on 4236 observed
data [I > 2s(I)], and R1 = 0.0951, wR2 = 0.134 on all data. Crystal data for
3?4-tert-butylpyridine: C60.5H46.5AlN5O2Cl1.5, M = 955.68, monoclinic,
˚
P21/n (No. 14), a = 11.8780(1), b = 10.4528(1), c = 42.0274(4) A, b =
94.962(1)u, V = 5198.5(1) A , Z = 4, m(Mo-Ka) = 0.164 mm21, T =
3
˚
220(2) K, 7369 independent measured reflections (Rint = 0.027), F2
refinement, R1 = 0.114, wR2 = 0.380 on 6355 observed data [I > 2s(I)],
and R1
6: C242H198Al4Cl4N20O16S8, M = 4140.34, triclinic, P1 (No. 2), a =
= 0.124, wR2 = 0.394 on all data. Crystal data for
¯
˚
17.0248(14), b = 18.7182(16), c = 19.0603(16) A, a = 104.949(1),
3
˚
b = 105.955(1), c = 94.339(1)u, V = 5571.7(8) A , Z = 1, m(Mo-Ka) =
0.21 mm21, T = 150(2) K, 15881 independent measured reflections (Rint
=
0.028), F2 refinement, R1 = 0.160, wR2 = 0.471 on 12262 observed data [I >
2s(I)], and R1 = 0.186, wR2 = 0.497 on all data.
1 A. K. Burrell, D. L. Officer, P. G. Plieger and D. C. W. Reid, Chem.
Rev., 2001, 101, 2751–2796.
2 J. K. M. Sanders, in The porphyrin handbook, ed. K. M. Kadish, K. M.
Smith and R. Guilard, Academic Press, New York, 2000, vol. 3,
pp. 347–368.
3 (a) J. K. M. Sanders, N. Bampos, Z. Clyde-Watson, S. L. Darling,
J. C. Hawley, H.-J. Kim, C. C. Mak and S. J. Webb, in The porphyrin
handbook, ed. K. M. Kadish, K. M. Smith and R. Guilard, Academic
Press, New York, 2000, vol. 3, pp. 1–48; (b) E. Stulz, S. M. Scott,
Y.-F. Ng, A. D. Bond, S. J. Teat, S. L. Darling, N. Feeder and
J. K. M. Sanders, Inorg. Chem., 2003, 42, 6564–6574.
4 For examples see: (a) H. Sugimoto, T. Kimura and S. Inoue, J. Am.
Chem. Soc., 1999, 121, 2325–2326; (b) Y. Hirai, T. Aida and S. Inoue,
J. Am. Chem. Soc., 1989, 111, 3062–3063; (c) K. Konishi, K. Makita,
T. Aida and S. Inoue, J. Chem. Soc., Chem. Commun., 1988, 643–645;
(d) S. Asano, T. Aido and S. Inoue, Macromolecules, 1985, 18,
2057–2061.
Fig. 5 Ball and stick representation of 6. The four peripheral aluminium
porphyrins are displayed as separate colours (purple, orange, yellow and
green). Hydrogens, solvent and coordinated DMSO solvent molecules
have been omitted for clarity.
5 P. P. Kumar and B. G. Maiya, New J. Chem., 2003, 27, 619–625.
6 T. N. Lomova, S. V. Zaitseva, O. V. Molodkina and T. A. Ageeva,
Russ. J. Coord. Chem., 1999, 25, 397–401.
7 J. C. Hawley, N. Bampos, R. J. Abraham and J. K. M. Sanders, Chem.
Commun., 1998, 661–662. For a review of tin(IV) porphyrins see:
D. P. Arnold and J. Blok, Coord. Chem. Rev., 2004, 248, 299–319.
8 Y. Iseki, E. Watanabe, A. Mori and S. Inoue, J. Am. Chem. Soc., 1993,
115, 7313–7317.
Single crystal X-ray analysis also revealed the desired structure,
6, as shown in Fig. 5.{ The molecule sits on a crystallographic
centre of symmetry positioned at the centre of the central
porphyrin ring. The four aluminium centres form a rectangle with
˚
sides measuring approximately 16.3 and 15.0 A. Each
aluminium(III) porphyrin has a coordinated DMSO bound
through the oxygen.
9 J. R. Miller and G. D. Dorough, J. Am. Chem. Soc., 1952, 74,
3977–3981.
In conclusion, we have shown that aluminium(III) porphyrin–
benzoate complexes are robust and the relative stabilities can be
predicted through a Hammett plot. These complexes are stable in
the presence of water and do not require an inert atmosphere.
Aluminium(III) porphyrins effectively have two independent faces
10 T. Aido and S. Inoue, J. Am. Chem. Soc., 1983, 105, 1304–1309.
11 In order to get around the insolubility of porphyrin 5, the reaction can
either be heated or a few drops of DMSO can be added.
12 C. C. Mak, N. Bampos, S. L. Darling, M. Montalti, L. Prodi and
J. K. M. Sanders, J. Org. Chem., 2001, 66, 4476–4486.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 3087–3089 | 3089