Facile assembly of a Cu9 amido complex: A new tripodal ligand design that promotes transition metal cluster formation

Article abstract of DOI:10.1039/b517531c

A tripodal amido ligand with a central non-chelating phosphorus donor allows for the facile assembly of a pentane soluble organometallic copper cluster with a central copper atom surrounded by a nonplanar chain of eight copper atoms and two terminal amido

Full text of DOI:10.1039/b517531c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Facile assembly of a Cu9 amido complex: a new tripodal ligand design
that promotes transition metal cluster formation{
Alana L. Keen, Meghan Doster, Hua Han and Samuel A. Johnson*
Received (in Berkeley, CA, USA) 9th December 2005, Accepted 26th January 2006
First published as an Advance Article on the web 10th February 2006
DOI: 10.1039/b517531c
copper complexes of 1 to determine if polynuclear complexes of
the low-valent late transition metals would be accessible that might
exhibit interesting photophysical properties and reactivities.15
A tripodal amido ligand with a central non-chelating phos-
phorus donor allows for the facile assembly of a pentane
soluble organometallic copper cluster with a central copper
atom surrounded by a nonplanar chain of eight copper atoms
and two terminal amido–copper bonds.
The reaction of 1 equiv of [P(CH2NArCF )3]H3 with 1 equiv of
3
mesitylcopper was monitored at low temperature by 31P{1H}
NMR spectroscopy; the major product exhibited a single
phosphine environment, but the complex decomposed in solution,
even when stored at 240 uC, and could not be isolated. A second
small signal in the 31P{1H} NMR spectrum indicated the presence
The design of ligands that support transition metal clusters or
organized assemblies of metals has multiple goals. Such complexes
can exhibit novel reactivities based on cooperativity1 or electron
transfer between metal centres as well as unique physical properties
such as luminescence2–4 or magnetism5 that result from interac-
tions between metal centres. Unfortunately, the design of ligands
that can support clusters or assemblies of metals with a variety of
transition metals is rarely straightforward.
of
a minor product, which was more thermally stable.
Optimization of this reaction demonstrated that the production
of the more thermally stable complex required 2 equiv of
P(CH2NHArCF
3
) and 9 equiv of mesitylcopper, as shown in
3
eqn. 2. The resultant solid was moderately soluble in pentane, and
recrystallization at 240 uC provided [P(CH2NArCF )3]2Cu9(m-
3
Our approach to design of such a ligand was to incorporate
common donor ligands, such as amide and phosphine donors, into
a ligand framework such that they could not chelate to the same
metal centre. The reaction of P(CH2OH)3 with an excess of
2,4,6-Me3C6H2)3 (2) as bright orange crystals in a 54% yield.{
Complex 2 is stable as a solid or in solution at 240 uC for extended
periods of time. However, in solution at room temperature, the
complex decomposed over the course of days, which resulted in a
colour change of the solution from bright orange to dark brown.
3,5-bis(trifluoromethyl)aniline leads to the formation of
P(CH2NHArCF
3
) as a spectroscopically pure white powder in
3
an 80% yield, as shown in eqn. 1. Related compounds have
previously been tested as flame retardants, but never as ligand
precursors.6
ð1Þ
ð2Þ
With high-valent electropositive early transition metals, chela-
tion by the amido donors is anticipated. In this bonding mode the
phosphine lone pair is directed away from the metal chelated by
the amido donors, and is available to bind a second metal; related
tripodal ligands are known but lack this additional phosphine
donor, and in all related amido–phosphino ligands chelation by
both the amido and phosphine ligands is typically preferred.7–14 In
the alternate conformation shown as mode A in Fig. 1, the
phosphine and amide donors all bind in the same direction.
Chelation to a single metal is impossible, because the donor
orbitals are nearly parallel. Such a conformation could be
stabilized by, or possibly encourage, metal–metal interactions.
Attachment of the phosphine donors of more than one
The solid-state structure of 2 was determined by X-ray
crystallography, and a simplified ORTEP depiction is shown in
Fig. 2.§ The structure demonstrates that two ligand fragments
encapsulate a Cu9 core. At the centre of the cluster, Cu(1) is bound
[P(CH2NArCF )3] moiety to a single metal centre could potentially
3
be used as a tool to generate larger polymetallic complexes, as
shown in bonding mode B in Fig. 1. We decided to examine
Department of Chemistry & Biochemistry, University of Windsor,
Windsor, ON, Canada N9B 3P4. E-mail: sjohnson@uwindsor.ca;
Fax: 1 519 973 7098; Tel: 1 519 253 3000 (x3769)
{ Electronic supplementary information (ESI) available: Full experimental
details. See DOI: 10.1039/b517531c
Fig. 1 Two possible binding modes of the P(CH2NArCF
3
) ligand that
3
could lead to polynuclear complexes.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 1221–1223 | 1221

View Article Online
˚
Fig. 3 Summary of Cu–Cu distances in 2. Bond lengths shown in A;
˚
standard deviations are 0.0012–0.0013 A.
˚
3.8667(12) A preclude any significant interactions between these
pairs of metals.
Fig. 2 ORTEP depiction of the solid-state structure of 2, with 50%
probability ellipsoids. Hydrogen atoms are omitted. Only the ipso carbons
of the 3,5-bis(trifluoromethyl)phenyl substituents are shown. The 1,3,5
methyl substituents of the mesityl groups and short contacts between the
central and outer copper atoms are omitted for clarity.
1
Low-temperature H, 19F, and 31P{1H} NMR spectroscopy in
toluene verifies that the C2 structure observed in the solid-state is
maintained in solution. A combination of 1-D homonuclear
decoupling experiments at 273 K, and 2-D 1H–1H COSY and
EXSY/NOESY NMR data obtained at 228 K was used to fully
assign the 1H NMR resonances; surprisingly, long-range couplings
between aromatic protons were readily observed in the low
temperature 1H–1H COSY spectrum, whereas the 2-bond
couplings of the ligand methylene protons were more easily
assigned with the aid of 1-D homonuclear decoupling experiments.
Two fluxional processes were observed by variable-temperature
NMR spectroscopy over the temperature range of 213–303 K. One
of the ligand 3,5-(CF3)2C6H3 substituents experiences a moderate
barrier to rotation about the N–C bond, as observed by coalescence
of a pair of ortho protons in the 1H NMR spectra and a pair of CF3
environments in the 19F NMR spectra. The activation energy for this
barrier to rotation was calculated to be 24 kJ/mol; we assign this as
the 3,5-(CF3)2C6H3 substituent on the terminal amide, which should
have slight N–Cipso double bond character due to delocalization of
its lone pair onto the aromatic ring.
to the phosphine donors of 2 P(CH2NArCF )3 moieties, and no
3
anionic ligands. The remaining 8 Cu atoms nearly completely
encircle Cu(1) and are bound to the amido donors of the ligands
held together by Cu(1) and three mesityl groups. These 8 Cu atoms
at the periphery of the complex are bonded in a nonplanar zig-zag
chain via d10–d10 bonding interactions,16–20 with Cu–Cu bond
˚
distances ranging from 2.4657(13) to 2.5125(13) A. If the central
Cu is considered as Cu(I), the peripheral Cu8 fragment must bear a
delocalized negative charge. The Cu–Cu–Cu angles in the zig-zig
chain tend to be close to either 90u or 180u, with Cu(2)–Cu(3)–
Cu(4), Cu(3)–Cu(4)–Cu(5) and Cu(4)–Cu(5)–Cu(6) angles of
94.65(4), 161.53(5), and 94.38(4)u respectively. The complex has
no crystallographic symmetry, but approximate C2 symmetry. The
N(1)–Cu(2) and the N(4)–Cu(9) bonds are rare examples of
terminal amido–copper bonds.21,22 The terminal-amido–ipso-
carbon bonds are slightly shorter than the related bridging-
amido–ipso-carbon bonds; the terminal-amido–ipso-carbon
The second fluxional process was observed near room
1
temperature, where all of the resonances in the H NMR of 2
were exchange broadened. The complex appears to be achieving
higher apparent symmetry. At higher temperatures the rate of
decomposition of 2 becomes significant, which prevented a fast-
exchange spectrum from being obtained. The room-temperature
1H NMR spectrum was not affected by the concentration of 2,
which is evidence against an intermolecular exchange process.
Unfortunately, at 228 K, the EXSY/NOESY spectrum provided
little information regarding chemical exchange, because the cross
peaks due to nuclear Overhauser effects had the same phase as the
exchange cross peaks, as expected for a sizeable molecule with a
large rotational correlation time. 1H spin-saturation transfer
experiments performed at 273 K demonstrated that all ligand
arms are exchanged in the fluxional process responsible for the
˚
N(1)–C(4) and N(4)–C(31) distances are 1.372(9) and 1.385(9) A,
respectively, whereas the bridging-amido–ipso-carbon bonds are
˚
on average 0.043 A longer. Such shortening of terminal-amido–
ipso-carbon bonds has been observed before in electron-rich
amido complexes, including a related Cu complex with a slightly
21
shorter N–C bond length of 1.354(9) A. This shortened bond
˚
length can be attributed to increased delocalization of the amido
lone pair into the aromatic ring, due to its destabilization via
antibonding interactions with occupied metal d-orbitals.23 Such
destabilizing interactions typically render late transition metal
amides highly nucleophilic and basic.24–26 In complex 2, the
p-systems of the aromatic rings associated with the terminal
amides are aligned appropriately to overlap with the lone pair of
electrons on the terminal-amido moieties.
1
broadening of the resonances in the H NMR spectrum at room
temperature. The mesityl substituent environments are also
exchanged. Such a fluxional process appears to require a breaking
of some of the Cu–Cu bonding interactions; however, inadequate
evidence is available from the 1H NMR data to provide a detailed
mechanism. A dissociative mechanism in which mesityl copper is
lost from 2 was ruled out by adding mesitylcopper to a solution of
2. The lack of line broadening in the mesitylcopper resonances of
the room temperature 1H NMR, combined with the results of 1H
NMR spin-saturation transfer experiments, indicates that any
The Cu–Cu distances are summarized in Fig. 3. Aside from
coordination to the two phosphine donors, the central Cu(1)
appears to have relatively short interaction with 6 of the 8 Cu
centres that encircle it (all except Cu(5) and Cu(6)); however, all of
these interactions are approximately 10 to 15% longer than the
Cu–Cu interactions in the 8-membered chain. For clarity these
short contacts are not drawn in the ORTEP depiction in Fig. 2.
The Cu(1)–Cu(5) and Cu(1)–Cu(6) distances of 3.8634(12) and
1222 | Chem. Commun., 2006, 1221–1223
This journal is ß The Royal Society of Chemistry 2006

View Article Online
§ A suitable single crystal of 2 was covered with paratone and mounted in
the 143 K N2 stream of a Bruker AXS P4/SMART 1000 CCD
chemical exchange process involving mesitylcopper is too slow to
account for the fluxional behaviour observed near room
temperature.
˚
diffractometer equipped with a Mo Ka radiation (l = 0.71073 A) source.
The structure was solved using full matrix least squares on F2. Crystal data
for 2: C96H99Cu9F36N6P2, monoclinic, a = 20.5700(30), b = 21.1510(30), c =
The absorption spectrum for 1 displays an absorbance (lmax) at
373 nm, and the emission spectrum displays a maximum at 418 nm.
The luminescence occurred with a quantum yield (W) of 0.27 in
toluene at 298 K, relative to 9,10-diphenylanthracene (W = 0.90).
This value is slightly lower than the recently reported copper amide
dimer of Peters.15 This reduced luminescence is not surprising,
considering the increased flexibility of 2, and the relatively lesser
steric bulk of the supporting ligand.
˚
28.4010(30) A, a = 90.000(0)u, b = 119.442(8)u, c = 90.000(0)u, V =
3
10761(2) A , Z = 4, Dcalcd = 1.638 g cm23. A total of 101105 reflections
˚
were collected of which 18926 were unique; wR2 = 0.167, R = 0.075. CCDC
292735. For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b517531c
1 B. Bosnich, Inorg. Chem., 1999, 38, 2554–2562.
2 V. W. W. Yam and K. K. W. Lo, Mol. Supramol. Photochem., 1999, 4,
31–112.
Although a few examples of amido complexes of Cu(I) are
known,27–35 the ease by which 2 assembles, and its structural
integrity in solution are remarkable. Most copper amido
complexes adopt dinuclear, trinuclear or tetranuclear structures.
Closely related tripodal silane ligands, which lack the central
phosphine in 1 have also been observed to form anticipated
trinuclear copper complexes with bridging amido ligands.36
Usually, bulky ligands and careful synthetic methodology are
required to synthesize terminal copper–amido bonds.21 We are
currently investigating the reactivity of this complex, to determine
if the terminal amido ligands display the strong nucleophilicity and
basicity common to electron-rich late-transition metals, and the
ability of this ligand to provide polymetallic complexes with other
transition metals.
3 P. C. Ford, E. Cariati and J. Bourassa, Chem. Rev., 1999, 99,
3625–3647.
4 V. W.-W. Yam and K. K.-W. Lo, Chem. Soc. Rev., 1999, 28, 323–334.
5 L. K. Thompson, Coord. Chem. Rev., 2002, 233–234, 193–206.
6 A. W. Frank and G. L. Drake, Jr, J. Org. Chem., 1972, 37, 2752–2755.
7 R. R. Schrock, S. W. Seidel, Y. Schrodi and W. M. Davis,
Organometallics, 1999, 18, 428–437.
8 L. H. Gade, Acc. Chem. Res., 2002, 35, 575–582.
9 L. Turculet and T. D. Tilley, Organometallics, 2002, 21, 3961–3972.
10 L. Jia, E. Ding, A. L. Rheingold and B. Rhatigan, Organometallics,
2000, 19, 963–965.
11 M. D. Fryzuk, Can. J. Chem., 1992, 70, 2839–2845.
12 M. D. Fryzuk, J. B. Love and S. J. Rettig, Chem. Commun., 1996,
2783–2784.
13 M. D. Fryzuk, S. A. Johnson and S. J. Rettig, J. Am. Chem. Soc., 1998,
120, 11024–11025.
14 L.-C. Liang, J.-M. Lin and C.-H. Hung, Organometallics, 2003, 22,
3007–3009.
15 S. B. Harkins and J. C. Peters, J. Am. Chem. Soc., 2005, 127, 2030–2031.
16 P. Pyykko¨, Chem. Rev., 1997, 97, 597–636.
17 P. K. Mehrotra and R. Hoffmann, Inorg. Chem., 1978, 17, 2187–2189.
18 K. M. Merz, Jr. and R. Hoffmann, Inorg. Chem., 1988, 27, 2120–2127.
19 P. Belanzoni, M. Rosi, A. Sgamellotti, E. J. Baerends and C. Floriani,
Chem. Phys. Lett., 1996, 257, 41–48.
Acknowledgement is made to the National Sciences and
Engineering Council (NSERC) of Canada and the Ontario
Research and Development Challenge Fund (University of
Windsor Centre for Catalysis and Materials Research) for their
financial support.
20 H. L. Hermann, G. Boche and P. Schwerdtfeger, Chem.–Eur. J., 2001,
7, 5333–5342.
Notes and references
21 E. D. Blue, A. Davis, D. Conner, T. B. Gunnoe, P. D. Boyle and
P. S. White, J. Am. Chem. Soc., 2003, 125, 9435–9441.
22 L. A. Goj, E. D. Blue, C. Munro-Leighton, T. B. Gunnoe and
J. L. Petersen, Inorg. Chem., 2005, 44, 8647–8649.
23 K. G. Caulton, New J. Chem., 1994, 18, 25–41.
24 J. R. Fulton, A. W. Holland, D. J. Fox and R. G. Bergman, Acc. Chem.
Res., 2002, 35, 44–56.
{ Characterization data for [P(CH2NArCF )3]H3 (1): 1H NMR (C6D6,
3
3
2
298 K, 300 MHz): d 2.55 (dd, JHH = 5.2 Hz, JPH = 5.2 Hz, 6H, CH2),
3.13 (br, 3H, NH), 6.59 (s, 6H, o-H), 7.27 (s, 3H, p-H). 13C{1H} NMR
(C6D6, 298 K, 125.8 MHz): d 39.4 (d, JPC = 12.2 Hz, PCH2), 111.2 (s, o-C),
112.5 (s, p-C), 122.1 (s, m-C), 132.9 (q, J = 32.9 Hz, CF3), 148.9 (d, J =
5.5 Hz, ipso-C). 31P{1H} NMR (C6D6, 121.5 MHz, 298 K): d 232.6 (s).
19F NMR (C6D6, 298 K, 282.48 MHz): d 14.71 (s). Anal. Calc’d for
C27H18F18N3P: C, 42.82; H, 2.40; N, 5.55. Found: C, 43.00; H, 2.49; N,
5.41.
25 M. D. Fryzuk and C. D. Montgomery, Coord. Chem. Rev., 1989, 95,
1–40.
26 H. E. Bryndza and W. Tam, Chem. Rev., 1988, 88, 1163–1188.
27 S. B. Harkins and J. C. Peters, J. Am. Chem. Soc., 2004, 126, 2885–2893.
28 M. Niemeyer, Acta Crystallogr., Sect. E, 2001, 57, m491–m493.
29 T. Tsuda, K. Watanbe, K. Miyata, H. Yamamoto and T. Saegusa,
Inorg. Chem., 1981, 20, 2728–2730.
30 A. M. James, R. K. Laxman, F. R. Fronczek and A. W. Maverick,
Inorg. Chem., 1998, 37, 3785–3791.
31 H. Chen, M. M. Olmstead, S. C. Shoner and P. P. Power, J. Chem. Soc.,
Dalton Trans., 1992, 451–457.
32 H. Link, P. Reiss, S. Chitsaz, H. Pfistner and D. Fenske, Z. Anorg. Allg.
Chem., 2003, 629, 755–768.
33 P. Reib and D. Fenske, Z. Anorg. Allg. Chem., 2000, 626, 2245–2247.
34 H. Eriksson and M. Hakansson, Organometallics, 1997, 16, 4243–4244.
35 S. Gambarotta and M. Bracci, J. Chem. Soc., Dalton Trans., 1987,
1883–1888.
Characterization data for [P(CH2NArCF )3]2Cu9(m-2,4,6-Me3C6H2)3 (2):
3
1H NMR (C7D8, 243 K, 300 MHz) assigned using 1H–1H COSY and
NOESY and identified by the crystal structure atom labels: d 1.75 (s, 3H,
CH3, Mes-p-C71), 2.01 (s, 6H, CH3, Mes-p-C62), 2.20 (s, 6H, CH3, Mes-
o-C63), 2.41 (s, 6H, CH3, Mes-o-C70), 2.63 (s, 6H, CH3, Mes-o-C61), 3.24
2
(overlapping, 4H, CH2-H3a/H2a), 3.51 (d, 2H, JHH = 13.5 Hz, CH2-1a),
3.63 (d, 2H, 2JHH = 13.5 Hz, CH2-1b), 3.78 (d, 2H, 2JHH = 12.5 Hz, CH2-
3b), 4.10 (d, 2H, 2JHH = 12.5 Hz, CH2-2b), 6.15 (s, 2H, terminal N-o-H ),
6.24 (s, 2H, Mes-m-H59), 6.30 (s, 2H, Mes-m-H66), 6.56 (s, 2H, Mes-
m-H57), 6.90 (s, 2H, terminal N-o-H), 7.26 and 7.27 (s, 3H total, bridging
N-p-H and terminal N-p-H), 7.30 (overlapping s, 8H total, bridging
N-o-H), 7.44 (s, 2H, bridging N-p-H). 31P{1H} NMR (C6D6, 300 K,
121.54 MHz): 9.23 (s). 19F NMR (C7D8, 228 K, 282.48 MHz): d 14.78 (s,
6F), 14.80 (s, 6F), 15.20 (br s, 12 F, non-terminal aryl CF3), 15.30 (s, 6F),
15.42 (s, 6F). Anal. Calc’d for C81H66Cu9F36N6P2: C, 39.85; H, 2.73; N,
3.44. Found: C, 39.98; H, 2.50; N, 3.40. UV/VIS: e = 33980 L mol21 cm21
.
36 M. Veith, O. Schutt and V. Huch, Z. Anorg. Allg. Chem., 1999, 625,
1155–1164.
lmax = 373 nm. lemit = 418 nm.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 1221–1223 | 1223