Template controlled self-assembly of bidentate phosphine complexes with hemilabile coordination behaviour

Article abstract of DOI:10.1039/b616102b

Template-assisted self-assembly of ditopic catechol phosphines creates complexes containing a chelating diphosphine ligand, which display hemilabile coordination properties with prospects for applications in catalysis. The Royal Society of Chemistry.

Full text of DOI:10.1039/b616102b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Template controlled self-assembly of bidentate phosphine complexes
with hemilabile coordination behaviour{
Samir Chikkali, Dietrich Gudat* and Mark Niemeyer
Received (in Cambridge, UK) 6th November 2006, Accepted 27th November 2006
First published as an Advance Article on the web 12th December 2006
DOI: 10.1039/b616102b
ligand. Here, we describe the feasibility of accessing such species by
self-assembly on a main-group template, and demonstrate that the
bisphosphine ligand in the assembly formed shows hemilabile
behaviour.
Template-assisted self-assembly of ditopic catechol phosphines
creates complexes containing a chelating diphosphine ligand,
which display hemilabile coordination properties with prospects
for applications in catalysis.
In view of the existence of stable spirocyclic catecholate com-
plexes of tin9 we began to study reactions of two molar equivalents
of the phosphine L1 with [PdCl2(COD)] (1) and SnCl4 or
Me2SnCl2, respectively. The reaction proceeded readily in DMF–
NEt3 via self-assembly to afford the complexes 2a,b (Scheme 2)
which were isolated in moderate to good yield.{ Both products
were characterised by analytical and spectroscopic data and 2a as
well by a single-crystal X-ray diffraction study.§ The formation of
2a,b was also observed when the reaction was carried out with an
excess, or less than the stoichiometric quantity, of 1.
Chelating bidentate phosphines are extremely useful ligands in
catalysis due to their ability to enforce a rigid coordination
environment at the metal and induce thus improved regio- and
enantioselectivity.1 In order to circumvent laborious syntheses
which are often required to construct bidentate ligands with
covalent backbones, lately a new concept based on the self-
assembly of monodentate fragments in the coordination sphere of
a metal atom was introduced.2 The ligand backbone is hence
formed via non-covalent interactions such as hydrogen bonding
between fragments with complementary binding motifs,2,3 or by
the assembly of two fragments on a template. Available strategies
for the latter include formation of coordinative bonds to a cataly-
The crystals of 2a (Fig. 1) contain discrete complexes that lie on
a crystallographic C2 axis along with disordered solvent molecules
(CH2Cl2). The tin atom features a distorted octahedral coordina-
tion by the oxygen atoms of two chelating catechol units and two
chlorides. The oxygen atoms O1/O19 act as m2-bridging ligands
towards Sn and Pd, with the Sn–O1 bonds in the resulting planar
SnO2Pd-ring being distinctly longer (2.17 s) than the Pd–O1
bonds (2.09 s ) or the Sn–O2 bonds to the terminal oxygen atoms
(2.03 s). The square-planar coordination sphere at Pd is
completed by the two phosphorus atoms. The constraints of the
polycyclic structure inflict a strictly pyramidal coordination at
the m2-bridging oxygen atoms (sum of angles 323u) with a
tically inactive metal ion such as Zn2+ 2,4 or anion sequestering.5
,
We have recently reported6 the synthesis of the phosphine L1
(Scheme 1) which has the potential to act as a ditopic ligand with a
hard catecholate and a soft phosphine binding site. Some time ago
it has been demonstrated that the disposition of a related catechol
phosphine L2 for site-selective binding of hard and soft metal ions
can be exploited for the designed synthesis of supramolecular
heterometallic clusters7 or the immobilisation of catalytically active
complexes on metal oxide supports.8 In view of the closer
proximity of the donor centres and higher conformational
flexibility of L1 as compared to rigid L2 we wondered if assembly
of the catechol moieties of two molecules of L1 to a suitable
template and simultaneous binding of the phosphine units to a soft
metal might allow the formation of molecular complexes in which
the template-supported phosphine fragments mimic a chelate
Scheme 1 Catechol phosphines.
Institut fu¨r Anorganische Chemie, Universita¨t Stuttgart, Stuttgart,
Germany. E-mail: gudat@iac.uni-stuttgart.de; Fax: +49 71168564241;
Tel: +49 711 68564186
{ Electronic supplementary information (ESI) available: Descriptions of
the syntheses and spectroscopic data of 2a,b; details of catalytic and
computational studies. See DOI: 10.1039/b616102b
Scheme 2 Molecular bimetallic chelate complexes by self-assembly. R =
Cl (2a), Me (2b);
L = solvent (DMF or formamide). Reagents
and conditions: (i) 1 equiv. R2SnCl2, 1 equiv. [(cyclooctadiene)PdCl2],
DMF–NEt3.
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 981–983 | 981

View Article Online
The 1H NMR signals of the CH2 and phenyl protons of 2a are
likewise broad at 30 uC and decoalesce at 250 uC into two sets
of signals, indicating that the geminal protons and phenyl rings
in the Ph2PCH2-units become chemically inequivalent. The
1
analogous H NMR signals of 2a9 (which were identified in 2D
NMR spectra) did not split up but it is currently unclear if this
results from the failure to reach the slow-exchange regime
(because of accidental near-degeneracy of the chemical shifts
at exchanging sites), or if the geminal protons and phenyl rings
remain equivalent.
Interpretation of the dynamic phenomena observed is feasible
by assuming that 2a9 arises from coordination of extra solvent
molecules to 2a in a process which is favoured by enthalpy and
disfavoured by entropy. As the available experimental data allow
no unequivocal assignment of the molecular structure of 2a9, we
performed DFT calculations11 in order to model the interaction of
2a with formamide in the gas phase and decide if the new donors
bind preferably to tin or to palladium. The results of these studies
suggested that formation of a macrocyclic complex with trans-
configuration at Pd (cf. Scheme 2 and Fig. 2) is favoured by
3.6 kcal mol21 over coordination of formamide to tin. Since
all attempts to locate further stationary points on the energy
hypersurface (e.g. representing complexes with cis-configuration at
Pd or trans-configuration at Sn) failed, we conclude that the found
molecular structure is most likely to represent that of the transient
complex 2a9. The failure to resolve the diastereotopic CH2 protons
in the low-temperature 1H NMR spectra and the high 31P
chemical shift of 2b compared to [PdCl2(L1)2] impose no plain
contradictions to this assignment as it cannot be ruled out that the
former is caused by rapid dynamic racemisation of the chiral
coordination environment at tin (which is facilitated by the
dissociation of the O-donors from Pd and removes the
inequivalence between diastereotopic geminal sites)12 or accidental
near-degeneracy of chemical shifts whereas the latter is attributable
to the zwitterionic nature of 2b.
Fig. 1 Molecular structure of 2a in the crystal (H-atoms omitted; 50%
probability thermal ellipsoids; atom labels with a prime (9) character
indicate that these atoms are at equivalent position (2 2 x, y, 1/2 2 z));
selected distances (s) and angles (u): Pd–O1 2.088(2), Pd–P 2.2519(10),
O1–Sn 2.171(2), O2–Sn 2.033(3), Sn–Cl 2.3901(10); O1–Pd–O19 76.95(14),
O1–Pd–P 168.55(7), P–Pd–P9 98.78(5), O1–Sn–O19 73.49(13), Cl–Sn–Cl9
104.28(5), O2–Sn–O1 103.58(11), O2–Sn–Cl 91.46(9).
trans-arrangement of the exocyclic O–C bonds at the SnO2Pd-ring
and a cis-arrangement of the phosphine moieties at Pd. The same
constraints are presumably responsible for the substantial bond
angle distortions around tin and palladium (cf. Fig. 1) which
suggest that the assembly of the complex goes along with the build-
up of some ring strain.
The 31P NMR spectra of 2a,b at 30 uC in DMF show
broadened lines (d 76.4 (2a), 74.6 (2b)) at substantially higher
chemical shifts than in the complex [trans-PdCl2(L1)2] (d 19.1)
which is accessible from reaction of L1 and 1; similar deshieldings
had previously been observed for chelate complexes of related
phosphinoalcohol ligands.10 The observation of tin-satellites in the
31P{1H} and a triplet in the 119Sn{1H} NMR spectrum of 2b
(d 2199.4, JSn,P = 15 Hz) disclosed that the bimetallic aggregate
remains kinetically stable and does not undergo ligand exchange (a
119Sn NMR signal of 2a was unobservable in solution owing to its
much lower solubility but the 119Sn CP-MAS NMR spectrum of a
solid sample shows one isotropic line at diso = 2455).
In view of the easy replacement of the catechol oxygens from
Pd, 2a can be viewed as a Pd-complex with a hemilabile ligand
that can switch between tetradentate O2P2 and bidentate P2
The 31P NMR signals of 2a,b sharpened eventually upon
cooling to 230 uC. Surprisingly, further cooling of a solution of 2a
lead to the appearance of a new 31P NMR signal (d 55.5) which
grew in intensity when the temperature was lowered to 255 uC. All
changes were reversible upon warming to ambient temperature.
We attribute the new signal to a second species 2a9 which is,
according to 2D 31P EXSY spectra, in dynamic exchange with 2a.
Assuming that this exchange can be described as a pseudo-
unimolecular reaction 2a O 2a9, evaluation of the temperature
dependence of the equilibrium constant K = [2a9]/[2a] yields
values of DH = 25.4 kcal mol21 and DS = 226 cal mol21 K21
.
Careful analysis of the results of low-temperature NMR studies
of solutions of 2b suggested that similar dynamic equilibria are
likewise present even if the unambiguous identification of a
transient species was in this case precluded due to its much lower
concentration.
Fig. 2 Wire-model of the energy optimised molecular structure of the
solvent adduct 2a?2H2NCHO (at the b3lyp/3-21g* level of theory).
982 | Chem. Commun., 2007, 981–983
This journal is ß The Royal Society of Chemistry 2007

View Article Online
1 Z. Freixa and P. W. N. M. van Leeuwen, Dalton Trans., 2003, 1890.
2 Selected reviews:B. Breit, Angew. Chem., 2005, 117, 6976, (Angew.
Chem., Int. Ed, 2005, 44, 6816); A. J. Sandee and J. N. H. Reek, Dalton
Trans., 2006, 3385.
3 B. Breit and W. Seiche, J. Am. Chem. Soc., 2003, 125, 6608; B. Breit and
W. Seiche, Angew. Chem., 2005, 117, 1666, (Angew. Chem., Int. Ed.,
2005, 44, 1640).
4 V. F. Slagt, P. W. N. M. van Leeuwen and J. N. H. Reek, Angew.
Chem., 2003, 115, 5781, (Angew. Chem., Int. Ed., 2003, 42, 5619);
V. F. Slagt, P. W. N. M. van Leeuwen and J. N. H. Reek, Chem.
Commun., 2003, 2474; J. M. Takacs, D. S. Reddy, S. A. Moteki, D. Wu
and H. Palencia, J. Am. Chem. Soc., 2004, 126, 4494; V. F. Slagt,
M. Ro¨der, P. C. J. Kamer, P. W. N. M. van Leeuwen and J. N. H. Reek,
J. Am. Chem. Soc., 2004, 126, 4056; M. J. Wilkinson, P. W. N. M. van
Leeuwen and J. N. H. Reek, Org. Biomol. Chem., 2005, 3, 2371.
5 P. A. Duckmanton, A. J. Blake and J. B. Love, Inorg. Chem., 2005, 44,
7708; L. K. Knight, Z. Freixa, P. W. N. M. van Leeuwen and
J. N. H. Reek, Organometallics, 2006, 25, 954.
6 S. Chikkali and D. Gudat, Eur. J. Inorg. Chem., 2006, 3005.
7 X. Sun, D. W. Johnson, D. L. Caulder, R. E. Powers, K. N. Raymond
and E. H. Wong, Angew. Chem., 1999, 111, 1386, (Angew. Chem. Int.
Ed., 1999, 38, 1303); X. Sun, D. W. Johnson, D. L. Caulder,
R. E. Powers, K. N. Raymond and E. H. Wong, J. Am. Chem. Soc.,
2001, 123, 2752.
coordination. As it appears likely that the solvent molecules in
2a9 are further displaceable by other donors, a similar catalytic
activity as for other Pd-phosphine complexes is anticipated. A
first confirmation of this hypothesis was obtained from the
successful application of 2a,b as catalysts in the Sonogashira
coupling13 of para-iodonitrobenzene with phenyl acetylene. The
stability of the bimetallic complex under catalytic conditions was,
in the case of 2b, indicated by the finding that the catalyst could be
recovered during the work-up and reused for further catalytic
transformations.
In summary, the synthesis of discrete complexes containing a
hard main-group and a soft transition metal by self-assembly of
simple metal precursors and bifunctional phosphine–catecholate
ligands has been demonstrated. The products are distinguished
from known heterobimetallic complexes of phosphinoalcohols
in requiring neither the dedicated synthesis of a polydentate chelate
ligand14 nor the need for introducing the metal atoms in separate
reaction steps.15 The coordination of the soft Pd-atom by
m-bridging catecholate oxygen atoms that are easily displaceable
by donor solvents can be viewed in terms of hemilabile
coordination properties. Studies of the impact of the hemilabile
behaviour on the catalytic properties of such complexes as well as
the assembly of complexes containing other main-group and
transition-metal atoms are currently under investigation.
8 N. T. Lucas, J. M. Hook, A. M. McDonagh and S. B. Colbran, Eur. J.
Inorg. Chem., 2005, 496–503; N. T. Lucas, A. M. McDonagh,
I. G. Dance, S. B. Colbran and D. C. Craig, Dalton Trans., 2006,
680.
9 C. Lamberth, J. C. Machell, D. M. P. Mingos and T. L. Stolberg,
J. Mater. Chem., 1991, 1, 775, and references therein.
10 D. J. Brauer, P. Machnitzki, T. Nickel and O. Stelzer, Eur. J. Inorg.
Chem., 2000, 65.
Notes and references
11 All computations were performed at the B3LYP/3-21g* level of theory
with the Gaussian program suite: M. J. Frisch, G. W. Trucks,
H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman,
V. G. Zakrzewski, J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin, M. C. Strain,
O. Farkas, J. Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci,
C. Pomelli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson,
P. Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz,
A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-
Laham, C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. G. Johnson, W. Chen, M. W. Wong, J. L. Andres,
M. Head-Gordon, E. S. Replogle and J. A. Pople, GAUSSIAN 98
(Revision A.7), Gaussian, Inc., Pittsburgh, PA, 1998. The computed
structures and energies of 3 and its formamide adducts are listed as
ESI{.
12 Although the dynamic stereochemistry of Sn-catecholates has not been
studied in detail it is known that catechol phosphates may racemise very
easily: cf. J. Lacour, C. Ginglinger, C. Grivet and G. Bernardinelli,
Angew. Chem., 1997, 109, 660, (Angew. Chem., Int. Ed. Engl., 1997, 36,
608), and references therein.
13 K. Sonogashira, in Comprehensive Organic Synthesis, ed. B. M. Trost
and I. Fleming, Pergamon, Oxford, 1991, vol. 3, p. 551; see also: J. Li,
X.-D. Zhang and Y.-Y. Xie, Eur. J. Org. Chem., 2005, 4256.
14 A. Kless, C. Lefeber, A. Spangenberg, R. Kempe, W. Baumann, J. Holz
and A. Bo¨rner, Tetrahedron, 1996, 52, 14599.
{ Experimental procedure: A mixture of L1 (600 mg, 1.95 mmol), SnCl4
(1.1 ml, 0.97 mmol), 1 (280 mg, 0.97 mmol) and NEt3 (0.8 ml, 5.7 mmol) in
dry DMF (20 ml) was stirred for 24 h at room temperature. The red
suspension was filtered through Celite. Solvents were evaporated in
vacuum and the residue washed with CH2Cl2 (10 ml) and dissolved in a
small amount of DMF. The same volume of CH2Cl2 and, finally, small
portions of diethyl ether were added until a precipitate began to form. The
mixture was stored overnight at 4 uC to yield a microcrystalline solid which
was collected by filtration and dried in vacuum for 4 h at 70 uC to yield
40% of 2a, mp 214 uC (decomp.). Elemental analysis (%). Calc. for
2a?DMF: C 50.16, H 3.80, N 1.43. Found: C 49.92, H 3.77, N 1.58. 2b was
obtained by an analogous procedure from L1, 1and Me2SnCl2, and isolated
as red crystals (mp 310 uC (decomp.)) in 95% yield. Elemental analysis (%).
Calc. for 2b?DMF: C 54.78, H 4.81, N 1.49. Found: C 54.68, H 4.78, N
1.41. Detailed descriptions of synthetic procedures and characterisation
data are available as ESI.{
§ Crystal structure analysis: single crystals of a solvate 2a?3CH2Cl2 were
obtained from DMF–CH2Cl2 (4 : 1); Siemens P4 diffractometer, T =
173(2) K, Mo-Ka radiation (l = 0.710 73 s), SHELX97 for structure
solution (direct methods) and refinement (full-matrix, least squares refined
against F2). The positions of the hydrogen atoms were refined with a riding
model; the contribution of the severely disordered CH2Cl2 molecules in
solvent accessible cavities of the structure was eliminated from the reflection
data, using the BYPASS method16 as implemented in the SQUEEZE
routine of the PLATON98 package; orange–red crystals,
C38H30Cl2O4P2PdSn; M = 908.56; crystal size 0.40 6 0.15 6 0.08 mm;
monoclinic, space group P2/c (no. 13); a = 12.227(2), b = 11.707(2), c =
15 G. S. Ferguson and P. T. Wolczanski, Organometallics, 1985, 4, 1601;
G. S. Ferguson and P. T. Wolczanski, J. Am. Chem. Soc., 1986, 108,
8293; G. S. Ferguson, P. T. Wolczanski, L. Pa´rka´nyi and M. C.
Zonnevylle, Organometallics, 1988, 7, 1967; O. Ku¨hl, S. Blaurock,
J. Sieler and E. Hey-Hawkins, Polyhedron, 2001, 20, 2171.
16 P. van der Sluis and A. L. Spek, Acta Crystallogr., Sect. A: Found.
Crystallogr., 1990, 46, 194.
3
21
,
˚
˚
16.475(3) A, b = 102.516(12)u, V = 2302.2(7) A , Z = 2, m = 1.149 mm
semi-empirical absorption correction by using y-scans, min/max. abs.
0.589/0.923, F(000) = 900, hmax = 28u; 6056 reflections of which 5523 were
independent; R1 = 0.045 (for I . 2s(I)), wR2 = 0.125 (all data). CCDC
616084. For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b616102b
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 981–983 | 983