Rational design and assembly of M2M′3L6 supramolecular clusters with C3h symmetry by exploiting incommensurate symmetry numbers

Article abstract of DOI:10.1021/ja0029376

A rational approach to heterometallic cluster formation is described that uses incommensurate symmetry requirements at two different metals to control the stoichiometry of the assembly. Critical to this strategy is the proper design and synthesis of hybrid ligands with coordination sites selective toward each metal. The phosphino-catechol ligand 4-(diphenylphosphino)benzene-1,2-diol (H₂L) possesses both hard catecholate and soft phosphine donor sites and serves such a role, using soft (C₂-symmetric) and hard (C₃-symmetric) metal centers. The ML₃ catecholate complexes (M = Fe^III, Ga^III, Ti^IV, Sn^IV) have been prepared and characterized as C₃-symmetry precursors for the stepwise assembly (aufbau) of heterometallic clusters. While the single-crystal X-ray structure of the Cs₂ [TiL₃] salt shows a C₁ mer-configuration in the solid-state, room-temperature solution NMR data of this and related complexes are consistent with either exclusive formation of the C₃-fac-isomer with all PPh₂ donor sites syn to each other or facile fac/mer isomerization. Coordination of these [ML₃]^2- (M = Ti^IV, Sn^IV) metallaligands via their soft P donor sites to C₂-symmetric PdBr₂ units gives exclusively pentametallic [M₂Pd₃Br₆L₆]^4- (M = Ti, Sn) clusters. These clusters have been fully characterized by spectral and X-ray structural data as C_3h mesocates with Cs⁺ or protonated 1,4-diazabicyclo-[2.2.2]octane (DABCO·H⁺) cations incorporated into deep molecular clefts. Exclusive formation of this type of supramolecular species is sensitive to the nature of the counterions. Alkali cations such as K⁺, Rb⁺, and Cs⁺ give high-yield formation of the respective clusters while NEt₃H⁺ and NMe₄⁺ yield none of the desired products. Extension of the aufbau assembly to produce related [M₂Pd₃Cl₆L₆]^4-, [M₂Pd₃I₆L₆]^4-, and [M₂Cr₃(CO)₁₂L₆]^4- (M = Ti, Sn) clusters has also been realized. In addition to this aufbau approach, self-assembly of several of these [M₂Pd₃Br₆L₆]^4- clusters from all eleven components (two M^IV, three PdBr₂, six H₂L) was also accomplished under appropriate reaction conditions.

Full text of DOI:10.1021/ja0029376

2752
J. Am. Chem. Soc. 2001, 123, 2752-2763
Rational Design and Assembly of M2M′3L6 Supramolecular Clusters
with C3h Symmetry by Exploiting Incommensurate Symmetry
Numbers§
Xiankai Sun,† Darren W. Johnson,‡ Dana L. Caulder,‡ Kenneth N. Raymond,*,‡ and
Edward H. Wong*,†
Contribution from the Departments of Chemistry, UniVersity of New Hampshire,
Durham, New Hampshire 03824, and UniVersity of California, Berkeley, California 94720-1460
ReceiVed August 7, 2000
Abstract: A rational approach to heterometallic cluster formation is described that uses incommensurate
symmetry requirements at two different metals to control the stoichiometry of the assembly. Critical to this
strategy is the proper design and synthesis of hybrid ligands with coordination sites selective toward each
metal. The phosphino-catechol ligand 4-(diphenylphosphino)benzene-1,2-diol (H2L) possesses both hard
catecholate and soft phosphine donor sites and serves such a role, using soft (C2-symmetric) and hard (C3-
symmetric) metal centers. The ML3 catecholate complexes (M ) FeIII, GaIII, TiIV, SnIV) have been prepared
and characterized as C3-symmetry precursors for the stepwise assembly (aufbau) of heterometallic clusters.
While the single-crystal X-ray structure of the Cs2[TiL3] salt shows a C1 mer-configuration in the solid -state,
room-temperature solution NMR data of this and related complexes are consistent with either exclusive formation
of the C3-fac-isomer with all PPh2 donor sites syn to each other or facile fac/mer isomerization. Coordination
of these [ML3]2- (M ) TiIV, SnIV) metallaligands via their soft P donor sites to C2-symmetric PdBr2 units
gives exclusively pentametallic [M2Pd3Br6L6]4- (M ) Ti, Sn) clusters. These clusters have been fully
characterized by spectral and X-ray structural data as C3h mesocates with Cs+ or protonated 1,4-diazabicyclo-
[2.2.2]octane (DABCO‚H+) cations incorporated into deep molecular clefts. Exclusive formation of this type
of supramolecular species is sensitive to the nature of the counterions. Alkali cations such as K+, Rb+, and
+
Cs+ give high-yield formation of the respective clusters while NEt3H+ and NMe4 yield none of the desired
products. Extension of the aufbau assembly to produce related [M2Pd3Cl6L6]4-, [M2Pd3I6L6]4-, and
[M2Cr3(CO)12L6]4- (M ) Ti, Sn) clusters has also been realized. In addition to this aufbau approach, self-
assembly of several of these [M2Pd3Br6L6]4- clusters from all eleVen components (two MIV, three PdBr2, six
H2L) was also accomplished under appropriate reaction conditions.
Introduction
the intrinsic beauty of these clusters and the challenges in ligand
design and cluster synthesis. However, many of these clusters
exhibit fascinating properties as well, such as guest binding,11-16
The synthesis of high-symmetry metal-ligand clusters is an
effective way to build large molecular constructs from small
subunits. These clusters form spontaneously via self-assembly
reactions from a mixture of labile components to yield the most
thermodynamically stable product. Among a wide array of
multinuclear supramolecular clusters, helicates and their meso
counterparts (mesocates), rings, catenanes,1,2 tetrahedra,3 cyl-
inders, cages,4 octahedra, cubes, and icosahedra,5 have been
generated in this manner.6-10 Much of this attention results from
(6) For all but the most recent papers on these molecules see the most
thorough review in this field: Leininger, S.; Olenyuk, B.; Stang, P. J. Chem.
ReV. 2000, 100, 853. For applications see pp 902-903.
(7) Piguet, C.; Bernardinelli, G.; Hopfgartner, G. Chem. ReV. 1997, 97,
2005 and references therein.
(8) Lehn, J.-M. Supramolecular Chemistry: Concepts and PerspectiVes;
VCH: Weinheim, 1995, and references therein.
(9) Raymond, K. N.; Caulder, D. L.; Powers, R. E.; Beissel, T.; Meyer,
M.; Kersting, B. Proc. 40th Robert A. Welch Found. Chem. Res. 1996, 40,
115 and references therein.
(10) (a) Caulder, D. L.; Raymond, K. N. J. Chem. Soc., Dalton Trans.
1998, 1185 and references therein. (b) Saalfrank, R. W.; Demleitner, B. In
Transition Metals in Supramolecular Chemistry; Sauvage, J.-P., Ed.; John
Wiley & Sons Ltd.: Chichester, England, 1999; Perspectives in Supramo-
lecular Chemistry, Vol. 5, pp 1-51 and references therein.
(11) Fleming, J. S.; Mann, K. L. V.; Carraz, C.-A.; Psillakis, E.; Jeffery,
J. C.; McCleverty, J. A.; Ward, M. D. Angew. Chem., Int. Ed. 1998, 37,
1279.
(12) Gokel, G. W. Molecular Recognition: Receptors for Cationic
Guests; Pergamon: New York, 1996; Vol. 1 and references therein.
(13) Aoyagi, M.; Biradha, K.; Fujita, M. J. Am. Chem. Soc. 1999, 121,
7457.
* Authors to whom correspondence should be addressed. E-mail:
ehw@christa.unh.edu, raymond@socrates.berkeley.edu.
§ Paper No. 19 in the series Coordination Number Incommensurate
Cluster Formation. For the previous paper see: Terpin, A. J.; Ziegler, M.;
Johnson, D. W.; Raymond, K. N. Angew. Chem. Int. Ed. 2000, 39, 157-
160.
† University of New Hampshire, Durham.
‡ University of California, Berkeley.
(1) Sauvage, J. P. Acc. Chem. Res. 1998, 31, 611 and references therein.
(2) Fujita, M. Acc. Chem. Res. 1999, 32, 53 and references therein.
(3) Fox, O. D.; Drew, M. G. B.; Beer, P. D. Angew. Chem., Int. Ed.
2000, 39, 136.
(4) Takeda, N.; Umemoto, K.; Yamaguchi, K.; Fujita, M. Nature 1999,
398, 794.
(14) Kusukawa, T.; Fujita, M. J. Am. Chem. Soc. 1999, 121, 1397.
(15) Parac, T.; Caulder, D. L.; Raymond, K. N. J. Am. Chem. Soc. 1998,
120, 8003.
(5) Olenyuk, B.; Whiteford, J. A.; Fechtenkotter, A.; Stang, P. J. Nature
1999, 398, 796.
10.1021/ja0029376 CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/22/2001

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2753
chiral resolution,17-21 dynamic interconversion of assemblies,22-25
mechanical coupling between metal centers,26-28 catalysis,29 and
magnetism,30-33 inter alia.6
We have developed a predictive design strategy10 resulting
in the synthesis of various high-symmetry coordination clusters
including M2L3 helicates26,34,35 and mesocates,22,36 M4L6
25,28,37,38
and M4L4 tetrahedra,39 M6L6 and M8L8 cylinders, and M8L6
40
octahedra.41 These examples all focus on the coordination of
three bidentate chelators to a tri- or tetravalent metal ion in a
pseudooctahedral fashion at the apexes of the clusters. This
coordination generates local 3-fold symmetry at these metal
centers. These chelators are contained in a rigid, symmetric
multi(bidentate)-ligand, which supplies the other symmetry
elements of the cluster (2-fold, 3-fold, or mirror plane). By
simultaneously fulfilling the symmetry requirements of both the
ligand and metal centers, discrete high-symmetry clusters are
generated under thermodynamic control.
Figure 1. Cartoon showing the 90° offset, incommensurate interaction
of a 3-fold symmetry element (blue spheres), and a C2 axis/mirror plane
(red spheres) linked by a hybrid ligand L (black) to generate a M2M′3L6
cluster.
Recently, we communicated the stepwise assembly of two
M2M′3L6 mesocates in which a second type of metal ion, rather
than a symmetric ligand, supplies the mirror plane symmetry
in the cluster.36 In principle, the ligand forms part of an
asymmetric unit of the cluster and must have two different
interaction sites (e.g., one hard and one soft donor) that can
each preferentially interact with one of the metal ions over the
other. Here we present details of both this stepwise assembly
as well as the spontaneous self-assembly from all 11 components
of a series of mesocates from such an asymmetric “hybrid”,
wherein two different metals are used to generate the two
incommensurate symmetry elements of the clusters (Figure 1).
This use of incommensurate metal-ligand interactions forms
the basis for our design strategy. Full synthetic details and
discussions on several single-crystal X-ray structures are also
presented.
(16) Saalfrank, R. W.; Burak, R.; Breit, A.; Stalke, D.; Herbst-Irmer,
R.; Daub, J.; Porsch, M.; Bill, E.; Mu¨ther, M.; Trautwein, A. X. Angew.
Chem., Int. Ed. 1994, 33, 1621.
(17) Knof, U.; von Zelewsky, A. Angew. Chem., Int. Ed. 1999, 38, 303.
(18) Kra¨mer, R.; Lehn, J.-M.; DeCian, A.; Fischer, J. Angew. Chem.,
Int. Ed. 1993, 32, 703.
(19) Enemark, E. J.; Stack, T. D. P. Angew. Chem., Int. Ed. 1998, 37,
932.
(20) Stang, P. J.; Olenyuk, B.; Muddiman, D. C.; Smith, R. D.
Organomet. 1997, 16, 3094.
(21) Masood, M. A.; Enemark, E. J.; Stack, T. D. P. Angew. Chem., Int.
Ed. 1998, 37, 928.
(22) Xu, J.; Parac, T.; Raymond, K. N. Angew. Chem., Int. Ed. 1999,
19, 2878.
(23) Lee, S. B.; Hwang, S.; Chung, D. S.; Yun, H.; Hong, J.-I.
Tetrahedron Lett. 1998, 39, 873.
(24) Hasenknopf, B.; Lehn, J.-M.; Kneisel, B. O.; Baum, G.; Fenske, D.
Angew. Chem., Int. Ed. 1996, 35, 1838.
(25) Scherer, M.; Caulder, D. L.; Johnson, D. W.; Raymond, K. N.
Angew. Chem., Int. Ed. 1999, 38, 1588.
(26) Kersting, B.; M. Meyer; Powers, R. E.; Raymond, K. N. J. Am.
Chem. Soc. 1996, 118, 7221.
(27) Pfeil, A.; Lehn, J.-M. Chem. Commun. 1992, 838.
(28) Beissel, T.; Powers, R. E.; Parac, T. N.; Raymond, K. N. J. Am.
Chem. Soc. 1999, 121, 4200.
(29) Sanders, J. K. M. Chem. Eur. J. 1998, 4, 1378.
(30) Waldmann, O.; Schulein, J.; Koch, A.; Muller, P.; Bernt, I.;
Saalfrank, R. W.; Andres, H. P.; Gudel, H. U.; Allenspach, P. Inorg. Chem.
1999, 38, 5879.
Results and Discussion
Ligand Design. The assembly of a mixed-metal helicate of
the type Ru2Cu3L6 was recently reported where a substitution-
inert RuII[3-(pyridin-2-yl)pyrazole]3 complex was used as the
3-fold interaction site.42 The uncoordinated nitrogen atom of
the pyrazole ring remains free to coordinate labile CuI to
generate a 2-fold axis and thus assemble a triple helicate. We
chose to explore the use of a hybrid ligand with more
distinguishing coordination sites, namely catechol and phos-
phine. Catechol is a hard chelating ligand that generates a C3
axis when coordinated to hard, tri- or tetravalent metals with
preferred octahedral coordination (e.g., AlIII, GaIII, FeIII, SnIV,
TiIV).43,44 Phosphine ligands, on the other hand, are soft donors
which can coordinate to square-planar metals (e.g., PdII or PtII)
in a trans fashion to generate a 2-fold axis or mirror plane.45,46
A properly designed hybrid ligand containing both these donors
could assemble an M2M′3L6 cluster, the smallest discrete species
that would simultaneously fulfill the two orthogonal symmetry
requirements. Molecular models (CAChe47) suggested a 4-phos-
phinocatechol ligand (Figure 2) would have the ideal geometry
for isolating the two types of coordination sites and to permit
cluster assembly.
(31) Gatteschi, D.; Caneschi, A.; Pardi, L.; Sessoli, R. Science 1994,
265, 1054.
Synthesis and Characterization of the Hybrid Ligand H2L.
The hybrid ligand H2L can be readily prepared in two steps
from 4-dichlorophosphino-veratrole (Scheme 1). The latter was
obtained from veratrole under Friedel-Crafts conditions by
using a literature synthesis.48 Grignard reaction of 4-dichloro-
phosphino-veratrole with phenylmagnesium bromide afforded
(32) Amoroso, A. J.; Jeffery, J. C.; Jones, P. L.; McCleverty, J. A.;
Thornton, P.; Ward, M. D. Angew. Chem., Int. Ed. 1995, 34, 1443.
(33) Taft, K. L.; Delfs, C. D.; Papaefthymiou, G. C.; Foner, S.; Gatteschi,
D.; Lippard, S. J. J. Am. Chem. Soc. 1994, 116, 823.
(34) Kersting, B.; Telford, J. R.; Meyer, M.; Raymond, K. N. J. Am.
Chem. Soc. 1996, 118, 5712.
(35) Caulder, D. L.; Raymond, K. N. Angew. Chem., Int. Ed. 1997, 36,
1439.
(36) Sun, X.; Johnson, D. W.; Caulder, D. L.; Powers, R. E.; Raymond,
K. N.; Wong, E. H. Angew. Chem. 1999, 38, 1303.
(37) Caulder, D. L.; Powers, R. E.; Parac, T.; Raymond, K. N. Angew.
Chem., Int. Ed. 1998, 37, 1840.
(42) Lam, M. H. W.; Cheung, S. T. C.; Fung, K.-M.; Wong, W.-T. Inorg.
Chem. 1997, 36, 4618.
(43) Pierpoint, C. G.; Buchanan, R. M. Coord. Chem. ReV. 1981, 38,
45.
(44) Pierpoint, C. G.; Lange, C. W. Inorg. Chem. 1994, 33, 331.
(45) McAuliffe, C. A.; Wilkinson, G.; Stone, F. G. A.; Abel, F. W., Eds.;
Pergamon: Oxford, 1987; Vol. 2, Chapter 14.
(46) Levason, W.; Hartley, F. R., Ed.; Wiley: New York, 1990; Vol. 1,
pp Chapter 16.
(38) Beissel, T.; Powers, R. E.; Raymond, K. N. Angew. Chem., Int. Ed.
1996, 35, 1084.
(39) Bru¨ckner, C.; Powers, R. E.; Raymond, K. N. Angew. Chem., Int.
Ed. 1998, 37, 1837.
(40) Johnson, D. W.; Xu, J.; Saalfrank, R. W.; Raymond, K. N. Angew.
Chem., Int. Ed. 1999, 38, 2882.
(41) Xu, J.; Raymond, K. N. 2000, manuscript in preparation.
(47) CAChe, Version 4.0; Oxford Molecular Group, Inc., U.S.A., 1997.
(48) Miles, J. A.; Beeny, M. T.; Ratts, K. W. J. Org. Chem. 1975, 40,
343.

2754 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Sun et al.
Figure 2. The 4-phosphino-catechol hybrid ligand.
Scheme 1
Figure 3. Depiction of the fac- and mer-isomers of Λ-[ML3]n-
.
H2L‚HBr to give tris-catecholate complexes without interference
from the P-donor groups. Thus [ML3]3- (M ) Fe, Ga) and
[ML3]2- (M ) Ti, Sn) were synthesized in good yields
according to Scheme 1. These products were characterized by
elemental and spectral analyses. Their IR spectra all show a
common strong catecholate C-O band at ca. 1260 cm-1
,
4-diphenylphosphino-veratrole in 70-80% yield as a white
solid. Demethylation with 48% aqueous HBr yielded a white
precipitate of H2L‚HBr. This was recrystallized from ethanol
with 48% aqueous HBr to give analytically pure H2L‚HBr in
70% yield. The proton NMR spectrum of this ligand in d7-DMF
revealed three well-resolved multiplets for the catechol pro-
tons: two doublet of doublets and a higher field triplet of
doublets. Assignment of these as H(6), H(3), and H(5) respec-
tively was made based on spectral simulation and selective
irradiation experiments (Figure 2). The catechol resonances in
the 13C{1H} NMR spectrum were also well-resolved with the
six unique ring carbons ranging from δ 147.7 to 116.6 (see
Experimental Section). 31P-13C couplings were observable for
all resonances except that of C(1) which is para to the PPh2
substituent.
Syntheses and Characterization of [ML3]n- Complexes.
The isolation of the soft phosphine donor site from the hard
dianionic catecholate chelate, crucial to our ligand design,
ensures that the ligand will bridge two metal centers when fully
coordinated, rather than forming a single metal complex using
both donor sets. While hard metal cations can exclusively bind
at the catecholate site, softer metal centers may not discriminate
sufficiently between phosphine and catecholate coordination
modes. Thus PdII, for example, reacted with H2L to give an
insoluble and intractable brown solid. This is presumably an
oligomeric or polymeric product containing indiscriminate P
and catecholate coordination to each PdII center. Exclusive
bonding of soft metals to the P site may eventually be realized
indirectly by first forming the 4-PPh2-veratrole complex with
subsequent deprotection of the methoxy groups under mild
conditions.49-51 Thus we opted to form the catecholate complex
first. Harder tri- and tetravalent metals such as AlIII, GaIII, FeIII,
TiIV, and SnIV reacted smoothly under basic conditions with
diagnostic of the coordinated catecholate groups.52,53 While the
FeIII complex is dark red in color, the TiIV complex is
red-orange, and the GaIII and SnIV complexes are both white
solids. The electronic spectrum of the Cs3FeL3 complex in
methanol solution exhibited a LMCT λmax at 490 nm (ꢀ ) 6050
L cm-1 mol-1).54,55 EPR spectra of the Cs3FeL2 complex in
the solid state or in methanol solution at room temperature both
exhibited a major g ) 4.25 signal typical of high-spin ferric
catecholate complexes.55 The trianionic complexes were con-
siderably more air-sensitive than the dianionic complexes, with
partial air oxidation of the PPh2 groups evident after overnight
exposure.
As anticipated, the PPh2 groups were found to remain
uncoordinated in accord with the observed 31P{1H} singlet
chemical shifts (δ -2.0 to -6.0) in the three diamagnetic GaIII,
TiIV, and SnIV [ML3]n- complexes.56,57 Further, room temper-
ature solution 1H and 13C{1H} NMR data of these products all
exhibited a single set of ligand catecholate and phenyl reso-
nances indicative of C3 symmetry. This is consistent with either
exclusive formation of the fac-isomer or rapid isomerization
involving the mer-isomer of C1 symmetry (Figure 3). All three
catecholate proton resonances shifted upfield upon metal
complexation by 0.26 ppm up to 0.73 ppm, as previously
observed for GaIII and RhIII tris-catecholate complexes.58,59
Significant 13C coordination shifts were found for the catechol
(52) Sofen, S. R.; Abu-Dari, K.; Freyberg, D. P.; Raymond, K. N. J.
Am. Chem. Soc. 1978, 100, 7882.
(53) Griffith, W. P.; Pumphrey, C. A.; Rainey, T.-A. 1986, 1125.
(54) Karpishin, T. B.; Gebhard, M. S.; Solomon, E. I.; Raymond, K. N.
J. Am. Chem. Soc. 1991, 113, 2977.
(55) Avdeef, A.; Sofen, S. R.; Bregante, T. L.; Raymond, K. N. J. Am.
Chem. Soc. 1978, 100, 5362.
(56) Gorenstein, D. G. Prog. NMR Spectrosc. 1983, 16, 1.
(57) Quin, L. D.; Verkade, J. G. Phosphorus-31 NMR Spectral Properties
in Compound Characterization and Structural Analysis; VCH: New York,
1994.
(49) Sembering, S. B.; Colbran, S. B.; Craig, D. C. Inorg. Chem. 1995,
34, 761.
(50) Sembering, S. B.; Colbran, S. B.; Craig, D. C.; Scudder, M. L. J.
Chem. Soc., Dalton Trans. 1995, 3731.
(51) Sembiring, S. B.; Colbran, S. B.; Craig, D. C. J. Chem. Soc., Dalton
Trans. 1999, 1543.
(58) McArdle, J. V.; Sofen, S. R.; Cooper, S. R.; Raymond, K. N. Inorg.
Chem. 1978, 17, 3075.
(59) Linas, M.; Wilson, D. M.; Neilands, J. B. Biochemistry 1973, 12,
3835.

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2755
Table 1. Selected Bond Distances (Å) and Bond Angles (deg) for
Cs2[Ti(L)3]a
Cs(1)-O(2)
Cs(1)-O(2′)
Cs(1)-O(6)
Cs(1)-O(7)
Cs(1)-O(8)
Cs(2)-O(3)
Cs(2)-O(3′)
Cs(2)-O(5)
Cs(2)-O(7)
Cs(2)-O(8)
Ti(1)-O(1)
Ti(1)-O(2)
Ti(1)-O(3)
Ti(1)-O(4)
Ti(1)-O(5)
Ti(1)-O(6)
P(3)-C(61)
P(4)-C(49)
3.097(8) O(1)-C(1)
3.305(7) O(2)-C(2)
3.051(8) O(3)-C(19)
1.35(2)
1.34(1)
1.36(1)
1.35(2)
1.34(2)
1.37(2)
1.26(2)
1.10(4)
1.83(1)
1.84(1)
1.87(1)
1.81(1)
1.82(1)
1.82(1)
1.84(2)
1.86(2)
1.92(3)
1.92(3)
3.21(1)
3.11(1)
O(4)-C(20)
O(5)-C(37)
3.164(8) O(6)-C(38)
3.272(9) O(7)-C(66)
3.034(9) O(8)-C(69)
3.30(1)
P(1)-C(5)
3.089(9) P(1)-C(7)
1.936(7) P(1)-C(13)
1.996(9) P(2)-C(22)
1.963(9) P(2)-C(25)
1.964(7) P(2)-C(31)
1.957(9) P(3)-C(41)
1.962(9) P(3)-C(43)
1.87(2)
1.90(2)
P(4)-C(40)
P(4)-C(55)
O(1)-Ti(1)-O(2)
79.8(3)
O(1)-Ti(1)-O(3)
O(1)-Ti(1)-O(5)
O(2)-Ti(1)-O(3)
88.5(3)
92.8(3)
98.6(4)
O(1)-Ti(1)-O(4) 162.8(4)
O(1)-Ti(1)-O(6) 103.1(3)
O(2)-Ti(1)-O(4)
O(2)-Ti(1)-O(6)
O(3)-Ti(1)-O(5)
89.3(3)
90.0(4)
92.0(4)
O(2)-Ti(1)-O(5) 166.8(3)
O(3)-Ti(1)-O(4) 80.0(3)
O(3)-Ti(1)-O(6) 166.7(4)
O(4)-Ti(1)-O(6) 90.1(3)
O(4)-Ti(1)-O(5) 100.2(3)
Figure 4. ORTEP of the solid-state structure of the Cs2[TiL3] complex.
For clarity only one of the two disordered PPh2 groups (P(3)) is shown
and only the oxygen atoms of coordinating solvents are shown (O(7)
and O(8)). Thermal ellipsoids are at 50% probability.
O(5)-Ti(1)-O(6) 81.1(4)
a Primes represent symmetry equivalent atoms generated by the
crystallographic inversion center.
complex. In addition, Cs(2) is coordinated by one DMF and
one disordered DMF or water solvent molecule, O(8) and O(7),
respectively. On the other hand, Cs(1) is coordinated by such a
solvent molecule from an adjacent complex. The Cs-O bond
distances range from 3.034(9) Å for Cs(2)-O(5) to 3.305(7) Å
for Cs(1)-O(2′). The intracomplex Cs‚‚‚Cs separation is 5.176-
(2) Å. The surprisingly low coordination numbers of these
cations are a direct result of the proximity of the phenyl
substituents on the appended PPh2 groups. In fact, each Cs cation
is shielded from further solvation by a phenyl ring only 3.7 Å
away. Other relevant bond distances and angles are listed in
Table 1.
Configuration of the [ML3]2- Complexes. In the absence
of steric, electronic, ion-pairing, or solvation effects, the C1 mer
isomers of [ML3]2- (M ) Ti, Sn) and [ML3]3- (M ) Fe, Al,
Ga) are statistically favored over the C3 fac isomers by a 3:1
ratio. Of these, only the fac configuration has all three
phosphino-donor groups disposed syn to each other, which is
essential for successful formation of any endohedral metal
cluster (Figure 3). While room temperature solution NMR data
are fully consistent with either exclusive presence of this high-
symmetry species in all cases or extremely facile fac/mer
equilibration, the X-ray structure of Cs2[TiL3] discussed above
revealed a mer configuration in the solid state. We examined
and discarded the possibility that the fac isomer is favored in
solution due to steric, electronic, ion-pairing, and/or solvation
factors.63
ring carbons only. Both C(1) and C(2) were shifted sizably
downfield by about 15 ppm upon metal coordination. Upfield
shifts of over 5 ppm were observed for C(3), C(4), and C(6).
Of the phenyl carbons, only the ipso-C shifted slightly,
downfield by 1.8 ppm, again confirming the uncoordinated status
of the PPh2 moiety.
X-ray Structure of the Cs2[TiL3] Complex. A single-crystal
X-ray structural study of Cs2[TiL3]‚2DMF‚0.5H2O revealed a
mer configuration in the solid state (Figure 4).60 While P(1)
and P(2) are in an anti configuration, the third phosphorus is
disordered equally over two sites, giving a mer arrangement at
either site.
The complex crystallizes as a racemic mixture of ∆ and Λ
configurations at the Ti centers. The average twist angle around
Ti is 42.3°, indicative of an intermediate octahedral/trigonal
prismatic coordination geometry. The six Ti-O bond lengths
average to 1.96 Å with only Ti-O(1) and Ti-O(2) significantly
different from the mean at 1.936(7) and 1.996(9) Å, respectively.
Of the six catecholate oxygens only O(1) is not coordinated to
a Cs cation, which may account for its short bond length to Ti.
No clear trend in the existence of any trans influence on Ti-O
or C-O bond lengths can be seen for the phenolic oxygens
meta to or trans to the PPh2 group. This is known to be a
moderately electron-withdrawing substituent.61,62 Similarly, no
meaningful differences in the catcholate C-O distances, which
average to 1.35 Å, were found.
The two cesium counterions are coordinated by catecholato
oxygens as well as by dimethylformamide (DMF) solvent
molecules. Each cation is coordinated by two catecholato
oxygens and a third catecholato oxygen from an adjacent
The alternative explanation for the observed high solution
2-
symmetry of ML3 must be facile solution fac/mer isomer-
ization processes that resulted in averaged spectra of C3
symmetry. Extensive variable-temperature 1H and 31P{1H}
studies of TiL32- and similar studies of SnL32-, including 119Sn
NMR spectroscopy in a variety of solvents, failed to yield any
unequivocal evidence for the slowing down of any fac/mer
exchange. Literature data on fluxional behavior for TiIV and
SnIV complexes with six-oxygen donors are relatively sparse.
(60) Cs2[TiL3]‚2DMF‚0.5H2O: fragment of a red/yellow block, crystal
size 0.30 × 0.07 × 0.07 mm3, FW ) 1345.73, T ) -115 °C, triclinic
space group, P1h, a ) 11.4876(2) Å, b ) 16.6820(3) Å, c ) 16.7549(4) Å,
R ) 96.759(1)°, â ) 99.298(1)°, γ ) 98.647(1)°, V ) 3099.3(1) Å3, Z )
2, R ) 0.064, Rw ) 0.083, GOF ) 2.13.
(61) van Doorn, J. Phosphorus Sulfur 1991, 62, 155.
(62) Louattani, E.; Lledos, A.; Suades, J.; Alvarez-Larena, A.; Piniella,
J. F. Organometallics 1995, 14, 1053.

2756 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Sun et al.
Fay and Lindmark64,65 reported that both TiIV tris(â-diketonate)
and bis(â-diketonate)(OR)2 complexes of unsymmetrical ligands
are fluxional down to -100 °C in methylene chloride solution.
A recent review on SnIV bidentate oxygen donor chelates
actually stated that tris(catecholate) SnIV complexes of unsym-
metrically substituted catechols (e.g. 4-cyano, 4-methyl, 4-nitro,
4-chloro, etc.) exist as fac/mer mixtures in solution according
solution spectral analyses confirmed the formation of a complex
with C3 or averaged-C3 symmetry. Specifically, a single 31P-
{1H} NMR signal at δ 22.10 and a single type of catecholate
1
group in the H and 13C{1H} NMR spectra were noted. In the
proton spectrum, the H(3) catechol proton multiplet of the
metallaligand moved dramatically downfield from δ 6.21 to 7.87
and became a broadened virtual triplet while the phenyl ortho
protons also shifted slightly downfield from δ 7.28 to 7.63. All
proton resonances were relatively broad, due probably to the
onset of dynamic behavior.
1
to H and 119Sn NMR data.66,67 This claim, however, appears
to be unfounded.68 Instead, only a single set of resonances was
obtained in all the cases studied. The implication is that the
dynamic behavior of TiIV and SnIV complexes with six oxygen
donors may indeed be quite fast on the NMR time scale even
at low temperatures. Possible acceleration of such dynamic
processes by the presence of trace amounts of protons in our
complexes can be ruled out since addition of Proton Sponge
had no effect on the resulting spectra.
For our purposes, the successful synthesis of the desired C3h
pentametallic clusters mandates that the fac isomer must be
accessible in solution whether due to its exclusive formation or
through a facile fac/mer exchange.
In the proton-decoupled 13C NMR spectrum of the product,
the most significant observation is the appearance of virtual
triplets (which were doublets in the [TiL3]2- spectrum) in
resonances for both catecholate and phenyl ring carbons. This
strongly suggests a trans-phosphine coordination mode at the
PdII centers.69 Further, the catecholate C(4) resonance shifted
upfield by over 5 ppm while a downfield coordination shift of
6 ppm is observed for C(3). Of the phenyl group carbons, similar
significant coordination shifts were found for the ipso (-6.1
ppm) and ortho (+2.4 ppm) ring carbon resonances.
Assembly of the Cs4[M2Pd3Br6L6] (M ) Ti, Sn) Clusters.
The [ML3]2- complexes have three divergent diphenylphosphino
groups that can coordinate to three separate metal centers (M′).
If these metal centers are C2 symmetric, supramolecular clusters
of the type M2M′3L6 (Figure 1) can be assembled. As monitored
by 31P{1H} NMR spectroscopy, titration of 3 equiv of a DMF
solution of PdBr2‚2PhCN into 2 equiv of Cs2[TiL3] in DMF
led to the exclusive formation of a single red product. No
exchange between this complex with excess free ligand was
observed on the NMR time scale at ambient temperature.
Isolation of the orange-red product in 95% yield followed by
Positive ion FAB mass spectrometry in DMF (nitrobenzyl
alcohol matrix) further confirmed the presence of [Ti2Pd3Br6L6]4-
clusters with various counterion compositions (see Experimental
Section). Other alkali metal salts of this cluster were prepared
analogously from the corresponding Li2[TiL3], K2[TiL3], or Rb2-
[TiL3] complexes. The (DABCO-H)4[Ti2Pd3Br6L6] cluster was
prepared from (DABCO-H)2[TiL3]. Spectral data of these
products are listed in the Experimental Section.
Similar assembly from the Cs2[SnL3] precursor led to nearly
quantitative formation of the Cs4[Sn2Pd3Br6L6] cluster. Finally,
single-crystal X-ray diffraction studies of Cs4[Ti2Pd3Br6L6],
Cs4[Sn2Pd3Br6L6] (cf. ref 36), and (DABCO-H)4[Sn2Pd3Br6L6]
confirmed the structures of these clusters in the solid state (vide
infra).
The use of DMF as a solvent was essential for clean formation
of a single product. Less polar solvents such as methanol and
acetonitrile yielded multiple products as well as insoluble
material. Interestingly, attempts at cluster formation with several
other cationic salts of [TiL3]2- were unsuccessful. For example,
tetraphenyl-phosphonium and PPN+ salts of [TiL3]2- yielded
only insoluble material with PdBr2 in all solvent systems
attempted. In addition, both triethylammonium and tetramethyl-
ammonium precursors gave multiple products in DMF, as
revealed by 31P NMR spectroscopy (Figure 5a). A FAB-mass
spectrum of the [NEt3H]2 [TiL3]/PdBr2 (2:3 mol ratio) reaction
mixture in DMF gave no evidence of any cluster formation.
However, addition of stoichiometric equivalents of cesium
triflate to this solution resulted in formation of a single species
whose NMR spectra and FAB mass spectrum confirmed the
formation of Cs4[Ti2Pd3Br6L6] (31P{1H} NMR spectrum, Figure
5b). Interestingly, the FAB mass spectrum revealed a propensity
for the Cs+ cations to be ionized with the cluster even when
run in the positive mode (see Experimental Section). This
suggests a strong interaction between the cluster and the Cs+
cations even under these harsh conditions. In light of this and
the solid-state structure of the complex discussed below, we
suggest that the cesium cations remain embedded into the three
cluster clefts even in solution, which may be essential for the
integrity of the cluster. Fast exchange of these cluster-bound
cesium cations with solvated cesium atoms was revealed by
titrating 4 equiv of CsOTf into the cluster and observing the
133Cs NMR spectra; only a single, averaged broad signal was
observed at ambient temperature.
(63) We examined three possible factors that may favor the fac-isomer
in solution. (1) Sterics: The remoteness of the three diphenylphosphino
substituents from the metal center and from each other renders significant
differences in fac versus mer steric strain energies unlikely. Molecular
mechanics indeed showed very similar strain energies. Steric effect as the
origin of any isomeric preference can be discounted. (2) Electronic effect:
Electronically the PPh2 group is known to be moderately electron withdraw-
ing,61,62 thus the existence of an intrinsic electronic effect (or trans influence)
favoring the fac-isomer is feasible. However, DFT molecular orbital
calculations (pBP-DN* with the Spartan 5.0 programs, Wavefunction, Inc.)
on [Ti(catecholato-4-PH2)3]2- actually favored the mer over the fac as the
ground-state configuration by 11.7 kJ/mol. Further, no convincing X-ray
data consistent with differing metal-oxygen bond distances due to such a
substituent effect have been found in either the Sn(4-NO2-catecholato)3 (fac)
(Lamberth, C.; Machell, J. C.; Mingos, D. M. P.; Stolberg, T. L. J. Mater.
Chem. 1991, 1, 775-780) or the Cs2TiL3 (mer) structure described above.
(3) SolVation effect: Since the fac-isomer can be expected to have a
substantially larger dipole moment than the mer-isomer (a substantially larger
molecular dipole was found for the fac-isomer relative to the mer-isomer;
4.13 D versus 1.56 D from PM3(tm) calculations, and 1.2 D versus 0.54 D
from DFT calculations). The resulting higher solvation energy in a polar
solvent may be the origin of its predominance. In this case, a significant
solvent polarity dependence of the speciation can be anticipated. Our NMR
spectral data do show some solvent dependence. For example, both Li+
and Cs+ salts of TiL32- showed broadened room temperature 1H signals in
CD2Cl2 compared to spectra in more polar CD3OD or d6-acetone. Interest-
ingly, for TMA2TiL3, both the TMA as well as the ligand signals are
broadened at room temperature in CD2Cl2 but not in CD3OD or d6-acetone.
These observations, however, can be attributed just as well to differing ion
pairing effects rather than to the emergence of a mer-isomer in a low polarity
solvent. We therefore sought to remove the former influence by studying
the PPN+ salt of [TiL3]2- since this bulky cation is not expected to
1
participate in tight ion pairing. Both the H and 31P{1H} NMR spectra of
PPN2TiL3 in CD2Cl2 and d6-acetone gave the sharp signals of a single C3
solution species. Thus tight ion pairing is most likely the origin of the
solvent-dependent spectra of the Li+, Cs+, and TMA+ salts.
(64) Fay, R. C. Coord. Chem. ReV. 1996, 154, 99.
(65) Fay, R. C.; Lindmark, A. F. J. Am. Chem. Soc. 1983, 105, 2118.
(66) Wong, C. Y.; Woollins, J. D. C. C. R. Coord. Chem. ReV. 1994,
130, 175.
(67) Denekamp, C. I. F.; Evans, D. F.; Parr, J.; Woollins, J. D. J. Chem.
Soc., Dalton Trans. 1993, 1489.
(68) Woollins, J. D.; Parr, J. Personal communication.
(69) Pregosin, P. S.; Kunz, R. W.; Springer-Verlag: New York, 1979;
p 89.

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2757
Figure 5. 31P{1H} NMR spectra of the reaction mixture of [NEt3H]2-
TiL3 and PdBr2‚2PhCN (2:3 ratio) in DMF: (bottom) before and (top)
after addition of 2 equiv of cesium triflate.
Figure 6. A view of the solid-state structure of Cs4[Ti2L6(PdBr2)3]‚
6THF viewed down the crystallographic 3-fold axis of the cluster. One
disordered Cs atom and all disordered solvent molecules and hydrogen
atoms have been omitted for clarity. The purple spheres represent P
atoms, brown spheres are Br atoms, green spheres are Cs atoms, yellow
spheres are Ti atoms, gray spheres represent carbon, and red spheres
are oxygen. The Pd ions are eclipsed by the coordinating phosphine
ligands.
2-
A 2:3 assembly of the C3-ML3 and C2-trans-PdBr2 can
yield either the D3 helicate or the C3h mesocate cluster. CAChe
molecular modeling revealed that the latter should be consider-
ably less sterically congested than the former.47 The X-ray
structural determinations of these clusters as described below
indeed confirm the exclusive formation of mesocates in the solid
state. Since the NMR data are consistent with the formation of
a unique species, we conclude that this solid state mesocate
structure is preserved in solution, although the possibility of a
fast mesocate/helicate equilibration cannot be unequivocally
discounted.
X-ray Structure of Cs4[Ti2Pd3Br6L6]‚9THF‚H2O‚1.5Et2O‚
1.5DMF. A single-crystal X-ray diffraction study confirmed the
successful assembly of the desired heterometallic cluster. This
has crystallographically imposed C3h symmetry and is therefore
a mesocate featuring one Ti center having ∆ and the second
the Λ configuration. The top and bottom TiL3 halves of the
cluster are linked by three pseudo-square-planar PdBr2 units,
each trans coordinated by one PPh2 group from each half (Figure
6). A 3-fold axis exists along the Ti‚‚‚Ti axis with the metal
centers separated by 6.76 Å. The twist angle around each Ti is
36.9°, almost halfway distorted from the octahedral twist angle
of 60° toward the trigonal prismatic coordination geometry of
0° (Figure 7). While the average Ti-catecholato oxygen distance
of 1.97 Å is normal, the Ti-O(1) distance of 1.949(5) Å is
shorter than the Ti-O(2) distance of 1.984(5) Å. This may be
a result of the strain imposed by formation of the cluster
whereby endo Ti-O(2) bonds are stretched while exo Ti-O(1)
bonds are compressed to accommodate the supramolecular
structure. In accord with this, exo cluster O(1)-Ti-O(1′) angles
are 88.3(2)° while endo O(2)-Ti-O(2′) are 89.2(2)°. Both O(1)
and O(2) are also coordinated to Cs+ cations.
Figure 7. ORTEP view highlighting Ti and Cs coordination spheres
in the solid-state structure of Cs4[Ti2L6(PdBr2)3]. All carbon, nitrogen,
phosphorus, palladium, bromine, and noncoordinating oxygen atoms
are omitted for clarity. Thermal ellipsoids are at 50% probability.
is one of the shortest reported.71 The three embedded cesium
cations are separated from each other by 4.45 Å. The fourth
Cs+ is coordinated to the exterior of the anionic cluster and
disordered equally over the two tris(catecholate)titanium cap
oxygen sites (Figure 7). The Cs(1)-O(1) distance of 3.080(5)
Å is again quite short. The remaining coordination sphere of
this cesium is probably filled by disordered ether and DMF
solvent molecules in the vicinity.
Of particular interest is the location of the four cesium
counterions. Three of these are deeply embedded in the three
clefts of the cluster. Each symmetry equivalent Cs(2) is
coordinated by four endo-catecholate oxygens in a rectangular
array and two exo-THF solvent molecules which are buried in
the clefts (Figure 6). The low coordination number (6) of these
cesium cations results from the peripheral Br atoms from the
three PdBr2 moieties which, while not in van der Waals contact
with the cations, effectively shield them from additional donors.
Therefore, relatively short Cs-O distances are observed.
Specifically, Cs(2) to catecholate O(2) has a distance of 3.138-
(5) Å while the 2.918(10) Å distance of Cs(2) to O(3) of a THF
X-ray Structure of Cs4[Sn2Pd3Br6L6]‚3.5DMF‚2H2O‚THF‚
x(solvent). Although the cluster has no crystallographically
imposed symmetry, it is also a mesocate and has idealized C3h
(70) Shannon, R. D. Acta Crystallogr. 1976, A 32, 751.
(71) Allen, F. H.; Kennard, O. Chem. Design Autom. News 1993, 8, 31.

2758 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Sun et al.
Figure 8. A view of the solid-state structure of Cs4[Sn2L6(PdBr2)3]
along the pseudomirror plane of the cluster. All disordered solvent
molecules, all hydrogen atoms, and one of the disordered Cs atoms
are omitted for clarity. The color scheme is the same as for Figure 6,
with Sn atoms represented by brown spheres and Pd atoms as light
purple spheres.
symmetry similar to the Cs4[Ti2Pd3Br6L6] structure described
above (Figure 8). Again, the two Sn atoms sit on an idealized
3-fold axis with a pseudomirror plane (through each of the three
Pd atoms) linking the ∆- and Λ-tris(catecholate)tin halves via
trans coordination of the PPh2 groups. The Sn coordination
spheres, with twist angles of 44.0° and 46.1°, are less distorted
from octahedral than in the Ti analogue. The Sn‚‚‚Sn separation
is 6.88 Å, slightly longer than the Ti‚‚‚Ti of 6.76 Å, presumably
a result of the larger ionic radius of SnIV (0.830 Å) compared
to TiIV (0.745 Å).70
Figure 9. ORTEP view highlighting Sn and Cs coordination spheres
in the solid-state structure of Cs4[Sn2L6(PdBr2)3]. All carbon, nitrogen,
phosphorus, palladium, bromine, and noncoordinating oxygen atoms
are omitted for clarity. Thermal ellipsoids are at 30% probability.
Three of the four cesium cations (Cs(1), Cs(2), and Cs(3))
are well-ordered while a fourth is disordered over three sites
(Cs(4), Cs(5), and Cs(6)). Both Cs(1) and Cs(3) are located in
cluster clefts and are seven-coordinate. Cs(1) is coordinated by
a rectangular array of endo-catecholate oxygens, two DMF
molecules, and a bromide, Br(3), from a PdBr2 moiety. Cs(3)
is coordinated by four endo-catecholate oxygens, one DMF, and
one water molecule, as well as a bromide, Br(5). Cs(2) sits atop
Sn(2), slightly off the pseudo 3-fold axis, and is coordinated to
three exo-catecholate oxygens, one DMF, and one water
molecule (Figure 9). In addition, disordered solvent molecules
probably complete its coordination sphere. The disordered fourth
cesium cation is located with 2/3 occupancy as Cs(4) in the third
1
cluster cleft. The remaining /3 occupancy is equally divided
Figure 10. A view down the 3-fold axis of the solid-state structure of
the (DABCO-H)4[Sn2Pd3Br6L6] cluster. All disordered solvent mol-
ecules, all hydrogen atoms, and a DABCO-H have been omitted for
clarity. The ligand is represented as a wireframe, Pd atoms are yellow
spheres, Sn atoms are orange spheres, oxygen atoms of encapsulated
solvent water molecules are represented as red spheres, Br atoms are
brown spheres, and blue and gray spheres represent nitrogen and carbon
atoms of the DABCO-H molecules, respectively.
between two capping exo sites above Sn(1) at Cs(5) and Cs(6).
The palladium ions have a pseudo-square-planar geometry
with normal Pd-P bond distances of 2.32-2.35 Å. An
interesting feature of the structure is the deflection of one Br
from each PdBr2 linker toward the cluster cleft to coordinate
the embedded Cs+ cations (Figure 8). The larger intracluster
Sn‚‚‚Sn separation of 6.88 Å may facilitate this interaction. The
average Cs-Br distance of 3.82 Å is well within the normal
range of reported Cs-Br distances in the Cambridge Structural
Database.71 Other important bond distances and bond angles
are listed in Table 3.
X-ray Structure of (DABCO-H)4[Sn2Pd3Br6L6]‚3H2O‚x-
(solvent).72 The core Sn2Pd3L6 cluster is again a C3h mesocate
as for the two structures discussed above (Figures 10 and 11).
The twist angle around the tin center is 41.9° while the average
Sn-O distance is 2.05 Å. Of interest is the significantly larger
Sn‚‚‚Sn separation of 7.47 Å compared to the 6.88 Å observed
in the Cs4[Sn2Pd3L6] structure, presumably necessary for
accommodating the three bulky DABCO groups within the
cluster clefts.
Again, the three pseudo-square-planar palladium centers are
trans coordinated by PPh2 groups with a normal Pd-P bond
length of 2.363(2) Å and nearly linear P-Pd-P angle of 174.5-
(1)°. While Br(1) sits 2.445(2) Å from the Pd center, the second
Br is disordered over Br(2) and Br(3) sites 0.63 Å apart.
Three protonated DABCO cations sit in the expanded cluster
clefts. In addition, three water molecules are sandwiched
between the two Sn atoms and each is hydrogen bonded to two
endo-catecholate oxygens as well as a protonated DABCO. A
fourth protonated DABCO is not associated with the cluster
and is treated as disordered x(solvent). Important bond distances
and angles are presented in Table 4.
(72) (DABCO-H)4[Sn2Pd3Br6L6]‚3H2O‚x(solvent): orange triangular
prism, crystal size 0.42 × 0.06 × 0.06 mm3, FW ) 3180.21, T ) -80 °C,
hexagonal space group, P63/m, a ) 23.2576(5) Å, c ) 16.7452(4) Å, V )
7844.1(3) Å3, Z ) 2, R ) 0.055, Rw ) 0.078, GOF ) 2.47.

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2759
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for
Cs4[Sn2Pd3Br6L6]
Cs(1)-Br(3)
Cs(1)-O(4)
Cs(1)-O(6)
Cs(1)-O(10)
Cs(1)-O(12)
Cs(1)-O(15)
Cs(2)-O(9)
Cs(2)-O(11)
Cs(2)-O(11′)
Cs(2)-O(13)
Cs(2)-O(20)
Cs(3)-Br(5)
Cs(3)-O(2)
Cs(3)-O(6)
Cs(3)-O(8)
Cs(3)-O(10)
Cs(3)-O(16)
Cs(4)-Br(1)
Cs(4)-O(2)
Cs(4)-O(4)
Cs(4)-O(8)
Cs(4)-O(12)
Cs(4)-O(23)
Cs(5)-O(3)
Cs(5)-O(19)
Cs(6)-O(3)
Cs(6)-O(19)
3.754(3) Sn(1)-O(1)
3.08(2) Sn(1)-O(2)
3.22(2) Sn(1)-O(3)
3.28(2) Sn(1)-O(4)
3.23(1) Sn(1)-O(5)
2.97(5) Sn(1)-O(6)
3.14(1) Sn(2)-O(7)
3.16(1) Sn(2)-O(8)
3.18(2) Sn(2)-O(9)
3.07(2) Sn(2)-O(10)
3.14(5) Sn(2)-O(11)
3.960(3) Sn(2)-O(12)
3.17(2) Pd(1)-Br(1)
3.25(2) Pd(1)-Br(2)
3.19(1) Pd(1)-P(1)
3.28(2) Pd(1)-P(2)
3.08(4) Pd(2)-Br(3)
3.741(5) Pd(2)-Br(4)
3.27(2) Pd(2)-P(3)
3.27(1) Pd(2)-P(4)
3.21(2) Pd(3)-Br(5)
3.23(1) Pd(3)-Br(6)
2.90(4) Pd(3)-P(5)
2.96(2) Pd(3)-P(6)
3.31(8) Cs(6)-O(1)
3.05(2) Cs(6)-O(5)
3.30(8)
2.05(2)
2.08(2)
2.06(2)
2.07(2)
2.06(2)
2.04(2)
2.06(2)
2.07(2)
2.03(1)
2.04(1)
2.07(1)
2.08(2)
2.451(4)
2.424(4)
2.331(7)
2.345(6)
2.442(4)
2.405(4)
2.336(8)
2.332(8)
2.446(2)
2.412(2)
2.331(7)
2.316(7)
3.10(2)
3.23(2)
Figure 11. A stereoview of the solid-state structure of the (DABCO-
H)4[Sn2Pd3Br6L6] cluster along the mirror plane of the cluster. Again
all disordered solvent molecules, all hydrogen atoms, and a DABCO-H
have been omitted for clarity. All but the Sn ions, DABCO-H, and
encapsulated solvent water molecules are shown as a wireframe. The
same color scheme as in Figure 10 is used.
Table 2. Selected Bond Distances (Å) and Bond Angles (deg) for
Cs4[Ti2Pd3Br6L6]a
Cs(1)-O(1)
Cs(2)-O(3)
Cs(2)-O(2)
Cs(2)-O(2′)
Pd(1)-P(1)
Pd(1)-Br(2)
Pd(1)-Br(1)
Pd(1)-Br(4)
Ti(1)-O(2)
3.080(5)
2.918(10) O(2)-C(2)
3.138(5)
3.181(5)
2.326(2)
2.405(2)
2.4309(13) Pd(1)-Br(3)
2.501(14) Ti(1)-O(1)
1.984(5)
O(1)-C(1)
1.340(8)
1.344(8)
1.28(2)
1.47(2)
1.36(3)
1.58(2)
2.468(9)
1.949(5)
O(3)-C(22)
O(3)-C(19)
O(4)-C(26)
O(4)-C(23)
O(1)-Cs(1)-O(1′) 52.3(2)
O(3)-Cs(2)-O(2) 91.4(3)
O(3)-Cs(2)-O(3′)
83.9(6)
O(3)-Cs(2)-O(2′) 141.4(2)
P(1)-Pd(1)-P(1′) 178.17(10)
P(1)-Pd(1)-Br(1) 90.12(5)
P(1)-Pd(1)-Br(3) 89.86(5)
P(1)-Pd(1)-Br(4) 89.67(5)
O(1)-Ti(1)-O(1′) 88.3(2)
P(1)-Pd(1)-Br(2) 90.17(5)
Br(2)-Pd(1)-Br(1) 161.9(2)
Br(1)-Pd(1)-Br(3) 179.0(9)
Br(1)-Pd(1)-Br(4) 166.7(9)
O(1′)-Ti(1)-O(2) 105.7(2)
O(1′)-Ti(1)-O(2) 161.3(2)
Br(3)-Cs(1)-O(4) 138.5(2)
Br(3)-Cs(1)-O(10) 89.9(2)
Br(3)-Cs(1)-O(15) 103.8(7)
O(4)-Cs(1)-O(10) 115.0(4)
Br(3)-Cs(1)-O(6)
Br(3)-Cs(1)-O(12) 128.8(3)
95.0(2)
O(4)-Cs(1)-O(6)
O(4)-Cs(1)-O(12)
O(4)-Cs(1)-O(10)
55.3(4)
92.1(3)
89.4(5)
O(1)-Ti(1)-O(2)
80.0(2)
O(4)-Cs(1)-O(15)
89.3(8)
O(6)-Cs(1)-O(12) 115.4(4)
O(10)-Cs(1)-O(12) 52.8(4)
O(12)-Cs(1)-O(15) 78.8(7)
O(6)-Cs(1)-O(15) 140.6(8)
a Primes represent symmetry equivalent atoms.
O(10)-Cs(1)-O(15) 124.4(7)
Summary of Solid-State Cluster Structures. All three
clusters are mesocates featuring both enantiomeric forms of the
C3-symmetry ML3 units linked by three trans-PdBr2 bridges to
give idealized C3h symmetry in the solid state. The three deep
molecular clefts of these clusters are occupied by the respective
counterions of the tetranionic assembly. The larger SnIV (83
pm compared to 74.5 pm for TiIV) ionic radius and longer
intracluster Sn‚‚‚Sn separation (6.88 Å compared to 6.76 Å for
Ti‚‚‚Ti) enables an unusual deflection of the PdBr2 units toward
the embedded Cs+ cations so that each coordinates to one
bromide. To accommodate the even larger protonated DABCO
cations, this Sn‚‚‚Sn separation is increased to 7.47 Å, though
these protonated tertiary amine cations are not directly coordi-
nated to the endo-catecholate oxygens. Some disorder of the
fourth cation as well as considerable disorder of the solvent
molecules are noted in all three structures.
Syntheses of the [M2Pd3X6L6] (X ) Cl, I) Clusters.
Reaction of Cs2TiL3 with PdCl2‚2PhCN (2:3 ratio) in DMF gave
Cs4[Ti2Pd3Cl6L6] as a dark-red solid in 74% yield. Elemental
and spectral analyses confirmed the formation of this cluster
(see Experimental Section). The iodo analogue was prepared
in quantitative yield by halide metathesis between Cs4[Ti2Pd3-
Br6L6] and CsI in DMF. Again both analytical and spectral data
confirmed its formulation as Cs4[Ti2Pd3I6L6]. A similar halide
metathesis yielded Cs4[Sn2Pd3I6L6] from Cs4[Sn2Pd3Br6L6] and
CsI. All these species exhibit the expected C3h symmetry in
solution based on NMR spectroscopy.
O(1)-Sn(1)-O(2)
O(1)-Sn(1)-O(4)
O(1)-Sn(1)-O(6)
O(2)-Sn(1)-O(4)
O(2)-Sn(1)-O(6)
O(3)-Sn(1)-O(5)
O(4)-Sn(1)-O(5)
O(5)-Sn(1)-O(6)
O(7)-Sn(2)-O(9)
O(7)-Sn(2)-O(11)
O(8)-Sn(2)-O(9)
O(8)-Sn(2)-O(11)
O(9)-Sn(2)-O(10)
O(9)-Sn(2)-O(12)
81.9(7)
165.2(5)
100.8(8)
89.2(7)
89.0(6)
90.2(7)
90.7(7)
81.7(7)
90.6(7)
88.6(6)
189.1(6)
97.8(5)
81.8(6)
96.8(7)
O(1)-Sn(1)-O(3)
O(1)-Sn(1)-O(5)
O(2)-Sn(1)-O(3)
O(2)-Sn(1)-O(5)
O(3)-Sn(1)-O(4)
O(3)-Sn(1)-O(6)
O(4)-Sn(1)-O(6)
O(7)-Sn(2)-O(8)
88.2(8)
90.7(7)
100.4(6)
166.8(7)
81.8(7)
167.8(9)
90.7(7)
82.3(6)
O(7)-Sn(2)-O(10) 101.2(6)
O(7)-Sn(2)-O(12) 167.8(4)
O(8)-Sn(2)-O(10)
O(8)-Sn(2)-O(12)
O(9)-Sn(2)-O(11)
91.4(5)
91.6(6)
90.3(5)
O(10)-Sn(2)-O(11) 89.5(6)
O(10)-Sn(2)-O(12) 81.7(6)
Br(1)-Pd(1)-Br(2) 172.4(1)
Br(1)-Pd(1)-P(1)
Br(2)-Pd(1)-P(1)
P(1)-Pd(1)-P(2)
Br(3)-Pd(2)-P(3)
Br(4)-Pd(2)-P(3)
P(3)-Pd(2)-P(4)
Br(5)-Pd(3)-P(5)
Br(6)-Pd(3)-P(5)
P(5)-Pd(3)-P(6)
86.0(2)
92.1(2)
172.7(2)
90.9(2)
89.4(2)
173.4(3)
87.0(1)
93.8(1)
171.5(3)
Br(1)-Pd(1)-P(2)
Br(2)-Pd(1)-P(2)
87.7(2)
93.6(2)
Br(3)-Pd(2)-Br(4) 162.9(2)
Br(3)-Pd(2)-P(4)
Br(4)-Pd(2)-P(4)
91.7(2)
89.9(2)
Br(5)-Pd(3)-Br(6) 174.4(2)
Br(5)-Pd(3)-P(6)
Br(6)-Pd(3)-P(6)
87.4(1)
91.2(1)
Cs(4)-Br(1)-Pd(1) 118.93(9) Cs(1)-Br(3)-Pd(2) 123.8(1)
Cs(3)-Br(5)-Pd(3) 120.6(1)
at δ +71.92 as well as the presence of unreacted Cs2TiL3 at δ
-4.74. In this case, the desired cluster Cs4[Ti2{Cr(CO)4}3L6]
was successfully isolated after workup in 66% yield. Analo-
gously, Cs4[Sn2{Cr(CO)4}3L6] was also synthesized in 74%
yield and both have been characterized by elemental analyses,
Syntheses of Other M2M′3L6 Clusters. Extension of the
aufbau assembly of the M2Pd3L6 cluster motif with other C2
linkers was attempted. Thus reaction of 3 equiv of cis-Cr(CO)4-
(piperidine)2 with 2 equiv of Cs2TiL3 in DMF gave a cloudy
orange solution after 5 days at room temperature. 31P{1H} NMR
analysis revealed formation of a new complex with a resonance
1
IR, and H, 13C{1H}, and 31P{1H} NMR spectroscopy. Com-
pared to spectral data for trans-Cr(CO)4(PPh3)2,69 all these data
are fully consistent with formation of the expected C3h mesocate
clusters featuring PPh2-coordinated trans-Cr(CO)4 linking groups.
The single strong IR carbonyl stretching band at 1855 cm-1

2760 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Sun et al.
Table 4. Selected Bond Distances (Å) and Bond Angles (deg) for
and self-assembly routes to the variety of C3h M2M′3L6 clusters
described herein. Extensions of this methodology to additional
hard/soft metals as well as other ligand combinations promise
to be an attractive and effective entry into other fascinating
supramolecular assemblies. The self-assembly of these hetero-
metallic clusters exemplifies our design strategy in which hard
and soft chemical bonding preferences are satisfied while
simultaneously forming a thermodynamically stable, discrete
cluster.
(H- DABCO)4[Sn2Pd3Br6L6]a
Sn(1)-O(1)
2.044(5) O(1)-C(1)
2.065(5) O(2)-C(2)
2.445(2) N(1)-C(19)
2.424(2) N(1)-C(21)
2.393(2) N(2)-C(20)
2.363(2) N(2)-C(22)
1.332(8)
1.371(8)
1.513(10)
1.49(2)
1.47(1)
1.44(2)
Sn(1)-O(2)
Pd(1)-Br(1)
Pd(1)-Br(2)
Pd(1)-Br(3)
Pd(1)-P(1)
O(1)-Sn(1)-O(1′)
89.7(2)
O(1)-Sn(1)-O(2)
80.4(2)
O(1)-Sn(1)-O(2′) 164.9(2)
O(1′)-Sn(1)-O(2) 101.6(2)
O(2)-Sn(1)-O(2′)
90.3(2)
P(1)-Pd(1)-Pd(1′) 174.5(1)
Br(1)-Pd(1)-Br(3) 171.3(5)
Br(1)-Pd(1)-Br(2) 173.7(1)
Experimental Section
Br(1)-Pd(1)-P(1)
88.88(6) Br(2)-Pd(1)-P(1)
91.39(6)
General Reaction Conditions and Reagent Sources. All operations
and manipulations were performed in standard Schlenk glassware under
a dry nitrogen atmosphere. Solvents were commercial reagent grade
and degassed before use. The following reagents were prepared
according to literature methods: 4-PCl2-veratrole,48 PdBr2‚2PhCN,74
PdCl2‚2PhCN,74 and Cr(CO)4(piperidine)2.75
Br(3)-Pd(1)-P(1)
90.73(6)
a Primes represent symmetry equivalent atoms generated by the
crystallographic 3-fold axis.
Scheme 2
Instrumentation and Methods. All NMR spectra were recorded
on a Bruker AM 360 MHz or a JEOL FX-90Q spectrometer. Chemical
shifts of 1H and 13C spectra are referenced to internal TMS. 31P spectra
were referenced to external 85% H3PO4. 119Sn and 133Cs NMR shifts
were referenced to external SnCl4 and CsCl, respectively. Electronic
spectra were measured on a Cary 219 spectrophotometer and EPR
spectra on a Varian E-4 instrument. Infrared spectra were run on a
Nicolet MX-1 FT-spectrophotometer with use of KBr pellets. Elemental
analyses were performed by the UNH Instrumentation Center on a
Perkin-Elmer 2400 Elemental Analyzer. FAB mass spectra were run
in an NBA matrix and carried out at the UC-Berkeley Mass Spectrom-
etry Facility. Single-crystal X-ray diffraction data were collected on a
Siemens SMART diffractometer in the Chexray facility at UC-Berkeley.
4-Diphenylphosphino-Veratrole (4-PPh2-Veratrole). A three-neck
flask under nitrogen was charged with magnesium (3.5 g, 0.14 mol)
and dry diethyl ether (80 mL) and chilled to 0 °C in an ice bath. A
solution of bromobenzene (14 mL, 0.13 mol) in 40 mL of dry diethyl
ether was added to the flask dropwise while the mixture in the flask
was stirred. After addition the mixture was heated to reflux for 1 h,
and then cooled to 0 °C. A solution of 4-dichlorophosphino-veratrole
(7.8 g, 33 mmol) in 30 mL of dry diethyl ether was added dropwise to
the vigorously stirring mixture. Upon addition a white precipitate was
produced. This mixture was heated to reflux again for 2 h, then cooled
to 0 °C. Water (50 mL) was added dropwise to quench the excess
Grignard reagent (CAUTION! The first few drops caused a rigorous
reaction producing plenty of heat). The mixture was then filtered in
air, and the filtrate was reduced by rotary evaporation and dried in a
vacuum to yield 10.2 g of a yellowish-white solid. Recrystallization of
this solid from ethanol (25 mL) afforded a pure white solid. Yield:
9.5 g, 29 mmol (88%).
and the triplet carbonyl 13C NMR resonance at δ +216 of the
latter product support a local D4h symmetry around the Cr
centers.
To prepare a Ga2Pd3L6 cluster, 2 equiv of Cs3GaL3 and 3
equiv of PdBr2‚2PhCN were combined in DMF solvent yielding
a clear, red solution that has a single 31P{1H} NMR resonance
at δ +22. Both the 1H and 13C{1H} NMR spectra also revealed
the expected patterns, although the signals were severely
broadened, indicative of significant dynamic behavior. Analyti-
cal data also supported a successful synthesis of the desired
Cs3Ga2(PdBr2)3L6 cluster.
The Self-Assembly of [M2Pd3Br6L6]4- Clusters. The ulti-
mate hybrid ligand should ideally exhibit sufficient discrimina-
tion between the two types of targeted incommensurate metal
centers to allow a one-pot self-assembly of a single heterome-
tallic cluster from a mixture of all the individual components.
In our case, this requires the spontaneous assembly of 11
components (15, including the 4 equiv of base needed): 2 “C3-
symmetry” MIV centers, 3 C2 PdBr2 units, as well as 6 equiv of
ligand L. We have already noted the propensity of PdII to form
very insoluble oligomeric material with H2L under basic
conditions (vide infra). We therefore sought alternative bases
with higher solubility than the alkaline metal carbonates.
Gratifyingly, successful self-assembly was achieved by using
aqueous ammonia or DABCO solutions as the base in DMF
solvent (Scheme 2). Both the (NH4)4[Ti2Pd3Br6L6]73 and
(DABCO-H)4[Sn2Pd3Br6L6] clusters can be isolated in excellent
yields via spontaneous self-assembly of the respective 11
components! The viability of this direct route broadens con-
siderably the scope of cluster formation chemistry from our
hybrid ligands.
1H NMR (360 MHz, CDCl3): δ 7.33 (m, 10 phenyl-H), 6.89 (m,
3-cat-H), 3.90 (s, -OCH3), 3.77 (s, OCH3). 13C {1H} NMR (89.9 MHz,
CDCl3): δ 149.8 (s, veratrole), 149.0 (d, Jpc ) 10.0 Hz, veratrole),
137.6 (broad, phenyl), 133.5 (d, Jpc ) 18.6 Hz, phenyl), 128.6 (s,
phenyl), 128.5 (d, Jpc ) 6.6 Hz, phenyl), 127.9 (broad, veratrole), 127.2
(d, Jpc ) 19.4 Hz, veratrole), 116.6 (d, Jpc ) 25.2 Hz, veratrole), 111.3
(d, Jpc ) 8.0 Hz, veratrole), 55.8 (s, OCH3). 31P {1H} NMR (36.3 MHz,
CDCl3): δ -5.35 (s). IR (KBr) υ(OCH3) 2957, 2833 cm-1, C-H bending
1507, 1438 cm-1, υ(C-O-C) 1254, 1235 cm-1. Anal. Calcd for C20H19-
PO2: C, 74.52; H, 5.94. Found: C, 74.31; H, 5.71.
4-Diphenylphosphino-Catechol, Hydrobromide Salt, 4-PPh2-
catechol‚HBr (H2L‚HBr). A mixture of 4-diphenylphosphino-veratrole
(5.0 g, 16 mmol) and 32 mL of 48% aqueous hydrobromic acid (0.28
mol) was refluxed under nitrogen for 18 h. Upon chilling, 4.7 g (13
mmol) of product precipitated. Recrystallization of this solid from
ethanol (20 mL) and 48% aqueous hydrobromic acid (2 mL) provided
a white solid. Yield: 4.1 g, 11 mmol (70%). 1H NMR (360 MHz, DMF-
d7): δ 7.40 (m, 6 phenyl-H), 7.28 (m, 4 phenyl-H), 6.93 (dd, J ) 7.9
and 1.53 Hz, 1 cat-H), 6.82 (dd, J ) 7.53 and 1.73 Hz, 1 cat-H), 6.71
(ddd, J ) 8.23, 8.23, and 1.53 Hz, 1 cat-H). 13C{1H} NMR (89.9 MHz,
DMF-d7): δ 147.6 (s, catechol), 146.5 (d, Jpc ) 8.6 Hz, catechol), 138.8
Conclusion
The hybrid ligand 4-PPh2-catechol combines a hard dianionic
catecholate chelating site with a disparate soft phosphine donor
group. The coordination of this ligand to metal centers with
incommensurate symmetry requirements has led to novel
supramolecular clusters through both stepwise aufbau synthesis
(73) A preliminary single-crystal X-ray structure has confirmed the
existence of this mesocate in the solid state: Sun, X.; Johnson, D. W.;
Clarke, K. M.; Raymond, K. N.; Wong, E. H. Unpublished results.
(74) Anderson, G. K.; Lin, M. Inorg. Synth. 1990, 28, 61.
(75) Atwood, J. L.; Darensbourg, D. J. Inorg. Chem. 1977, 16, 2314.

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2761
(d, Jpc ) 11.3 Hz, phenyl), 133.4 (d, Jpc ) 19.2 Hz, phenyl), 128.9 (s,
phenyl), 128.8 (d, Jpc ) 3.3 Hz, phenyl), 126.3 (d, Jpc ) 25.2 Hz,
catechol), 125.9 (d, Jpc ) 7.3 Hz, catechol), 121.4 (d, Jpc ) 18.5 Hz,
catechol), 116.5 (d, Jpc ) 10.0 Hz, catechol); 31P{1H} NMR (36.3 MHz,
DMF-d7): δ -6.73 (s). IR (KBr) υ(OH) broad 3290 cm-1, υ(aromatic-H)
3077, 3049, 3029 cm-1, C-H bending 1509, 1439 cm-1, υ(C-OH) 1291,
1266 cm-1. Anal. Calcd for C18H16PO2Br: C, 57.62; H, 4.30. Found:
C, 57.37; H, 4.38.
concentrated to half volume under vacuum and filtered through a frit
under nitrogen atmosphere to give an orange-red residue. This solid
was then washed with degassed water three times and dried in a vacuum
overnight. An orange-red solid (460 mg) was obtained in a yield of
1
73%. H NMR (360 MHz, d7-DMF): δ 7.30 (m, 3 × 10 phenyl-H),
6.55 (ddd, J ) 10.0, 7.7 and 1.8 Hz, 3 × 1 cat-H), 6.26 (dd, J ) 7.7
and 1.6 Hz, 3 × 1 cat-H), 6.12 (dd, J ) 7.3 and 1.7 Hz, 3 × 1 cat-H),
3.22 (s, 2 × 12 DABCO-H). 13C{1H} NMR (89.9 MHz, d7-DMF): δ
Cs3[(4-PPh2-Catecholato)3Ga] (Cs3[GaL3]). A mixture of Ga-
(NO3)3‚6H2O (264 mg, 0.731 mmol), 4-PPh2-Catechol‚HBr (815 mg,
2.17 mmol), and Cs2CO3 (1063 mg, 3.26 mmol) was stirred in 30 mL
of degassed methanol under nitrogen atmosphere at room temperature
for 3 days, giving a slightly cloudy solution. The solution was filtered
through a frit with Celite and glasswool (pre-purged by nitrogen) under
nitrogen to give a colorless clear filtrate, which was then evaporated
in a vacuum to give a white powder. The white solid was washed with
degassed water three times, and dried overnight in a vacuum. Yield:
842 mg (86%). 1H NMR (360 MHz, d4-methanol): δ 7.24 (m, 3 × 10
162.2 (s, catechol), 160.6 (d, Jpc ) 8.0 Hz, catechol), 140.2 (d, Jpc
)
12.5 Hz, phenyl), 133.3 (d, Jpc ) 18.7 Hz, phenyl), 128.6 (d, Jpc ) 6.2
Hz, phenyl), 128.3 (s, phenyl), 125.5 (d, Jpc ) 31.6 Hz, catechol), 121.6
(d, Jpc ) 4.7 Hz, catechol), 115.9 (d, Jpc ) 14.1 Hz, catechol), 111.2
(d, Jpc ) 13.2 Hz, catechol), 45.1 (s). 31P{1H} NMR (36.3 MHz, DMF/
d6-benzene) δ -5.35 (s). IR (KBr): υ(N-H) 3431 cm-1, υ(aromatic
C-H)
3051, 3010 cm-1, υ(-CH2CH2-) 2961, 2952, 2884 cm-1, C-H bending
1469 cm-1, υ(C-O-Ti) 1249 cm-1. Elemental Anal. Calcd for TiN4-
C66H65P3O6: C, 68.87; H, 5.69; N, 4.87. Found: C, 68.74; H, 5.90; N,
4.68.
1
phenyl-H), 6.51 (m, 3 × 3 cat-H). H NMR (360 MHz, d7-DMF): δ
Cs2[(4-PPh2-Catecholato)3Sn] (Cs2[SnL3]). To a colorless solution
of SnCl4 (295 mg, 1.13 mmol) in degassed methanol (40 mL) under
nitrogen were added 4-PPh2-catechol‚HBr (1.27 g, 3.40 mmol) and Cs2-
CO3 (1.66 g, 5.10 mmol). (CAUTION! The first few drops of methanol
must be added slowly to avoid SnCl4 boiling over due to the rigorous
methanolysis of SnCl4.) The mixture was stirred under a nitrogen
atmosphere at room temperature for 3 days giving a white suspension.
This was concentrated to half volume under vacuum and filtered through
a frit under a nitrogen atmosphere to give a white solid. The solid was
then washed with degassed water three times, and dried in vacuo
overnight to yield white solid. This solid was recrystallized from
7.29 (m, 3 × 10 phenyl-H), 6.46 (broad d, J ) 8.5 Hz, 3 × 1 cat-H),
6.38 (broad d, J ) 6.7 Hz, 3 × 1 cat-H), 6.27 (broad t, J ) 8.5 Hz, 3
× 1 cat-H). 13C{1H} NMR (89.9 MHz, d4-methanol): δ 158.0 (s,
catechol), 156.5 (d, Jpc ) 10.0 Hz, catechol), 141.5 (d, Jpc ) 10.0 Hz,
phenyl), 134.3 (d, Jpc ) 18.5 Hz, phenyl), 129.2 (d, Jpc ) 6.3 Hz,
phenyl), 129.0 (s, phenyl), 125.2 (d, Jpc ) 27.4 Hz, catechol), 120.6
(s, catechol), 119.0 (d, Jpc ) 18.4 Hz, catechol), 113.9 (d, Jpc ) 12.4
Hz, catechol). 31P{1H} NMR (36.3 MHz, d4-methanol) δ -6.02 (s).
Elemental Anal. Calcd for Cs3GaC54H39P3O6: C, 48.21; H, 2.92.
Found: C, 48.25; H, 3.02.
Cs3[(4-PPh2-Catecholato)3Fe] (Cs3[FeL3]). A mixture of Fe(NO3)3‚
9H2O (135 mg, 0.333 mmol), 4-PPh2-Catechol‚HBr (375 mg, 1.00
mmol), and Cs2CO3 (490 mg, 1.50 mmol) was stirred in 20 mL of
degassed methanol under a nitrogen atmosphere at room temperature
for 4 days, giving a dark purple-red mixture. The mixture was filtered
through a frit under nitrogen, and the blood-red filtrate was then
evaporated in vacuo to give a dark-red residue. Acetone (3 × 10 mL)
was used to extract the product. Evaporation of the dark-red acetone
solution in vacuo gave a dark-red solid. Yield: 112 mg (25%). No
NMR spectrum for this product is available due to the paramagnetism
of ironIII. EPR spectrum (1.0 mM in methanol): g ) 8.18, 4.25, 2.34.
UV/vis spectrum (methanol solution): λmax 323 nm (ꢀ ) 31 905), λmax
490 nm (ꢀ ) 6050). IR (KBr): C-H bending 1474 cm-1, υ(C-O-Fe)
1254 cm-1. Anal. Calcd for Cs3FeC54H39P3O6: C, 48.72; H, 2.95.
Found: C, 48.78; H, 2.94.
Alkali Metal Salts of [(4-PPh2-Catecholato)3Ti]2-([TiL3]2-). The
respective alkali metal carbonate was used in each preparation. The
following procedure for Cs2[(4-PPh2-Catecholato)3Ti] was followed:
A mixture of Ti(OMe)4 (387 mg, 2.14 mmol), 4-PPh2-Catechol‚HBr
(2420 mg, 6.44 mmol), and Cs2CO3 (1760 mg, 5.40 mmol) was stirred
in 60 mL of degassed methanol under nitrogen atmosphere at room
temperature for 42 h, giving an orange-red suspension. The suspension
was filtered through a frit under nitrogen to give an orange-red solid,
which was then washed with degassed water at least three times and
dried under vacuum overnight. The product (ca.. 2.0 g) was obtained
in yields of around 78%. 1H NMR (360 MHz, d6-acetone): δ 7.25 (m,
3 × 10 phenyl-H), 6.53 (ddd, J ) 9.6, 7.9 and 1.8 Hz, 3 × 1 cat-H),
6.39 (dd, J ) 8.1 and 1.8 Hz, 3 × 1 cat-H), 6.35 (dd, J ) 7.8 and 1.5
Hz, 3 × 1 cat-H). 13C{1H} NMR (89.9 MHz, d6-acetone): δ 162.5 (s,
catechol), 161.8 (d, Jpc ) 11.0 Hz, catechol), 140.8 (d, Jpc ) 12.5 Hz,
phenyl), 133.9 (d, Jpc ) 18.8 Hz, phenyl), 129.0 (d, Jpc ) 6.3 Hz,
phenyl), 128.7 (s, phenyl), 125.9 (d, Jpc ) 23.8 Hz, catechol), 122.7
(d, Jpc ) 3.8 Hz, catechol), 117.2 (d, Jpc ) 21.9 Hz, catechol), 112.3
(d, Jpc ) 10.9 Hz, catechol). 31P{1H} NMR (36.3 MHz, d6-acetone) δ
-4.88 (s). IR (KBr): υ(aromatic-H) 3051, 3009, 1468 cm-1, υ(C-O-Ti) 1248
cm-1. Elemental Anal. Calcd for Cs2TiC54H39P3O6: C, 54.50; H, 3.30.
Found: C, 54.46; H, 3.24.
1
acetone/diethyl ether. Yield: 898 mg, 62%. H NMR (360 MHz, d6-
acetone): δ 7.27 (m, 3 × 10 phenyl-H), 6.63 (dd, J ) 8.7 and 1.5 Hz,
3 × 1 cat-H), 6.56 (dd, J ) 7.7 and 1.6 Hz, 3 × 1 cat-H), 6.40 (dd, J
) 9.4, 7.8 and 1.9 Hz, 3 × 1 cat-H). 13C{1H} NMR (89.9 MHz, d6-
acetone): δ 154.9 (s, catechol), 153.5 (d, Jpc ) 10.6 Hz, catechol),
141.0 (d, Jpc ) 12.5 Hz, phenyl), 134.0 (d, Jpc ) 18.7 Hz, phenyl),
129.0 (d, Jpc ) 6.1 Hz, phenyl), 128.7 (s, phenyl), 124.7 (d, Jpc ) 24.3
Hz, catechol), 121.5 (d, Jpc ) 4.0 Hz, catechol), 119.5 (d, Jpc ) 21.6
Hz, catechol), 114.4 (d, Jpc ) 10.9 Hz, catechol). 31P{1H} NMR (36.3
MHz, d6-acetone): δ -4.04 (s). IR (KBr): υ(aromatic
3051, 3002
C-H)
cm-1, C-H bending 1473 cm-1, υ(C-O-Ti) 1246 cm-1. Elemental Anal.
Calcd for Cs2SnC54H39P3O6: C, 51.42; H, 3.12. Found: C, 51.32; H,
3.16.
Cs4[Ti2L6Pd3Br6]. A mixture of Cs2[TiL3] (416 mg, 0.350 mmol)
and PdBr2‚2PhCN (248 mg, 0.525 mmol) was stirred in 20 mL of DMF
under a nitrogen atmosphere at room temperature for 2 h giving a clear
orange solution. Addition of 200 mL of THF precipitated an orange-
yellow solid, which was filtered off and dried under vacuum. Yield:
526 mg (95%). Redissolving about 50 mg of this product in a mixed
solvent of DMF/THF (v/v 1/5, 10 mL) gave a clear orange-red solution.
Slow diffusion of diethyl ether into this solution gave well-formed X-ray
1
quality orange-red crystals. H NMR (360 MHz, d7-DMF): δ 7.89 (t,
J ) 6.4 Hz, 6 × 1 cat-H), 7.63 (broad, 6 × 4 phenyl-H), 7.37 (broad,
6 × 6 phenyl-H), 6.40 (broad, 6 × 1 cat-H), 6.25 (broad d, J ) 7.9
Hz, 6 × 1 cat-H). 13C{1H} NMR (89.9 MHz, d7-DMF): δ 163.8 (s,
catechol), 159.9 (virtual t, JPC ) 10.3 Hz, catechol), 134.7 (virtual t,
JPC ) 5.6 Hz, phenyl), 134.4 (virtual t, JPC ) 20.2 Hz, phenyl), 129.7
(s, phenyl), 127.6 (virtual t, JPC ) 5.0 Hz, phenyl), 125.0 (s, catechol),
121.9 (virtual t, JPC ) 13.9 Hz, catechol), 115.7 (virtual t, JPC ) 28.2
Hz, catechol), 111.0 (virtual t, JPC ) 6.0 Hz, catechol). 31P{1H} NMR
(36.3 MHz, d7-DMF) δ 22.10 (s). IR (KBr): C-H bending 1474 cm-1
,
υ(C-O-Ti) 1258 cm-1. Anal. Calcd for Cs4Ti2C108H78P6O12Pd3Br6: C,
40.80; H, 2.47. Found: C, 40.65; H, 2.41. FAB+ MS m/z (nitrobenzyl
alcohol matrix in DMF) (2 ) [Ti2L6Pd3]2+) {species, observed m/z
(calculated m/z)}: [2 + 6Br- + 5Cs+]1+, 3311 (3312.6); [2 + 6Br-+
4Cs+ + 1H+]1+, 3179 (3180.7); [2 + 6Br-+ 3Cs+ + 2H+]1+, 3049
(3048.8); [2 + 5Br-+ 3Cs+ + 1H+]1+, 2968 (2967.9); [2 + 4Br-
+
(DABCO-H)2[(4-PPh2-Catecholato)3Ti] ((DABCO‚H)2[TiL3]). A
mixture of Ti(OMe)4 (100 mg, 0.550 mmol), 4-PPh2-Catechol‚HBr (626
mg, 1.67 mmol), and DABCO (3.0 g, in excess of 10 equiv) was stirred
in 20 mL of degassed methanol under nitrogen atmosphere at room
temperature for 3 days, giving an orange-red suspension. It was
3Cs+]1+, 2886 (2887.0); [2 + 5Br-+ 2Cs+ + 2H+]1+, 2835 (2836.0);
[2 + 6Br-+ 1Cs+ + 3H+-+ 1Na+]1+, 2807 (2807.0); [2 + 4Br-+
2Cs+ + 1H+]1+, 2755 (2755.1); [2 + 6Br- + 4H+-+ 1Na+]1+, 2675
(2675.1).

2762 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Sun et al.
Cs4[Sn2L6Pd3Br6]. A DMF (15 mL) solution of Cs2[SnL3] (233 mg,
0.176 mmol) and PdBr2‚2PhCN (124 mg, 0.263 mmol) was stirred
under a nitrogen atmosphere at room temperature for 2 h giving a dark-
red solution. Addition of 200 mL of THF precipitated an orange-yellow
solid, which was filtered off and dried in a vacuum. Yield: 270 mg
(89%). Redissolving about 50 mg of the product in 10 mL of DMF
gave a clear orange-red solution. Slow diffusion of diethyl ether into
this solution gave well-formed orange-red crystals of X-ray quality.
1H NMR (360 MHz, d7-DMF): δ 8.15 (broad, 6 × 1 cat-H), 7.66
(broad, 6 × 4 phenyl-H), 7.39 (broad, 6 × 6 phenyl-H), 6.57 (bd, J )
7.6 Hz, 6 × 1 cat-H), 6.25 (broad, 6 × 1 cat-H). 13C{1H} NMR (89.9
MHz, d7-DMF): δ 156.2 (s, catechol), 151.5 (virtual t, JPC ) 10.0 Hz,
catechol), 134.9 (virtual t, JPC ) 5.6 Hz, phenyl), 134.2 (virtual t, JPC
) 24.8 Hz, phenyl), 129.8 (s, phenyl), 127.6 (virtual t, JPC ) 5.0 Hz,
phenyl), 123.8 (s, catechol), 120.6 (virtual t, JPC ) 13.4 Hz, catechol),
114.7 (virtual t, JPC ) 28.2 Hz, catechol), 113.4 (virtual t, JPC ) 4.0
Hz, catechol). 31P{1H} NMR (36.3 MHz, d7-DMF): δ 21.61 (s). IR
(KBr): C-H bending 1479 cm-1, υ(C-O-Sn) 1254 cm-1. Elemental Anal.
Calcd for Cs4Sn2C108H78P6O12Pd3Br6: C, 39.05; H, 2.37. Found: C,
38.91; H, 2.29.
was evaporated in a vacuum to give a dark-red solid. Yield: 212 mg
(97%). Redissolving about 30 mg of the product in 10 mL of DMF
gave a clear dark-red solution. Slow diffusion of diethyl ether into this
1
solution gave well-formed dark-red crystals. H NMR (360 MHz, d7-
DMF): δ 7.78 (broad t, J ) 7.3 Hz, 6 × 1 cat-H), 7.65 (broad m, 6 ×
4 phenyl-H), 7.38 (broad, 6 × 6 phenyl-H), 6.55 (broad m, 6 × 1 cat-
H), 6.32 (broad d, J ) 7.9 Hz, 6 × 1 cat-H). 13C{1H} NMR (89.9
MHz, d7-DMF): δ 162.9 (s, catechol), 158.0 (virtual t, JPC ) 10.0 Hz,
catechol), 136.3 (virtual t, JPC ) 25.2 Hz, phenyl), 134.2 (virtual t, JPC
) 5.6 Hz, phenyl), 129.7 (s, phenyl), 127.3 (virtual t, JPC ) 5.0 Hz,
phenyl), 124.5 (s, catechol), 122.9 (virtual t, JPC ) 18.9 Hz, catechol),
119.9 (virtual t, JPC ) 28.9 Hz, catechol), 110.7 (broad m, catechol).
31P{1H} NMR (36.3 MHz, d7-DMF): δ 12.89 (s). IR (KBr): C-H
bending 1473 cm-1, υ(C-O-Ti) 1253, 1264 cm-1. Anal. Calcd for Cs4-
Ti2C108H78P6O12Pd3I6: C, 37.47; H, 2.27. Found: C, 37.40; H, 2.18.
Cs4[Sn2L6Pd3I6]. Addition of CsI (468 mg, 1.80 mmol) to 20 mL
of DMF solution of Cs4[Sn2L6Pd3Br6] (200 mg, 0.0602 mmol) resulted
in a cloudy mixture that was stirred overnight. The mixture was filtered
through a frit with Celite and glasswool, giving a dark-red filtrate that
was evaporated in vacuo to give the product in quantitative yield (216
mg). Redissolving about 30 mg of the product in 10 mL of DMF gave
a clear dark-red solution. Slow diffusion of diethyl ether into this
(DABCO-H)4[Sn2L6Pd3Br6]. (a) Aufbau Route. A DMF (20 mL)
solution of (DABCO-H)2[SnL3] (424 mg, 0.350 mmol) and PdBr2‚
2PhCN (248 mg, 0.525 mmol) was stirred under nitrogen atmosphere
at room temperature. Initial turbidity disappeared to give a dark-red
solution in 4 h. Addition of 200 mL of THF precipitated an orange-
yellow solid, which was filtered off and dried under vacuum. Yield:
390 mg (69%).
1
solution gave well-formed dark-red crystals. H NMR (360 MHz, d7-
DMF): δ 8.12 (td, J ) 7.6 and 1.9 Hz, 6 × 1 cat-H), 7.66 (m, 6 × 4
phenyl-H), 7.35 (m, 6 × 6 phenyl-H), 6.67 (broad d, J ) 7.9 Hz, 6 ×
1 cat-H), 6.51 (broad m, 6 × 1 cat-H). 13C{1H} NMR (89.9 MHz,
d-DMF): δ 156.1 (s, catechol), 151.3 (virtual t, JPC ) 9.3 Hz, catechol),
137.4 (virtual t, JPC ) 25.2 Hz, phenyl), 134.6 (virtual t, JPC ) 5.3 Hz,
phenyl), 129.6 (s, phenyl), 127.4 (virtual t, JPC ) 4.6 Hz, phenyl), 125.3
(virtual t, JPC ) 11.3 Hz, catechol), 123.9 (s, catechol), 117.0 (m,
catechol), 113.0 (m, catechol). 31P{1H} NMR (36.3 MHz, d7-DMF) δ
12.25 (s). IR (KBr): C-H bending 1474, 1487 cm-1, υ(C-O-Sn) 1254
cm-1. Anal. Calcd for Cs4Sn2C108H78P6O12Pd3I6: C, 36.00; H, 2.18.
Found: C, 35.96; H, 2.12.
(b) Self-Assembly Approach. A suspension of SnCl4, PPh2-Catechol‚
HBr, PdBr2‚2PhCN, and DABCO in a molar ratio of 2:6:3:18 was
stirred in degassed DMF under a nitrogen atmosphere at room
temperature for a week giving a dark-red cloudy solution. This was
filtered through a frit with Celite and glasswool to give a dark-red
filtrate. The filtrate was evaporated in vacuo to give a dark-red solid
in over 75% yield. Redissolving ca. 20 mg of the product in 5 mL of
DMF gave a clear orange-red solution. Slow diffusion of diethyl ether
into this solution gave well-formed orange-red crystals of X-ray
quality.1H NMR (360 MHz, d7-DMF): δ 9.98 (flat, 4 × N-H), 8.31
(dd, J ) 7.6 and 1.8 Hz, 6 × 1 cat-H), 7.58 (broad, 6 × 4 phenyl-H),
7.35 (broad, 6 × 6 phenyl-H), 6.74 (dt, J ) 8.2 and 1.6 Hz, 6 × 1
cat-H), 6.50 (broad m, 6 × 1 cat-H), 3.16 (s, 4 × 12 DABCO-H).
13C{1H} NMR (89.9 MHz, d7-DMF): δ 157.0 (s, catechol), 152.1
(virtual t, JPC ) 10.0 Hz, catechol), 134.8 (virtual t, JPC ) 25.9 Hz,
phenyl), 134.6 (virtual t, JPC ) 5.3 Hz, phenyl), 129.6 (s, phenyl), 127.4
(virtual t, JPC ) 5.0 Hz, phenyl), 125.6 (virtual t, JPC ) 2.3 Hz,
catechol), 124.5 (virtual t, JPC ) 13.9 Hz, catechol), 113.9 (virtual t,
JPC ) 4.2 Hz, catechol), 112.6 (virtual t, JPC ) 27.2 Hz, catechol),
45.3 (s, DABCO). 31P{1H} NMR (36.3 MHz, d7-DMF): δ 20.73 (s).
IR (KBr): υ(N-H) 3425 cm-1, υ(aromatic C-H) 3051, 3006 cm-1, υ(-CH CH -)
Self -Assembly of (NH4)4[Ti2Pd3Br6L6]. A suspension of Ti(OMe)4,
PPh2-Catechol‚HBr, PdBr2‚2PhCN, and NH4OH in a molar ratio of
2:6:3:10 was stirred in degassed DMF under a nitrogen atmosphere at
room temperature for a week giving a dark-red cloudy solution that
was filtered through a frit filled with Celite and glasswool to give a
dark-red filtrate. The filtrate was evaporated under vacuum to give a
dark-red solid in over 80% yield. Redissolving about 20 mg of the
product in 10 mL of DMF gave a clear dark-red solution. Slow diffusion
of diethyl ether into this solution gave well-formed dark-red crystals
1
of X-ray quality. H NMR (360 MHz, d7-DMF): δ 8.42 (broad m, 6
× 1 cat-H), 7.57 (broad, 6 × 4 phenyl-H), 7.34 (broad, 6 × 6 phenyl-
H), 6.63 (broad d, J ) 7.9 Hz, 6 × 1 cat-H), 6.40 (broad, 6 × 1 cat-
H). 13C{1H} NMR (89.9 MHz, d7-DMF): δ 156.3 (s, catechol), 151.7
(virtual t, JPC ) 8.0 Hz, catechol), 135.1 (virtual t, JPC ) 14.6 Hz,
phenyl), 134.5 (broad, phenyl), 129.4 (broad, phenyl), 128.1 (broad,
phenyl), 127.4 (broad, catechol), 125.2 (m, catechol), 125.0 (m,
catechol), 113.7(broad, catechol). 31P{1H} NMR (36.3 MHz, d7-
DMF): δ 22.10 (s). IR (KBr): υ(NH4, N-H) 3326 cm-1, υ(aromatic C-H) 3052,
3024 cm-1; C-H bending 1468 cm-1, υ(C-O-Ti) 1259 cm-1. Anal. Calcd
for Ti2C108H94N4P6O12Pd3Br6: C, 47.69; H, 3.48; N, 2.06. Found: C,
47.52; H, 3.59; N, 2.17.
Cs4[Ti2L6{Cr(CO)4}3]. A DMF (20 mL) solution of Cs2[TiL3] (237
mg, 0.199 mmol) and Cr(CO)4(piperidine)2 (100 mg, 299 mmol) was
stirred under a nitrogen atmosphere at room temperature for 7 days
giving a clear orange-red solution. Addition of 50 mL of acetone
precipitated a yellow solid, which was filtered off and washed with
acetone (3 × 10 mL). The yellow solid was then dried under vacuum.
Yield: 210 mg (66%). 1H NMR (360 MHz, d7-DMF): δ 7.76 (broad,
6 × 1 cat-H), 7.49 (broad, 6 × 4 phenyl-H), 7.38 (broad, 6 × 6 phenyl-
H), 7.34 (broad, 6 × 1 cat-H), 6.14 (broad). 31P{ 1H} NMR (36.3 MHz,
d7-DMF): δ 72.09 (s). IR (KBr): υ(CO) 1856 cm-1; C-H bending 1474
cm-1, υ(C-O-Ti) 1260 cm-1. Anal. Calcd for Cs4Ti2C120H78P6O24Cr3: C,
50.16; H, 2.74. Found: C, 50.02; H, 2.79.
2
2
2952, 2883 cm-1, C-H bending 1478 cm-1, υ(C-O-Sn) 1253 cm-1. Anal.
Calcd for C132H130P6O12N8Sn2Pd3Br6: C, 48.90; H, 4.04; N, 3.46.
Found: C, 48.83; H, 3.96; N, 3.49.
Cs4[Ti2L6Pd3Cl6]. A DMF (10 mL) solution of Cs2[TiL3] (208 mg,
0.175 mmol) and PdCl2‚2PhCN (100 mg, 0.261 mmol) was stirred under
a nitrogen atmosphere at room temperature for 2 h giving a clear dark-
red solution. Addition of 100 mL of diethyl ether precipitated a dark-
red solid that was filtered off and dried in vacuo. Yield: 220 mg (74%).
1H NMR (360 MHz, d7-DMF): δ 7.86 (broad t, J ) 7.2 Hz, 6 × 1
cat-H), 7.63 (broad m, 6 × 4 phenyl-H), 7.40 (broad m, 6 × 6 phenyl-
H), 6.32 (broad, 6 × 1 cat-H), 6.23 (broad m, 6 × 1 cat-H). 13C{1H}
NMR (89.9 MHz, d7-DMF): δ 163.9 (s, catechol), 159.8 (virtual t,
JPC ) 10.0 Hz, catechol), 134.8 (virtual t, JPC ) 6.0 Hz, phenyl), 132.8
(virtual t, JPC ) 24.6 Hz, phenyl), 130.0 (s, phenyl), 127.8 (virtual t,
JPC ) 5.0 Hz, phenyl), 124.6 (s, catechol), 121.0 (m, catechol), 114.6
(virtual t, JPC ) 27.9 Hz, catechol), 111.2 (m, catechol). 31P{1H} NMR
(36.3 MHz, d7-DMF): δ 24.03 (s). IR (KBr): C-H bending 1475 cm-1
,
υ(C-O-Ti) 1264 cm-1. Anal. Calcd for Cs4Ti2C108H78P6O12Pd3Cl6: C,
44.53; H, 2.70. Found: C, 44.25; H, 2.51.
Cs4[Ti2L6Pd3I6]. Addition of CsI (491 mg, 1.89 mmol) to 20 mL of
DMF solution of Cs4[Ti2L6Pd3Br6] (200 mg, 0.0629 mmol) resulted in
a cloudy mixture, which was stirred overnight. The mixture was filtered
through a frit with Celite and glasswool, giving a dark-red filtrate that
Cs4[Sn2L6{Cr(CO)4}3]. A DMF (20 mL) solution of Cs2[SnL3] (251
mg, 0.199 mmol) and Cr(CO)4(piperidine)2 (100 mg, 0.299 mmol) was
stirred under a nitrogen atmosphere at room temperature for 7 days
giving a clear orange-red solution. Addition of 50 mL of acetone

M2M′3L6 Supramolecular Clusters
J. Am. Chem. Soc., Vol. 123, No. 12, 2001 2763
precipitated a yellow solid, which was filtered off and washed with
(refining on F). Hydrogen atoms were included but not refined.
Crystallographic data (excluding structure factors) have been deposited
with the Cambridge Crystallographic Data Centre as supplementary
publication numbers 139110 ((DABCO-H)4[Sn2Pd3Br6L6]) and 139159
(Cs2[TiL3]). Copies of the data can be obtained free of charge on
application to The Director, CCDC, 12 Union Road, Cambridge
CB21EZ, UK (fax: international code + (1223)336-033; e-mail:
deposit@chemcrys.cam.ac.uk).
acetone (3 × 10 mL). The yellow solid was then dried under vacuum.
1
Yield: 230 mg (74%). H NMR (360 MHz, d7-DMF): δ 7.71 (broad
m, 6 × 1 cat-H), 7.50 (broad, 6 × 4 phenyl-H), 7.39 (broad, 6 × 6
phenyl-H), 6.46 (broad d, J ) 4.0 Hz, 6 × 1 cat-H), 6.15 (broad, 6 ×
1 cat-H). 13C{1H} NMR (89.9 MHz, d7-DMF): δ 155.0 (s, catechol),
152.7 (virtual t, 9.0 Hz, catechol), 140.2 (virtual t, 5.6 Hz, phenyl),
132.6 (virtual t, 4.6 Hz, phenyl), 128.8 (s, phenyl), 128.4 (m, phenyl),
128.0 (virtual t, 4.0 Hz, catechol), 121.4 (m, catechol), 120.3 (m,
catechol), 112.8(m, catechol). 31P{1H} NMR (36.3 MHz, d7-DMF): δ
Acknowledgment. We thank the National Science Founda-
tion (CHE-9709621), NSF/NATO (exchange grant INT-9603212/
SRG951516), and Research Corporation (Research Opportunity
Grant to E.H.W.) for financial support. We also thank Dr.
Frederick J. Hollander for assistance with the crystallography
and Dr. Martin Michels for helpful discussions and a thorough
review of the manuscript.
71.55 (s). IR (KBr) υ(CO) 1855 cm-1 , C-H bending 1477 cm-1
(C-O-Sn) 1251 cm-1. Anal. Calcd for Cs4Sn2C120H78P6O24Cr3: C, 47.81;
H, 2.61. Found: C, 48.00; H, 2.65.
,
υ
X-ray Crystal Structure Analysis. Experimental crystal data for
Cs4[Sn2Pd3Br6L6] and Cs4[Ti2Pd3Br6L6] have been reported previ-
ously.36 Crystal data for the other two structures reported herein were
collected by using a Siemens SMART76 diffractometer equipped with
a CCD area detector with Mo KR (λ ) 0.71073 Å) radiation. Data in
the frames corresponding to an arbitrary hemisphere of data were
integrated by using SAINT.77 Data for both structures were corrected
for Lorentz and polarization effects. An empirical absorption correction
based on the measurement of redundant and equivalent reflections and
an ellipsoidal model for the absorption surface was applied with use
of SADABS.78 The structure solution and refinement for both structures
was performed by using the teXsan79 crystallographic software package
Supporting Information Available: The X-ray structure
reports of (DABCO-H)4[Sn2Pd3Br6L6] and Cs2[TiL3] (including
tables of crystal data, data collection parameters, and discussions
on modeling of the disorder in the molecules) (PDF). This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA0029376
(76) SMART, Area Detector Software Package; Siemens Industrial
Automation, Inc.: Madison, 1995.
(77) SAINT, SAX Area Detector Integration Program, 4.024 ed.; Siemens
Industrial Automation, Inc.: Madison, 1995.
(78) Sheldrick, G. SADABS, Siemens Area Detector ABSorption Cor-
rection Program; Advanced Copy ed.; Personal Communication, 1996.
(79) teXsan, Crystal Structure Analysis Package; Molecular Structure
Corporation: The Woodlands, TX, 1992.