Structural aspects, IR study, and molecular photophysics of the cumulene-bridged complexes with Rhenium(I), Ruthenium(II), and Osmium(II) centers

Article abstract of DOI:10.1021/ic0006910

The synthesis of a series of Re^I, Ru^II, and Os^II complexes that contain rigid polyphosphine/cumulene spacers is reported here. These cumulenic ligands, namely, 1,1′,3,3′ -tetrakis(diphenylphosphino)allene (C₃P₄) and 1,1′,4,4′-tetrakis(diphenylphosphino)cumulene (C₄P₄), utilize diphenylphosphino linkage components to coordinate to the metal-polypyridyl or metal-carbonyl units. Characterization of all mono-, homo-, and heterobimetallic complexes is achieved using ³¹P{¹H} NMR, IR, and fast atom bombardment mass spectroscopy (FAB/MS) and elemental analysis. The two Re^I homobimetallic complexes were also characterized by single-crystal X-ray structure determination, which provided the structural evidence of a 90° rotation between the C₃ and C₄ adducts causing a change in the electrochemical behavior. The ground-state electronic absorption and redox interactions, along with the excited-state photophysical characteristics, are also explored. Electrochemical studies showed that an increase in the carbon chain length resulted in a greater amount of σ-donation from the ligand to the metal centers, as well as a greater amount of electronic communication between the metal termini of the bimetallic species. The electronic absorption and emission spectra of the new complexes were also determined and characterized. The lifetimes of the excited-state luminescence of the Re^I mono- and homobimetallic complexes were found to be an order of magnitude shorter than the lifetimes of the heterobimetallic complexes containing the Ru^II and Os^II moieties. Excited-state energy transfer was observed from the higher MLCT excited state of the Re^I centers to the lower energy MLCT excited state of the Ru^II and Os^II centers on the following basis: no Re^I-based emission was detected in the steady-state emission measurements, the time-resolved decay traces were fitted to only single-exponential decays, and the quantum yields were identical for each compound at two different excitation wavelengths where different percentages of the metal-based chromophores were excited.

Full text of DOI:10.1021/ic0006910

6038
Inorg. Chem. 2000, 39, 6038-6050
Structural Aspects, IR Study, and Molecular Photophysics of the Cumulene-Bridged
Complexes with Rhenium(I), Ruthenium(II), and Osmium(II) Centers
Jeffrey V. Ortega,† Kay Khin, Wytze E. van der Veer, Joseph Ziller, and Bo Hong*
Department of Chemistry, University of California, Irvine, California 92697-2025
ReceiVed June 23, 2000
The synthesis of a series of ReI, RuII, and OsII complexes that contain rigid polyphosphine/cumulene spacers is
reported here. These cumulenic ligands, namely, 1,1′,3,3′-tetrakis(diphenylphosphino)allene (C3P4) and 1,1′,4,4′-
tetrakis(diphenylphosphino)cumulene (C4P4), utilize diphenylphosphino linkage components to coordinate to the
metal-polypyridyl or metal-carbonyl units. Characterization of all mono-, homo-, and heterobimetallic complexes
is achieved using 31P{1H} NMR, IR, and fast atom bombardment mass spectroscopy (FAB/MS) and elemental
analysis. The two ReI homobimetallic complexes were also characterized by single-crystal X-ray structure
determination, which provided the structural evidence of a 90° rotation between the C3 and C4 adducts causing
a change in the electrochemical behavior. The ground-state electronic absorption and redox interactions, along
with the excited-state photophysical characteristics, are also explored. Electrochemical studies showed that an
increase in the carbon chain length resulted in a greater amount of σ-donation from the ligand to the metal centers,
as well as a greater amount of electronic communication between the metal termini of the bimetallic species. The
electronic absorption and emission spectra of the new complexes were also determined and characterized. The
lifetimes of the excited-state luminescence of the ReI mono- and homobimetallic complexes were found to be an
order of magnitude shorter than the lifetimes of the heterobimetallic complexes containing the RuII and OsII
moieties. Excited-state energy transfer was observed from the higher MLCT excited state of the ReI centers to
the lower energy MLCT excited state of the RuII and OsII centers on the following basis: no ReI-based emission
was detected in the steady-state emission measurements, the time-resolved decay traces were fitted to only single-
exponential decays, and the quantum yields were identical for each compound at two different excitation
wavelengths where different percentages of the metal-based chromophores were excited.
Introduction
can be fine-tuned and investigated. To achieve this, the spacers
utilized must have desired characteristics such as rigidity,
photostability, and molecular orbitals with suitable energies to
overlap with those of the tethered metal units.1,4-6 Under the
appropriate conditions, the degree of interaction between the
The combination of metal-based units with rigid organic
spacers constitutes a fundamental class of functional supramo-
lecular systems, namely, “molecular wires” or rigid rodlike
molecules.1-10 With the incorporation of redox-active and
photoactive metal-based chromophores, functional supramo-
lecular systems can be constructed, and their redox character-
istics, excited-state properties, and signal transport processes
(4) (a) Benniston, A. C.; Harriman, A.; Grosshenny, V.; Ziessel, R. New
J. Chem. 1997, 21, 405-408. (b) Grosshenny, V.; Harriman, A.;
Ziessel, R. Angew. Chem., Int. Ed. Engl. 1995, 34, 2705-2708. (c)
Benniston, A. C.; Grosshenny, V.; Harriman, A.; Ziessel, R. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1884-1885. (d) Grosshenny, V.;
Ziessel, R. J. Chem. Soc., Dalton Trans. 1993, 817-819. (e) Ziessel,
R.; Juris, A.; Venturi, M. Inorg. Chem. 1998, 37, 5061-5069. (f)
Harriman, A.; Ziessel, R. Chem. Commun. 1996, 1707-1716. (g)
Grosshenny, V.; Harriman, A.; Gisselbrecht, J. P.; Ziessel, R. J. Am.
Chem. Soc. 1996, 118, 10315-10316. (h) Grosshenny, V.; Harriman,
A.; Ziessel, R. Angew. Chem., Int. Ed. Engl. 1995, 34, 1100-1102.
(5) Bunz, U. H. F. Angew. Chem., Int. Ed. Engl. 1996, 35, 969-971.
(6) Hong, B. Comments Inorg. Chem. 1999, 20, 177-207.
(7) (a) Coat, F.; Guillevic, M. A.; Toupet, L.; Paul, F.; Lapinte, C.
Organometallics 1997, 16, 5988-5998. (b) Guillemot, M.; Toupet,
L.; Lapinte, C. Organometallics 1998, 17, 1928-1930. (c) Le Narvor,
N.; Toupet, L.; Lapinte, C. J. Am. Chem. Soc. 1995, 117, 7129-7138.
(d) Le Narvor, N.; Lapinte, C. J. Chem. Soc., Chem. Commun. 1993,
357-359. (e) Coat, F.; Lapinte, C. Organometallics 1996, 15, 477-
479.
(8) Woodworth, B. E.; White, P. S.; Templeton, J. L. J. Am. Chem. Soc.
1997, 119, 828-829.
(9) (a) El-ghayoury, A.; Harriman, A.; Khatyr, A.; Ziessel, R. Angew.
Chem., Int. Ed. 2000, 39, 185-189. (b) Harriman, A.; Ziessel, R.
Coord. Chem. ReV. 1998, 171, 331-339. (c) Ziessel, R.; Hissler, M.;
El-Ghayoury, A.; Harriman, A. Coord. Chem. ReV. 1998, 180, 1251-
1298.
(10) (a) Derege, P. J. F.; Therien, M. J. Inorg. Chim. Acta 1996, 242, 211-
218. (b) Lin, V. S. Y.; Therien, M. J. Chem. Eur. J. 1995, 1, 645-
651.
* To whom all correspondence should be addressed.
† Currently at Department of Chemistry, Northwestern University,
Evanston, IL.
(1) (a) Falloon, S. B.; Szafert, S.; Arif, A. M.; Gladysz, J. A. Chem. Eur.
J. 1998, 4, 1033-1042. (b) Weng, W. Q.; Bartik, T.; Brady, M.; Bartik,
B.; Ramsden, J. A.; Arif, A. M.; Gladysz, J. A. J. Am. Chem. Soc.
1995, 117, 11922-11931. (c) Bartik, B.; Dembinski, R.; Bartik, T.;
Arif, A. M.; Gladysz, J. A. New J. Chem. 1997, 21, 739-750. (d)
Falloon, S. B.; Weng, W. Q.; Arif, A. M.; Gladysz, J. A. Organo-
metallics 1997, 16, 2008-2015. (e) Brady, M.; Weng, W. Q.; Zhou,
Y. L.; Seyler, J. W.; Amoroso, A. J.; Arif, A. M.; Bohme, M.;
Frenking, G.; Gladysz, J. A. J. Am. Chem. Soc. 1997, 119, 775-788.
(f) Brady, M.; Weng, W. Q.; Gladysz, J. A. J. Chem. Soc., Chem.
Commun. 1994, 2655-2656. (g) Falloon, S. B.; Arif, A. M.; Gladysz,
J. A. Chem. Commun. 1997, 629-630. (h) Bartik, T.; Bartik, B.; Brady,
M.; Dembinski, R.; Gladysz, J. A. Angew. Chem., Int. Ed. Engl. 1996,
35, 414-417.
(2) Mayr, A.; Yu, M. P. Y.; Yam, V. W. W. J. Am. Chem. Soc. 1999,
121, 1760-1761.
(3) (a) Ortega, J. V.; Hong, B.; Ghosal, S.; Hemminger, J. C.; Breedlove,
B.; Kubiak, C. P. Inorg. Chem. 1999, 38, 5102-5112. (b) Hong, B.;
Woodcock, S. R.; Saito, S. K.; Ortega, J. V. J. Chem. Soc., Dalton
Trans. 1998, 2615-2623. (c) Hong, B.; Ortega, J. V. Angew. Chem.,
Int. Ed. 1998, 37, 2131-2134.
10.1021/ic0006910 CCC: $19.00 © 2000 American Chemical Society
Published on Web 12/01/2000

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6039
metal termini can be probed by studying parameters such as
the rate of photoinduced energy/electron transfer, the electronic
coupling matrix element (HAB), and the comproportionation
constant (Kc). These rigid rodlike molecules may serve as media
to explore the basic properties that govern the fundamental
aspects of electronic communication and photophysical char-
acteristics in molecular systems.
resulted in the facilitation of rapid intramolecular energy transfer.
Porphyrins have also been combined with rigid acetylenic chains
by Therien and co-workers to coordinate with different metal
centers.10 These are examples describing the versatility that can
be engineered with the utilization of different coordination or
linkage units.
Recently our group has become interested in the development
of rigid rodlike systems with ReI, RuII, and OsII termini and
ditopic/tetratopic polyphosphines with cumulenic and acetylenic
carbon chains. Herein, we report the synthesis, X-ray structure
determination, redox characteristics, and photophysical studies
of a new family of monometallic (ReCnP4, n ) 3, 4),
homobimetallic (ReCnP4Re), and heterobimetallic species
(ReCnP4M, M ) Ru, Os) with ReI, RuII, and/or OsII centers.
The polyphosphine spacers, namely, 1,1′,3,3′-tetrakis(diphenyl-
phosphino)allene (C3P4) and 1,1′,4,4′-tetrakis(diphenylphos-
phino) cumulene (C4P4), are used to link together the metal
centers. The sp-hybridized carbon chains in the spacers not only
control the conformational rigidity of the overall complex but
also determine the planarity and the distance between the two
terminal units. The mono-, homo-, and heterobimetallic com-
plexes with only the RuII and/or OsII centers were described
elsewhere.3,6 In this study, we report the first structural evidence
of a 90° rotation between the ReI homobimetallic C3 and C4
adducts that were initially demonstrated by the electrochemical
studies of the homobimetallic RuII and OsII adducts. The redox
chemistry and excited-state properties are discussed and com-
pared with the analogous monometallic species as well as the
model complex containing 1,1′-bis(diphenylphosphino)ethene
(C2P2e). Other characterization techniques, including infrared
spectroscopy (IR), 31P{1H} nuclear magnetic resonance (NMR)
spectroscopy, fast atom bombardment mass spectroscopy (FAB/
MS), and elemental analysis (EA), were also used to structurally
characterize all new complexes. In addition, the ground-state
electronic absorption properties, the excited-state decay kinetics,
and the energy transfer process (using ReI/RuII and ReI/OsII
donor/acceptor pairs) are also explored in this system with short
sp carbon chains.
Specifically, the linear, one-dimensional forms of carbon have
attracted growing interest as suitable molecular building blocks
for the construction of supramolecular systems, especially the
rigid rodlike molecules.1-5 These carbon chains can exist as
sp-hybridized carbon atoms either in an uninterrupted series of
double bonds (cumulenes) or as alternating triple and single
bonds (polyynes). They can serve as passive bridging component
or active conduit for signal transport between various combina-
tions of metal-based moieties. The unsaturation can cause these
carbon chains to exhibit extraordinary properties.1c-d,4b Such
properties are manifested in the rigid molecule’s ability to
convey electrons efficiently from one end to the other as well
as the capability of unidirectional electron/energy transfer. In
addition, by “capping” the rigid molecular rods with photoactive
and redox-active metal-based units, it is then possible to study
the electronic, optical, and electrochemical interactions between
two linked organometallic centers.1e,f,3,4c-d,6 These metal-capped
carbon chains are of great interest due to their potentially useful
materials properties. Conjugated carbon rods are expected to
have a high degree of mechanical strength and stability, which
is important for making robust and resilient materials.5 Their
expected electrical conductivity could lead to applications such
as “molecular wires” in electronic devices.1g,4b-h,5 This property
would also render them useful for studying how charge is
transported from one metal-based moiety to another over a long
molecular distance.
Several laboratories have focused on the development of one-
dimensional molecular wires. For the majority of the research
in this area, the focus of the investigation has been primarily
on the acetylenic form of sp carbon chains.1,2,4-11 Lapinte and
co-workers have bridged two iron centers by an elemental four-
carbon chain7a-d and an eight-carbon chain7e consisting of sp-
hybridized polyynes that both facilitated a strong coupling
between FeII/FeII and FeII/FeIII centers. Gladysz and co-workers1h
have synthesized systems with rigid acetylenic carbon chains
between metal centers and a chain length as long as 20 carbons.
Gladysz11 has also developed a system in situ that contained a
cumulenic five-carbon rigid spacer that acted as a Fischer
carbene and facilitated the charge transfer from ReI to MnI.
Templeton and co-workers8 have bridged molybdenum and
tungsten tris(pyrazolyl)borate carbonyl units by a C4 cumulenic
carbon chain that converted into an acetylenic bridge with
metal-carbon triple bonds. In those complexes, the coordination
site is the Cn chain itself, where the terminal carbon atoms are
bonded directly to the metal centers. However, no excited-state
properties have been reported for such systems.
Other types of linkage components have been incorporated
to couple the rigid acetylenic bridge with metal centers.
Harriman and Ziessel9 have utilized polypyridyl units connected
to alkyne bridges to coordinate to metal centers. Specifically,
both terpyridyl and bipyridyl groups were used in conjunction
with C2 and C4 acetylenic units. This combination of π-conju-
gated acetylenic groups with the aryl polypyridyl groups has
Experimental Section
General Methods. All experiments were carried out under a nitrogen
atmosphere using standard glovebox and Schlenk techniques. All
purifications by column chromatography were performed in the absence
of light using basic alumina and either CHCl3 or MeCN as eluent. The
monometallic species were collected as the first fraction, while the
bimetallic complexes were collected as the second fraction.
(11) (a) Weng, W. G.; Bartik, T.; Gladysz, J. A. Angew. Chem., Int. Ed.
Engl. 1994, 33, 2199-2202. (b) Bartik, T.; Weng, W. Q.; Ramsden,
J. A.; Szafert, S.; Falloon, S. B.; Arif, A. M.; Gladysz, J. A. J. Am.
Chem. Soc. 1998, 120, 11071-11081.

6040 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
Materials. 1,1′,3,3′-Tetrakis(diphenylphosphino)allene (C3P4),12a
1,1′,4,4′-tetrakis(diphenylphosphino)cumulene (C4P4),12b 1,1′-bis(diphe-
of C2P2e (155.1 mg, 0.392 mmol) in THF (10 mL) over 30 min. After
stirring for 24 h at reflux, the reaction mixture was cooled to room
temperature before the solvent was removed by rotary evaporation. The
yellow product was extracted using MeCN. After the solvent was
removed, the yield was quantitative. 31P{1H} NMR (202.5 MHz,
benzene-d6): δ -8.41 (s) ppm. FT-IR (KBr): νCO 2029 (vs), 1952
(vs), 1895 (vs) cm-1. Anal. Calcd for C29H22O3P2ClRe: C, 49.61; H,
3.15. Found: C, 49.71; H, 3.10.
Cl(CO)3Re(C3P4) (ReC3P4). A solution of Re(CO)5Cl (60.7 mg,
0.168 mmol) in THF (10 mL) was added to a refluxing solution of
C3P4 (236.1 mg, 0.300 mmol) in THF (20 mL). After refluxing for 24
h, the reaction mixture was cooled to room temperature. The volume
of the solution was reduced down to ∼2 mL before being added
dropwise into a stirred solution of hexanes (50 mL). The resulting
yellow precipitate was collected by filtration, washed with 3 × 15 mL
of hexanes, and dried in Vacuo. Purification was performed by column
chromatography using basic alumina as the support and CHCl3 as the
eluent, and the product was collected as the first fraction. Yield: 124.4
mg (68.4%). 31P{1H} NMR (202.5 MHz, acetone-d6): δ 22.80 (s),
-13.30 (s) ppm. FT-IR (KBr): νCO 2026 (vs), 1944 (vs), 1991 (vs)
cm-1. Anal. Calcd for C54H40O3P4ClRe: C, 59.91; H, 3.72. Found: C,
59.75; H, 3.68.
nylphosphino)ethene (C2P2e),12c and cis-Os(bpy)2Cl2 were prepared
13
according to literature methods. [(bpy)2M(CnP4)](PF6)2 (M ) Ru, Os;
n ) 3, 4) were synthesized as previously described.3 cis-Ru(bpy)2Cl2
and Re(CO)5Cl were purchased from Strem and used as received. All
spectroscopic grade solvents were purchased from Fisher and used
without further purification. Tetrahydrofuran (THF) was distilled under
nitrogen over sodium benzophenone ketyl. Acetonitrile (MeCN) was
distilled under nitrogen over calcium hydride.
Physical Measurements. Electrochemical Analysis. Cyclic volta-
mmograms were recorded on a CHI 630 electrochemical analyzer. The
redox behavior can be monitored in “dry” aprotic solvents that contain
a supporting electrolyte. Typical experiments were run at 100 mV/s in
acetonitrile with 0.1 M tetrabutylammonium hexafluorophosphate as
the electrolyte. A Ag/AgCl wire was used as a pseudoreference
electrode, a platinum wire as the counter electrode, and a 1.0 mm
platinum disk electrode as the working electrode. Great care was taken
to remove oxygen as well as water in order to ensure the quality of the
measurements. A ferrocene standard was used to reference the observed
potential vs SCE.
NMR, IR, and Mass Spectral Analysis. 31P{1H} NMR spectra were
obtained either in acetone-d6 or in acetonitrile-d3 on an Omega 500
MHz spectrometer, referenced to an external standard solution of 85%
H3PO4 in D2O. Infrared absorption (IR) spectra (KBr pellets) were
collected as the average of 16 scans by using a Nicolet Impact Model
410 FT-IR with a DTGS (dueterated triglyme sulfate) detector. Fast
atom bombardment mass spectral analysis (FAB/MS) data were
obtained on a Micromass (Altrincham, UK) Autospec mass spectrometer
at the UC, Irvine Mass Spectral Laboratory. Cesium ions at 25 kV
were the bombarding species, and the matrix was meta-nitro benzyl
alcohol (mNBA).
Photophysical Measurements. The electronic absorption spectra
were recorded on a Hewlett-Packard 8453 diode array spectrophotom-
eter. Steady-state emission spectra were obtained on a Hitachi 4500
fluorescence spectrometer or an AMINCO-Bowman Series 2 lumines-
cence spectrometer. Luminescence quantum yields of all complexes
were measured in spectrograde acetonitrile relative to [Ru(bpy)3][PF6]2
(Φ ) 0.06214 in acetonitrile).
The time-resolved emission spectroscopic studies were performed
on a nanosecond laser flash photolysis unit equipped with a Continuum
Surelite II-10 Q-switched Nd:YAG laser and a Surelite OPO (optical
parametric oscillator) tunable visible source, a LeCroy 9350A oscil-
loscope, and a Spex 270 MIT-2x-FIX high-performance scanning and
imaging spectrometer. Only lifetimes longer than 6 ns were measured
accurately on this setup. For lifetimes shorter than 6 ns, they were
obtained on a SLM-Aminco 48000 MHF Fourier transform spectro-
fluorometer. As a light source, the 488 nm line of a Coherent Innova
90 argon ion laser was used. The laser beam was modulated with a
comb function with an interval spacing of 5 MHz, with a maximum
frequency of 250 MHz. The resulting beam was imaged on the sample;
the resulting fluorescence was detected with a photomultiplier via a
Schott filter (OG 515). The phase and intensity of each component of
the comb function was determined; the required reference signal was
obtained by the utilization of a small portion of the incident beam.
The resulting signals were fitted with a single exponential, which
obtained the best fit with respect to both the recorded phase and intensity
information. For each sample a series of five measurements was
obtained, each consisting of 1000 scans.
Cl(CO)3Re(C3P4)Re(CO)3Cl (ReC3P4Re). A solution containing
both Re(CO)5Cl (48.0 mg, 0.132 mmol) and C3P4 (50.0 mg, 0.064
mmol) in THF (20 mL) was heated at reflux 24 h. Upon completion,
the solution was allowed to cool to room temperature, and the solvent
reduced down to ∼2 mL with the use of a rotary evaporator. The 2
mL sample was added dropwise into a stirred solution of hexanes (50
mL). The resulting yellow precipitate was collected by filtration, washed
with 3 × 15 mL of hexanes, and dried in Vacuo. Purification was
performed by column chromatography using basic alumina as the
support and CHCl3 as the eluent, and the product was collected as the
second fraction. Crystals were obtained by first dissolving the yellow
powder in a minimum amount of THF and then adding an equivalent
amount of benzene. The solution mixture was stored at room temper-
ature in a vented container. As the THF evaporated, yellow needles
formed. Yield: 80.0 mg (87.3%). 31P{1H} NMR (202.5 MHz, acetone-
d6): δ -1.43 (d), -12.75 (d) ppm. 13C NMR (125.7 MHz, acetone-
d6): δ 136.23 and 135.25 (d, JCP ) 11.3-12.6 Hz, phenyls Ci and
Ci′), 134.47 and 131.44 (d, JCP ) 2.0 Hz, phenyls Cp and Cp′), 133.68
and 131.98 (d, JCP ) 12.6-13.9 Hz, dCP2), 132.70 (dd, JCP ) 17.6
and 2.3 Hz, CdCdC), 130.15-129.68 (m, phenyls Cm, Cm′, Co, and
Co′). FT-IR (KBr): νCO 2026 (vs), 1946 (vs), 1914 (vs), 1878 (w) cm-1
.
Anal. Calcd for C57H40O6P4Cl2Re2: C, 49.32; H, 2.90. Found: C, 49.33;
H, 2.95.
[(bpy)2Ru(C3P4)Re(CO)3Cl](PF6)2 (RuC3P4Re). A solution con-
taining both Re(CO)5Cl (15.0 mg, 0.042 mmol) and RuC3P4 (42.2 mg,
0.029 mmol) in MeCN (20 mL) was heated at reflux for 48 h. Upon
completion, the solution was allowed to cool to room temperature, and
the solvent reduced down to ∼2 mL with the use of a rotary evaporator.
The 2 mL sample was added dropwise into a stirred solution of diethyl
ether (50 mL). The resulting orange precipitate was collected by
filtration, washed with 3 × 10 mL of diethyl ether, and dried in Vacuo.
Purification was performed by column chromatography using basic
alumina as the support and MeCN as the eluent, and the product was
collected as the second fraction. Yield: 45.2 mg (81.1%). 31P{1H} NMR
(202.5 MHz, MeCN-d3): δ 25.11 (s), -16.37 (s) ppm. FT-IR (KBr):
ν
CO 2020(vs),1911(vs),1880(vs)cm-1.Anal.CalcdforC74H56F12N4O3P6-
ClReRu: C, 49.77; H, 3.16; N, 3.13. Found: C, 49.73; H, 3.34; N,
3.16.
Cl(CO)3Re(C2P2e) (ReC2P2e). To a refluxing solution of Re(CO)5Cl
(128.8 mg, 0.356 mmol) in THF (20 mL) was added dropwise a solution
[(bpy)2Os(C3P4)Re(CO)3Cl](PF6)2 (OsC3P4Re). Re(CO)5Cl (26.0
mg, 0.071 mmol) was refluxed with OsC3P4 (52.0 mg, 0.033 mmol)
in a MeCN (20 mL) and THF (5 mL) mixture for 24 h. The reaction
mixture was then cooled to room temperature before the solvent was
removed by a rotary evaporator. After the resulting brownish-green
material was dissolved in ∼2 mL of MeCN, it was added dropwise
into a 50 mL stirred solution of diethyl ether. The resulting green
precipitate was collected by filtration, rinsed with 3 × 15 mL of diethyl
ether, and dried in Vacuo. Purification was performed by column
chromatography using basic alumina as the support and MeCN as the
eluent, and the product was collected as the second fraction. Yield:
(12) (a) Schmidbaur, H.; Pollok, T. Angew. Chem., Int. Ed. Engl. 1986,
25, 348-349. (b) Schmidbaur, H.; Manhart, S.; Schier, A. Chem. Ber.
1993, 126, 6, 2 389-2391. (c) Colquhoun, I. J.; McFarlane, W. J.
Chem. Soc., Dalton Trans. 1982, 1915-1920.
(13) (a) Buckingham, D. A.; Dwyer, F. P.; Goodwin, H. A.; Sargeson, A.
M. Aust. J. Chem. 1964, 17, 325. (b) Kober, E. M.; Casper, J. V.;
Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1988, 27, 4587.
(14) (a) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583-
5590. (b) Caspar, J. V.; Sullivan, B. P.; Meyer, T. J. Inorg. Chem.
1984, 23, 2104-2109.

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6041
54.0 mg (87.1%). 31P{1H} NMR (202.5 MHz, MeCN-d3): δ 96.57 (s),
-24.11 (s) ppm. FT-IR (KBr): νCO 2032 (vs), 1941 (vs), 1908 (vs)
cm-1. Anal. Calcd for C74H56F12N4O3P6ClOsRe: C, 47.40; H, 3.01; N,
2.98. Found: C, 47.34; H, 3.27; N, 3.01.
0.20 × 0.37 mm was mounted on a glass fiber and transferred to a
Bruker CCD platform diffractometer. The SMART15 program package
was used to determine the unit cell parameters and for data collection
(30 s/frame scan time for a hemisphere of diffraction data). The raw
frame data was processed using SAINT16 and SADABS17 to yield the
reflection data file. Subsequent calculations were carried out using the
SHELXTL18 program. The diffraction symmetry was 2/m and the
systematic absences were consistent with the monoclinic space groups
Cc or C2/c. It was later determined that the centrosymmetric space
group C2/c was correct.
The structure was solved by direct methods and refined on F2 by
full-matrix least-squares techniques. The analytical scattering factors19
for neutral atoms were used throughout the analysis. Hydrogen atoms
were located from a difference Fourier map and refined (x,y,z) and
U(iso) or included using a riding model. The molecule was a dimer,
which was located on a 2-fold rotation axis. Atom C(26) was located
directly on the 2-fold and was assigned a site-occupancy factor of 0.5.
There were two molecules of benzene solvent present per dimeric
formula unit. The solvent molecule defined by atoms C(30)-C(32)
was located about a 2-fold axis, while that defined by atoms C(33)-
C(35) was located about an inversion center. At convergence, wR2 )
0.0441 and GOF ) 1.107 for 475 variables refined against 7460 unique
data (as a comparison for refinement on F, R1 ) 0.0195 for those
6911 data with I > 2.0σ(I)).
X-ray Structure Determination for ReC4P4Re. A red crystal of
Cl(CO)3Re(C4P4)Re(CO)3Cl‚4(C4H8O) with the dimensions 0.03 × 0.16
× 0.17 mm was mounted on a glass fiber and transferred to a Bruker
CCD platform diffractometer. The SMART15 program package was
used to determine the unit-cell parameters and for data collection (30
s/frame scan time for a hemisphere of diffraction data). The raw frame
data was processed using SAINT16 and SADABS17 to yield the
reflection data file. Subsequent calculations were carried out using the
SHELXTL18 program. The diffraction symmetry was 2/m, and the
systematic absences were consistent with the monoclinic space groups
Cc or C2/c. It was later determined that the centrosymmetric space
group C2/c was correct.
The structure was solved by direct methods and refined on F2 by
full-matrix least-squares techniques. The analytical scattering factors19
for neutral atoms were used throughout the analysis. Hydrogen atoms
were included using a riding model. The molecule was a dimer, which
was located about an inversion center. There were four molecules of
THF solvent present per dimeric formula unit. The solvent molecules
were disordered and were included with two components for each atom
which were assigned site-occupancy factors of 0.5. The SAME18
command was used to fix the geometry of the rings relative to that
defined by O(5)-C(37). At convergence, wR2 ) 0.0918 and GOF )
1.073 for 405 variables refined against 8421 unique data (as a
comparison for refinement on F, R1 ) 0.0454 for those 6507 data
with I > 2.0σ(I)).
Cl(CO)3Re(C4P4) (ReC4P4). To a refluxing solution of C4P4 (261.7
mg, 0.332 mmol) in THF (40 mL) was added Re(CO)5Cl (57.0 mg,
0.157 mmol) in THF (20 mL) dropwise. After refluxing for 24 h, the
reaction mixture was cooled to room temperature. The volume of the
solution was reduced down to ∼2 mL before being added dropwise
into a stirred solution of hexanes (75 mL). The resulting orange
precipitate was collected by filtration, washed with 3 × 15 mL of
hexanes, and dried in Vacuo. Purification was performed by column
chromatography using basic alumina as the support and CHCl3 as the
eluent, and the product was collected as the first fraction. Yield: 82.5
mg (64.2%). 31P{1H} NMR (202.5 MHz, acetone-d6): δ 21.40 (s),
-8.92 (s) ppm. FT-IR (KBr): νCO 2026 (vs), 1948 (vs), 1907 (vs) cm-1
.
Anal. Calcd for C55H40O3P4ClRe: C, 60.35; H, 3.68. Found: C, 60.57;
H, 3.64.
Cl(CO)3Re(C4P4)Re(CO)3Cl (ReC4P4Re). A solution containing
C4P4 (100.5 mg, 0.127 mmol) and Re(CO)5Cl (100.0 mg, 0.277 mmol)
in THF (40 mL) was refluxed for 24 h, and the reaction mixture was
cooled to room temperature. The volume of the solution was reduced
down to ∼2 mL before being added dropwise into a stirred solution of
hexanes (125 mL). The resulting red precipitate was collected by
filtration, washed with 3 × 15 mL of hexanes, and dried in Vacuo.
Purification was performed by column chromatography using basic
alumina as the support and CHCl3 as the eluent, and the product was
collected as the second fraction. Crystals were obtained by first
dissolving the red powder in a minimum amount of THF and then
adding an equivalent amount of benzene. The solution mixture was
stored at room temperature in a vented container. As the THF
evaporated, red plates formed. Yield: 139.4 mg (78.4%). 31P{1H} NMR
(202.5 MHz, acetone-d6): δ -2.53 (s) ppm. 13C NMR (125.7 MHz,
THF-d8): δ 135.18 (m, CdCdC), 133.76 (m, phenyl Ci), 132.25 (d,
JCP ) 7.6 Hz, dCP2), 129.93 (s, phenyl, Cp), 129.3 (m, phenyl Cm and
Co). FT-IR (KBr mull): νCO 2030 (vs), 1960 (vs), 1900 (vs) cm-1
.
Anal. Calcd for C58H40O6P4Cl2Re2: C, 49.75; H, 2.87. Found: C, 49.83;
H, 2.88.
[(bpy)2Ru(C4P4)Re(CO)3Cl](PF6)2 (RuC4P4Re). A solution con-
taining both Re(CO)5Cl (14.0 mg, 0.039 mmol) and RuC4P4 (31.0 mg,
0.021 mmol) in a MeCN (20 mL) and THF (20 mL) mixture was heated
at reflux for 24 h. Upon completion, the solution was allowed to cool
to room temperature, and the solvent removed with the use of a rotary
evaporator. The resulting reddish-brown material was dissolved in ∼2
mL of MeCN. The 2 mL sample was dropped into a stirred solution of
diethyl ether (50 mL). The resulting reddish-orange precipitate was
collected by filtration, washed with 3 × 10 mL diethyl ether, and dried
in Vacuo. Purification was performed by column chromatography using
basic alumina as the support and MeCN as the eluent, and the product
was collected as the second fraction. Yield: 29.2 mg (77.4%). 31P-
{1H} NMR (202.5 MHz, MeCN-d3): δ 20.18 (s), 13.56 (s) ppm. FT-
IR(KBr): νCO 2015(vs),1880(vs)cm-1.Anal.CalcdforC75H56F12N4O3P6-
ClReRu: C, 50.10; H, 3.13; N, 3.11. Found: C, 50.14; H, 3.23; N,
3.12.
[(bpy)2Os(C4P4)Re(CO)3Cl](PF6)2 (OsC4P4Re). A solution of both
Re(CO)5Cl (18.5 mg, 0.051 mmol) and OsC4P4 (40.5 mg, 0.025 mmol)
in a MeCN (20 mL) and THF (20 mL) mixture was heated at reflux
for 24 h. Upon completion, the solution was allowed to cool to room
temperature, and the solvent removed with the use of a rotary
evaporator. The resulting brown material was dissolved in ∼2 mL of
MeCN and then added dropwise into a stirred solution of diethyl ether
(50 mL). The resulting brown precipitate was collected by filtration,
washed with 3 × 10 mL of diethyl ether, and dried in vacuo. Purification
was performed by column chromatography using basic alumina as the
support and MeCN as the eluent, and the product was collected as the
second fraction. Yield: 35.5 mg (73.5%). 31P{1H} NMR (202.5 MHz,
MeCN-d3): δ 81.14 (s), -7.54 (s) ppm. FT-IR (KBr): νCO 2021 (vs),
1890 (vs) cm-1. Anal. Calcd for C75H56F12N4O3P6ClOsRe: C, 47.73;
H, 2.99; N, 2.96. Found: C, 47.62; H, 3.01; N, 3.01.
Results and Discussion
Synthetic Aspects. For the mono- and homobimetallic
reactions with ReI, the metal-to-ligand ratios can be altered in
order to favor the production of the desired adduct. The addition
of the metal starting material to a heated solution of the ligand,
with the metal-to-ligand ratio 1:(1.7-2.2), favors the production
of the monometallic complex. By having an excess amount of
spacer, the formation of the monometallic complex was
dominating. Conversely, if the ligand was added to a heated
(15) SMART Software Users Guide, Version 4.21; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, 1997.
(16) SAINT Software Users Guide, Version 4.05; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, 1997.
(17) Sheldrick, G. M. SADABS; Bruker Analytical X-ray Systems, Inc.:
Madison, WI, 1997.
(18) Sheldrick, G. M. SHELXTL Version 5.10; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, 1997.
X-ray Structure Determination for ReC3P4Re. A yellow crystal
of Cl(CO)3Re(C3P4)Re(CO)3Cl‚2(C6H6) with the dimensions 0.17 ×
(19) International Tables for X-ray Crystallography, Vol. C; Kluwer
Academic Publishers: Dordrecht, 1992.

6042 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
Table 1. Fast Atom Bombardment Mass Spectral (FAB/MS) Data
compound MW (m/z), [rel int, %]
fragment assignment
ReC2P2e
702 [15]
674 [30]
667 [100]
646 [30]
638 [15]
M+
M+ - CO
M+ - Cl
M+ - 2CO
M+ - CO - Cl
M+ - 3CO
618 [60]
582 [10]
M+ - 3CO - Cl
M+
ReC3P4
1082 [55]
1054 [20]
1046 [15]
1026 [15]
1018 [15]
998 [5]
915 [40]
898 [10]
879 [90]
863 [20]
M+ - CO
M+ - Cl
M+ - 2CO
M+ - CO - Cl
M+ - 3CO
M+ - PPh2 + O
M+ - PPh2
M+ - PPh2 - Cl + O
M+ - PPh2 - Cl
M+ - PPh2 - 2CO + O
M+ - PPh2 - 2CO
M+ - PPh2 - 2CO + O
M+ - PPh2 - CO - Cl
M+ - PPh2 - 2CO - Cl
M+
857 [100]
841 [100]
832 [100]
832 [35]
Figure 1. Representative synthesis of a heterobimetallic complex.
solution of the metal precursor with the metal-to-ligand ratio
(2-2.2):1, the homobimetallic species was the major product
isolated.
806 [15]
ReC4P4
1095 [60]
1067 [35]
1059 [100]
1038 [80]
1031 [30]
1010 [55]
1003 [30]
910 [10]
M+ - CO
To synthesize the heterometallic complexes RuC3P4Re and
RuC4P4Re (Figure 1), the ligand was first attached to the RuII
center, which subsequently reacted with Re(CO)5Cl in refluxing
acetonitrile (82° C) for 24 h with a Ru:Re ratio of 1:1.5. The
corresponding heterobimetallic complexes OsC3P4Re and
OsC4P4Re were obtained under the same conditions but with a
slightly more excess of Re(CO)5Cl (Os:Re ratio of 2:1). More
of the ReI starting material was needed in the reaction of the
OsII complexes because the OsII adducts were less reactive than
the RuII ones. The presence of an excess of ReI favored the
reaction in the forward direction to give OsCnP4Re (n ) 3, 4).
Upon completion of reactions and the removal of the THF
portion of the solvent mixture, the remaining concentrated
product/solvent mixture was added dropwise into a saturated
aqueous solution of KPF6 to obtain the solid products. There
are two reasons to use the KPF6/H2O solution. First, the metal
complexes with either +2 to +4 charges were slightly soluble
in water. Saturating the water with KPF6 salt increased the
polarity of the solution to the point that the organic portion of
the multicomponent complexes controlled the solubility of the
overall complex and precipitated out of aqueous solution. The
second reason was to ensure that the product was isolated as
the PF6 salt rather than a chloride salt. While the complexes
with the PF6- counteranions are luminescent, the corresponding
adducts with Cl- counteranions are not.
M+ - Cl
M+ - 2CO
M+ - CO - Cl
M+ - 3CO
M+ - 2CO - Cl
M+ - PPh2
845 [45]
825 [95]
M+ - PPh2 - CO - Cl
M+ - PPh2 - 3CO
M+
ReC3P4Re
ReC4P4Re
1388 [30]
1360 [20]
1053 [100]
1332 [25]
1297 [10]
1400 [85]
1072 [10]
1365 [15]
1316 [10]
1281 [15]
1217 [45]
1181 [45]
1125 [95]
1095 [45]
1668 [45]
1476 [100]
1331 [60]
1186 [50]
1472 [30]
1440 [100]
1287 [60]
1267 [45]
1259 [60]
1231 [30]
1090 [45]
M+ - CO
M+ - Cl
M+ - 2CO
M+ - 2CO - Cl
M+
M+ - CO
M+ - Cl
M+ - 2CO - Cl
M+ - 3CO - Cl
M+ - 4CO - 2Cl
M+ - PPh2 - Cl
M+ - PPh2 - 2CO - Cl
M+ - PPh2 - 3CO - Cl
M+ - 3CO - Cl
M+ - Re(CO)3Cl
M+ - Re(CO)3Cl - PF6
M+ - Re(CO)3Cl - 2PF6
M+ - 2PF6
RuC3P4Re
RuC4P4Re
M+ - 2PF6 - Cl - CO
M+ - 2PF6 - PPh2 - Cl
M+ - 2PF6 - PPh2 - 2CO
M+ - 2PF6 - PPh2 - Cl - CO
M+ - 2PF6 - PPh2 - Cl - 2CO
M+ - 2PF6 - 2PPh2 - Cl -
CO + O
Chromatographic methods were required in order to purify
the desired products. In the isolation of the monometallic
complexes, the major side products to be separated were the
homobimetallic complexes and the monosubstituted complex,
where a chloride was still coordinated to the RuII or OsII center.
The chromatographic separation in this system was possible
partly due to the different cationic charges on these organome-
tallic complexes. For the system containing only the ReI centers,
chloroform was used in the chromatographic separation. How-
ever, for the heterobimetallic ReI/RuII and ReI/OsII complexes,
acetonitrile was a better solvent for the chromatographic
separation. This was due to the solubility characteristics imparted
by the OsII and RuII moieties connected via the cumulenic spacer
to the ReI moiety.
OsC3P4Re
OsC4P4Re
1748 [100]
1606 [60]
1586 [15]
1566 [100]
1568 [10]
1543 [15]
1536 [40]
1509 [25]
1491 [100]
1475 [95]
M+ - PF6 + O
M+ - 2PF6 + O
M+ - 2PF6
M+ - 2PF6 - Cl + O
M+ - 2PF6 - CO
M+ - 2PF6 - 2CO
M+ - 2PF6 - CO - Cl
M+ - 2PF6 - 2CO - Cl
M+ - 2PF6 - 3CO - Cl + O
M+ - 2PF6 - 3CO - Cl
as the neutral homobimetallic ones (ReC3P4Re and ReC4P4Re).
For the cationic heterobimetallic complexes containing either
RuII or OsII, the overall structures were found to be less robust
than the complexes containing solely the ReI center. The loss
FAB/MS Analysis. A summary of the FAB/MS data is in
Table 1. There are several apparent trends of the features in the
data. The molecular ions were detected for the neutral mono-
metallic complexes (ReC2P2e, ReC3P4, and ReC4P4) as well
of the counterion PF6 occurred readily, as did the loss of F-
-
atoms. Even though the core structures were stable, loss of PPh2

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6043
was 10.050 Å, which is 1.33 Å longer than the ReI-ReI distance
in ReC3P4Re. The dihedral angle for ReC4P4Re, defined by
the two planes containing P(1)-C(25)-P(2) and P(1′)-C(25′)-
P(2′), was found to be 0.0°, which is expected for an even-
numbered cumulenic chain. Because of a longer carbon chain
length, the phenyl groups are no longer close enough to each
other as to cause the distortion away from the coplanar
arrangement. The bond angles for C(25)-C(26)-C(26′) and
C(26)-C(26′)-C(25′) of the cumulenic carbon chain were
found to be more linear (178.8°) than for the C3 carbon chain
in ReC3P4Re (176.9°), presumably due to the decrease in steric
interactions among the phenyl groups. The 2-fold axis of the
molecule was found to pass the center of the double bond
between C(26) and C(26′). The CdC bond lengths of the
cumulenic group contain the characteristic “long-short-long”
pattern found for other butatrienes.12b As in ReC3P4Re, each
ReI moiety in ReC4P4Re is octahedral, with the three carbonyl
ligands in a facial arrangement with the chloride perpendicular
to the P(1)-Re(1)-P(2) plane.
Table 2. Crystal Data and Structure Refinement for ReC3P4Re and
ReC4P4Re
ReC3P4Re
ReC4P4Re
empirical formula
C57H40Cl2O6P4Re2‚2- C58H40Cl2O6P4Re2‚4-
(C6H6) (C4H8O)
1688.50
fw
1544.29
monoclinic
C2/c
cryst syst
space group
wavelength (Å)
temp (K)
a (Å)
monoclinic
C2/c
0.71073
158
0.71073
158
18.0275(9)
18.9805(9)
19.3010(9)
90
25.3118(12)
15.1708(7)
19.7377(9)
90
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
volume (Å3)
Z
110.5770(10)
90
109.9210(10)
90
6182.9(5)
4
7125.8(6)
4
F(000)
3032
1.659
4.154
0.37 × 0.20 × 0.17
3360
1.574
3.616
F
calc (mg m-3
)
µ (mm-1
)
cryst size (mm3)
0.17 × 0.16 × 0.03
no. of reflns collected 19960
22278
For both crystal structures the bite angles, defined by the
atoms P(1)-Re(1)-P(2), were found to be 70.33° and 70.43°
for the C3 and C4 adducts, respectively. These values are
consistent with but slightly smaller than the average bite angle
(71.53°) found for diphenylphosphinomethane (dppm) coordi-
nated to different transition metal centers.20 Although the angles
about a sp2-hybridized carbon atom should theoretically be 120°,
the angle of P(1)-C(25)-P(2) for ReC3P4Re and ReC4P4Re
was found to be 100.73° and 101.92°, respectively. These
observed decreases in the angles are presumably caused by the
four-membered ring structure at the ReI center.
no. of unique reflns
7460
1.107
R1 ) 0.0195
wR2 ) 0.0428
R1 ) 0.0227
wR2 ) 0.0441
8421
1.073
R1 ) 0.0454
wR2 ) 0.0828
R1 ) 0.0711
wR2 ) 0.0918
S(F2)a
final R indicesb
[I > 2σ(I)]
R indices (all data)b
a S ) goodness of fit ) [∑[w(Fo - Fo )2]/(n - p)]1/2, where n is
the number of reflections and p is the total number of parameters
2
2
refined. b R1 ) ∑||Fo| - |Fc||/∑|Fo| and wR2 ) [∑[w(Fo - Fo )2]/
2
2
∑[w(Fo )2]1/2, w ) 1/σ2(|Fo|).
2
groups was observed, coinciding with the results observed for
the complexes with only the RuII or OsII centers.3 The loss of
CO and Cl units from the ReI core was also detected. The
addition of one, two, or three oxygen atoms was found in
fragments, even though all of the complexes were shown to be
air stable from both 31P{1H} NMR and crystallographic data.
Such additions of oxygen atoms were observed to be charac-
teristic for systems with polyphosphine spacers.3,6
IR Analysis. The IR carbonyl stretches for compounds
containing the Re(CO)3Cl units are listed in Table 4. Except
for RuC4P4Re and OsC4P4Re, every compound possessed three
very strong signals in the carbonyl region between 2032 and
1880 cm-1 (Figure 4a). These data represent an apparent shift
to lower energies with respect to the starting material Re(CO)5Cl
that appear between 2155 and 1854 cm-1 21
The similarity in
.
CrystalStructureDetermination.ThestructuresofReC3P4Re
and ReC4P4Re were obtained by single-crystal X-ray structure
determination. A summary of the crystallographic data appears
in Table 2. Selected bond lengths and bond angles are listed in
Table 3. The ORTEP plot of ReC3P4Re is shown in Figure 2
(excluding the two benzene molecules, which crystallize as part
of the unit cell). The refinements converged to R1 ) 1.95%
and wR2 ) 4.24% with I > 2σ(I) and R1 ) 2.27%, wR2 )
4.41%, and S ) 1.107 for 7460 unique reflections, 475 variables,
and 60 restraints. The molecular unit consists of two Re(CO)3Cl
units bridged by the C3P4 spacer. The metal-to-metal distance
was 8.721 Å, with a dihedral angle (θ) of 91.9° defined by the
two planes resulting from P(1)-Re(1)-P(2) and P(1′)-Re(1′)-
P(2′). For an odd-numbered cumulenic chain, θ was expected
to be 90.0°. The observed distortion is presumably due to steric
interactions among the bulky phenyl groups. The C(25)-C(26)-
C(25′) bond angle of the cumulenic carbon chain was found to
be roughly linear (176.9°). The central C(26) was found to lie
directly on the 2-fold axis of the molecule; therefore both Cd
C bond lengths were equivalent. Each ReI moiety was octahedral
with three carbonyl ligands in a facial arrangement.
the IR spectra of these compounds suggests that they all have
the same core structure around the ReI center. From the
aforementioned crystal structures for ReC3P4Re and ReC4P4Re,
it was observed that each ReI center independently possessed
Cs symmetry with the carbonyl ligands in a facial arrangement,
one carbonyl trans to the chlorine atom and the other two
carbonyls trans to the chelating diphenylphosphino groups
(Figure 5a). Three IR-active carbonyl modes are expected for
the ReI centers which contain the local Cs symmetry (2A′ +
A′′).22,23 The highest energy peak between 2032 and 2015 cm-1
has been assigned as the ν(CO) for the CO that is trans to the
chlorine atom. Since the chlorine ligand is less electron donating
than the diphenylphosphino groups, less back-bonding will occur
from the ReI metal to the transCl CO antibonding orbital. This
effect will give rise to a higher bond order in the CO unit,
shifting the ν(CO) to a higher energy relative to the two ν(CO)
bands (between 1960 and 1880 cm-1) from the CO groups that
are trans to the diphenylphosphino units.24,25
Compounds RuC4P4Re and OsC4P4Re exhibit only two CO
stretches; the higher energy peak is also assigned (as above) to
the transCl CO, while the lower energy peak appears to be a
The ORTEP plot of ReC4P4Re is shown in Figure 3
(excluding the four THF molecules, which crystallize as part
of the unit cell). The refinements converged to R1 ) 4.54%
and wR2 ) 8.28% with I > 2σ(I) and R1 ) 7.11%, wR2 )
9.18%, and S ) 1.073 for 8421 unique reflections, 405 variables,
and 60 restraints. The molecular unit consists of two Re(CO)3Cl
units bridged by the C4P4 spacer. The metal-to-metal distance
(20) Dierkes, P.; van Leeuwen, P. J. Chem. Soc., Dalton Trans. 1999,
1519-1529.
(21) Kaesz, H. D.; Bau, R.; Hendrickson, D.; Smith, J. M. J. Am. Chem.
Soc. 1967, 89, 2844-2851.
(22) Benkstein, K. D.; Hupp, J. T.; Stern, C. L. Inorg. Chem. 1998, 37,
5404-5405.
(23) Abel, E. W.; Tyfield, S. P. Can. J. Chem. 1969, 47, 4627-4633.

6044 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for Complexes ReC3P4Re and ReC4P4Re
bond
ReC3P4Re
ReC4P4Re
angle
ReC3P4Re
ReC4P4Re
Re(1)-C(27)
Re(1)-C(28)
Re(1)-C(29)
Re(1)-P(1)
Re(1)-P(2)
Re(1)-Cl(1)
O(1)-C(27)
O(2)-C(28)
O(3)-C(29)
P(1)-C(25)
P(2)-C(25)
C(25)-C(26)
C(26)-C(26)′
1.918(2)
1.966(2)
1.950(2)
2.4789(5)
2.4654(5)
2.4869(6)
1.139(3)
1.137(3)
1.142(3)
1.842(2)
1.855(2)
1.303(2)
2.009(6)
1.954(6)
1.950(6)
2.4616(13)
2.4641(13)
2.4673(14)
1.017(6)
1.144(6)
1.145(7)
1.832(5)
1.826(5)
1.324(6)
1.260(9)
C(27)-Re(1)-Cl(1)
C(28)-Re(1)-P(2)
C(29)-Re(1)-P(1)
P(1)-Re(1)-P(2)
P(1)-C(25)-P(2)
C(25)-C(26)-C(25′)
C(25)-C(26)-C(26′)
179.30(7)
167.62(7)
168.77(7)
70.33(17)
100.73(9)
176.9(3)
179.07(15)
167.50(17)
167.52(17)
70.43(4)
101.9(2)
178.8(7)
Figure 2. ORTEP plot of ReC3P4Re with thermal ellipsoids plotted at 50% probability. For clarity hydrogen atoms have been omitted.
broaden combination of the two transP CO (Figure 4b). This
effect can be explained by the increased electronic communica-
tion associated with the C4P4 spacer.3 An increase in electron
delocalization from the diphenylphosphino chelates in the
polyphosphine spacer will cause a sufficient perturbation on the
ν(CO) vibrational modes, rendering them indistinguishable.
Previously for systems that contained three CO in a facial
arrangement with chelating N-donating ligands, only one band
was observed for the two CO trans to the chelating N-
ligands.26,27 Another possible interpretation of the observation
of only two ν(CO) bands for the complexes RuC4P4Re and
OsC4P4Re could be that the carbonyl groups are not in a facial
arrangement but rather a meridial arrangement.23 This arrange-
ment would place two of the carbonyl groups trans with respect
to each other, rendering their ν(CO) stretches equivalent, while
the third CO would be trans to one of the diphenylphosphino
groups in the spacer (Figure 5b). Since the phosphine would
be a better electron donor than the π-acidic carbonyl ligands, it
would be expected, however, that the transP ν(CO) would be
shifted to a lower energy than the broad ν(CO) band that
corresponds to the CO groups trans to each other. This is a
contradiction to the observed IR spectra that exhibit the broad
signal to be of lower energy. Hence, the actual observation is
consistent with the facial arrangement argument. In addition,
the facial arrangement would also be the more thermodynami-
cally stable one of the two possible configurations.23
As mentioned earlier, there was very little change of the three
ν(CO) bands in most of the ReI complexes. The three ReI
monometallic complexes exhibited very similar ν(CO) frequen-
cies, with 2026, 1944, and 1911 cm-1 for ReC3P4, 2026, 1948,
and 1907 cm-1 for ReC4P4, as well as 2029, 1952, and 1895
cm-1 for ReC2P2e. Likewise, the ReI homobimetallic complexes
had ν(CO) similar to their corresponding monometallic ad-
ducts: 2026, 1946, and 1914 cm-1 for ReC3P4Re, and 2030,
1960, and 1900 cm-1 for ReC4P4Re. In the OsII-containing
heterobimetallic complexes, there was a minor shift to lower
energy for the transP ν(CO). For the RuII-containing heterobi-
metallic species RuCnP4Re (n ) 3, 4), a shift to even lower
energy was observed in the transP ν(CO). For RuC3P4Re, the
shift was about 35 cm-1. For RuC4P4Re, the combined transP
ν(CO) occurred at 1880 cm-1, about 80 cm-1 shift toward lower
energy when compared with transP ν(CO) in ReC4P4Re. The
shift from the lower energy peak of the transP ν(CO) in
(24) Huhmann-Vincent, J.; Scott, B. L.; Kubas, G. J. Inorg. Chem. 1999,
38, 115-124.
(25) Edwards, D. A.; Marshalsea, J. J. Organomet. Chem. 1977, 131, 73-
90.
(26) Pfennig, B. W.; Cohen, J. L.; Sosnowski, I.; Novotny, N. M.; Ho, D.
M. Inorg. Chem. 1999, 38, 606-612.
(27) Schoonover, J. R.; Strouse, G. F.; Dyer, R. B.; Bates, W. D.; Chen,
P. Y.; Meyer, T. J. Inorg. Chem. 1996, 35, 273-274.

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6045
Figure 3. ORTEP plot of ReC4P4Re with thermal ellipsoids plotted at 50% probability level. For clarity hydrogen atoms have been omitted.
Table 4. IRa Spectroscopic Data of the Carbonyl Signalsb for ReI
Complexes
compound
vCO transCl (cm-1
)
vCO transP (cm-1
)
ReC3P4
ReC4P4
ReC2P2e
ReC3P4Re
ReC4P4Re
RuC3P4Re
RuC4P4Re
OsC3P4Re
OsC4P4Re
2026
2026
2029
2026
2030
2020
2015
2032
2021
1944
1948
1952
1946
1960
1911
1911
1907
1895
1914
1900
1880
1880
1890
1941
1908
b
a Measured as KBr pellets. Every signal measured was very strong
(vs).
ReCnP4Re (n ) 3, 4) to the combined transP ν(CO) in
RuC4P4Re was 20 cm-1. These shifts are due to an increase of
electron donation from the RuII center through the spacer to
the ReI center and subsequently into the antibonding orbital of
the CO ligand. A greater shift (of the carbonyl stretches) to
lower energies occurred for RuC4P4Re and OsC4P4Re, along
with the observation of the overlapping ν(CO) signals. These
are indicative of a greater electronic communication through
the C4P4 with respect to complexes of C3P4.
31P{1H} NMR Analysis. Analysis of the new complexes by
31P{1H} NMR has proven to provide substantial information
as to the identification and the purity of samples.6,28-30 The 31P-
{1H} NMR data for all new ReI complexes are listed in Table
5. Singlets were observed for both the noncoordinated diphe-
nylphosphino groups and the coordinated M-PPh2 units, except
in ReC3P4Re. Regarding the two observed doublets in the 31P
NMR spectrum of ReC3P4Re, the aforementioned X-ray
structural determination revealed a structural distortion possibly
due to the steric interaction between the bulky phenyl groups.
Such steric distortion may cause the two cis-PPh2 units on the
same ReI centers to be unequivalent and results in two 31P NMR
signals with a small cis-coupling JP-P constant of ca. 36 Hz.
Furthermore, the maximum symmetry the molecule may have
in solution is C2, which may also result in the pairwise
Figure 4. (a) Representative carbonyl region of the IR spectrum for
ReC2P2e depicting three very strong signals. (b) Region of the IR
spectrum for OsC4P4Re depicting only two very strong signals.
inequivalence of the four P atoms. As a comparison, the steric
interaction within the corresponding homobimetallic species
ReC4P4Re is decreased due to an increased carbon chain length
in the bridging unit, and the single-crystal structure determi-

6046 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
Redox Characteristics and Electronic Communication.
Cyclic voltammetry (CV) has been a very important analytical
tool for studying the electrochemical behavior of a variety of
chemical systems in the past two to three decades.32-34 It can
be used to study the ground-state electronic communication of
multicomponent supramolecular systems.1h,4b,g,7e,35 By following
the redox behavior of a given system, it then becomes possible
to devise a way to control that behavior. For a system to be
considered useful in the regime of molecular devices, that system
must have electronic properties that are not only observable but
also tunable.1h,2,7c,35-37 By varying the redox potentials and
consequently modulating the energetics of electron transport,
the movement and storage of electrons within supramolecular
systems can be possible. Several groups have been concerned
with the redox behavior of one-dimensional rigid rodlike
supramolecular systems. Our group have reported the modula-
tion of redox potentials and the level of electronic communica-
tions between redox-active termini using RuII and OsII com-
plexes with polyphosphine/cumulene spacers.3,6 Gladysz et
al.1f,h,11b have reported the modification of redox communication
between two metal units directly linked by acetylenic (polyyne)
units with variable chain lengths. Lapinte and co-workers7a,38
have also described the effect of different spacer lengths on the
electrochemical behavior of the same metal-based unit in
different oxidation states. It is anticipated that the rigidity and
orientation of the spacers with organic linkage components can
alter and manipulate the redox behavior of the metal termini.3,39
The results of electrochemical studies on the new compounds
containing the mixed ReI/RuII and ReI/OsII units are listed in
Table 6. No reversible redox potential was observed for the
ReI metal center, but those attributed to the OsII/III and RuII/III
were observed to be pseudoreversible. Upon comparison of the
heterobimetallic MCnP4Re (n ) 3, 4, M ) Ru, Os) with the
corresponding OsII and RuII monometallic species MCnP4 (n
) 3, 4, M ) Ru, Os) previously reported,3 similar characteristics
were observed. For RuC3P4Re and OsC3P4Re, there were three
redox waves corresponding to the metal oxidation at a positive
potential, as well as the reduction of the C3P4 spacer and the
bipyridyl ligand at negative potentials. Alternatively, cyclic
voltammograms for RuC4P4Re and OsC4P4Re displayed one
metal-based oxidation and two bipyridine reductions, as ob-
served before for the corresponding monometallic complexes
with the same spacer. Interestingly, the longer C4-spacer was
not reduced within the CV scan range of this study, but the
shorter C3P4 was. A similar observation was also found in RuII
and OsII mono- and homobimetallic species with the same
spacers.3,6
Figure 5. The two possible arrangements for the ReI core.
nation confirmed an ideal coplanar arrangement, with a dihedral
angle of 0.0°, of the two Re(CO)3Cl(P)2 units. As a result, only
one singlet was observed in the 31P NMR spectrum.
For the assignment of signals, the chemical shifts for the ReI-
PPh2 units appeared between -1.43 and -12.75 ppm in
ReC2P2e, ReC3P4Re, and ReC4P4Re, and at -8.92 and -13.30
ppm in ReC3P4 and ReC4P4, respectively. The two other signals
at 22.80 and 21.40 ppm are assigned to the free noncoordinated
-PPh2 units. For the monometallic complexes ReC3P4, ReC4P4,
and ReC2P2e the signals for the noncoordinated phosphines in
the 31P{1H} NMR spectra were shifted downfield, while the
coordinated phosphine signals were shifted upfield with respect
to the free ligands. This behavior is indicative of an inductive
“pulling” of the electrons away from the noncoordinated -PPh2
units. Presumably, this effect is caused by the CO units on the
ReI center withdrawing electron density across the spacer.
Therefore, the -PPh2 units coordinated to the ReI center
experienced an increase in shielding from the electrons induc-
tively withdrawn from the noncoordinated -PPh2 units.
Similarly, the 31P NMR signals for the heterobimetallic
complexes can also be assigned. The signals at negative ppm
chemical shifts represented the ReI-PPh2 units, while the signals
at positive chemical shifts signify the RuII-PPh2 and OsII-
PPh2 units. While the OsII resulted in signals significantly shifted
to above +80 ppm, the RuII-PPh2 appeared between +20 and
+25 ppm. Even though RuC4P4Re had two signals at positive
chemical shifts, the signal at 20.18 ppm was assigned to RuII-
PPh2 due to the proximity with the signal of RuC4P4Re at 25.11
ppm. As observed for the monometallic complexes, the carbonyl
groups of the ReI unit inductively shield the ReI-PPh2 units
while deshielding the RuII-PPh2 and OsII-PPh2 units.
It was shown previously25,31 that in similar rhenium tricar-
bonyl systems the coordinated phosphorus signal of the chelating
bisphosphine ligand (e.g., dppm) was shifted upfield with respect
to the free ligand. This is in agreement with the above
assignments for the ReI-PPh2 signals in this system. The
electrons of the noncoordinating phosphines were inductively
“pulled” through the cumulenic backbone toward the coordi-
nated phosphines due to the back-bonding of the rhenium metal
to the two trans carbonyls. Likewise, the phosphines coordinated
to the ReI center in the heterobimetallic complexes were also
shifted upfield.
The longer C4P4 spacer also caused the oxidation potentials
of both OsII and RuII to shift to lower values with respect to the
corresponding complexes with C3P4. The RuII/III redox couple
of RuC3P4Re (+1.46 V) was shifted to a lower potential by
450 mV in RuC4P4Re (+1.01 V). Similarly, the OsII/III redox
couple was shifted 520 mV to a less positive potential when
(32) Kissinger, P. T.; Heineman, W. R. J. Chem. Educ. 1983, 60, 702-
706.
(33) Heinze, J. Angew. Chem., Int. Ed. Engl. 1984, 23, 831-918.
(34) Maloy, J. T. J. Chem. Educ. 1983, 60, 285-289.
(35) Patoux, C.; Launay, J. P.; Beley, M.; Chodorowski-Kimmes, S.; Collin,
J. P.; James, S.; Sauvage, J. P. J. Am. Chem. Soc. 1998, 120, 3717-
3725.
(36) Lehn, J. M. Angew. Chem., Int. Ed. Engl. 1990, 29, 1304-1319.
(37) Harriman, A.; Ziessel, R. Coord. Chem. ReV. 1998, 171, 331-339.
(38) Le Narvor, N.; Lapinte, C. C. R. Acad. Sci. Ser. II C 1998, 1, 745-
749.
(39) Balzani, V.; Juris, A.; Venturi, M.; Campagna, S.; Serroni, S. Chem.
ReV. 1996, 96, 759-833.
(28) Pregosin, P. S.; Kunz, R. W. 31P and 13C NMR of Transition Metal
Phosphine Complexes; Springer-Verlag: Berlin, 1979.
(29) Beringhelli, T.; Dalfonso, G.; Freni, M.; Minoja, A. P. Inorg. Chem.
1996, 35, 2393-2395.
(30) Benn, R. Transition Metal Nuclear Magnetic Resonance; Pregosin,
P. S., Ed.; Amsterdam, Netherlands, 1991; Chapter 4.
(31) Carriedo, G. A.; Rodriguez, M. L.; Garciagranda, S.; Aguirre, A. Inorg.
Chim. Acta 1990, 178, 101-106.

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6047
Ru-P or Os-P
Table 5. 31P{1H} NMR Spectroscopic Data of ReI Complexesa
noncoord
complex
31P (ppm)
Re-P 31P (ppm)
complex
Re-P 31P (ppm)
31P (ppm)
ReP4C3
ReP4C4
ReP2C2e
ReP4C3Re
22.80 (s)
21.40 (s)
-13.30 (s)
RuP4C3Re
RuP4C4Re
OsP4C3Re
OsP4C4Re
-16.37 (s)
13.56 (s)
-24.11 (s)
-7.54 (s)
25.11 (s)
20.18 (s)
96.57 (s)
81.14 (s)
-8.92 (s)
-8.41 (s)
-12.75 (d, JP-P ) 35.7 Hz)
-1.43 (d, JP-P ) 35.7 Hz)
-2.53 (s)
ReP4C4Re
a Only the chemical shifts associated with the core structure are presented here. A characteristic septet was observed for each complex corresponding
-
to the PF6 counteranions centered between δ -144 and -145 ppm.
Table 6. Formal Potentials (vs SCE) for RuII and OsII Complexes Containing ReI
RuII/III E1/2, V
(∆Ep, mV)
OsII/III E1/2, V
(∆Ep, mV)
P4C3
E
, V
1st bpy
E
, V
2nd bpy
E , V
1/2
0/-
0/-
0/-
1/2
1/2
complex
(∆Ep, mV)
(∆Ep, mV)
(∆Ep, mV)
RuP4C3Re
+1.46
-1.18
-1.42
(1e-, 100)
+1.01
(1e-, 70)
(1e-, 82)
-1.40
RuP4C4Re
OsP4C3Re
OsP4C4Re
-1.63
(1e-, 146)
(1e-, 74)
-1.35
(1e-, 65)
+1.13
-1.12
(1e-, 104)
+0.61
(1e-, 88)
(1e-, 74)
-1.20
-1.42
(1e-, 80)
(1e-, 84)
(1e-, 94)
Table 7. Absorption Maxima from the Electronic Spectra of the
comparing OsC3P4Re and OsC4P4Re. Presumably the longer
C4P4 had a greater number of delocalized π-electrons and was
a better σ-donor than C3P4. When an increased amount of
σ-donation was contributed to the metal center, there will be
more electron density residing on the metal, thereby causing
the redox couple to shift to a less positive value.3a
ReI Complexes
λ
abs, nm (ꢀ, M-1 cm-1
)
ligand-centered CT
Re, Ru, or Os-based
1MLCT
complex
N-bpy
P-spacer
Os 3MLCT
ReC2P2e
330
420
The metal-based oxidations for the heterobimetallic MCnP4Re
(M ) RuII, OsII) complexes, though slightly different, were
found to be consistent with the oxidation potentials of the
previously reported monometallic complexes MCnP4 (M ) RuII,
OsII).3,6 For RuC3P4Re, the RuII/III oxidation occurred at +1.46
V, which was a 90 mV shift from that of the corresponding
monometallic species RuC3P4 (+1.55 V). For RuC4P4Re, the
RuII/III oxidation occurred at +1.01 V, only a 60 mV change
from the corresponding RuC4P4 monometallic species (+1.07
V). This similarity was also observed for the OsII/III redox couple
in OsC3P4Re and the corresponding OsC3P4, where there was
only a 10 mV difference in the oxidation potentials (+1.13 V
and +1.12 mV, respectively). However, the complex OsC4P4Re
(+0.61 V for OsII/III) experienced a much larger shift in the
oxidation potential (190 mV) with respect to the corresponding
OsC4P4 complex (+0.80 V). Such a low oxidation potential
(+0.61 V) suggested that the OsII center in OsC4P4Re was much
easier to oxidize.
(9535)
320
(11 889)
300
(21 280)
395
(8400)
350
(8120)
300
(14 520)
370
(7640)
360
(6070)
420
(4790)
465
ReC3P4
ReC4P4
(7990)
ReC3P4Re
ReC4P4Re
435
(sh, 530)
475
(2260)
RuC3P4Re 285
450
(2155)
450
(3530)
440
(8100)
460
(30 580) (6780)
RuC4P4Re 290
330
(35 635) (14 400)
OsC3P4Re 290
345
550
(sh, 1250)
595
(64 100) (20 430)
OsC4P4Re 290
390
(100 950) (17 240)
(15 540)
(3480)
Electronic Absorption, Emission, and Lifetimes. The
ground-state absorption maxima and extinction coefficients from
the electronic absorption spectra of all new compounds are listed
in Table 7. All of the complexes exhibit ligand-centered charge
transfer (1LC) features involving the bridging polyphosphine
spacer in the visible region of 300-395 nm. The extinction
coefficients (ꢀ ∼ 103-104 M-1 cm-1) are an order of magnitude
more intense than those in a similar system containing non-
polyene polyphosphine ligands (ꢀ ≈ 102-103 M-1 cm-1).21 This
observation may result from the extensive delocalization of
electrons through the cumulenic framework. The mono- and
homobimetallic complexes ReC4P4 and ReC4P4Re have one
1
polypyridyl group is attached to C4P4, the higher energy LC
band is no longer observable. Instead, they exhibit a second
stronger type of 1LC charge transfer involving the ancillary bpy
ligands with well-known and typical extinction coefficients (ꢀ
≈ 104-105 M-1 cm-1).22,23
The mono- and homobimetallic ReI complexes also exhibit
absorption maxima in the region 420-460 nm (Figure 6a) which
have been previously assigned as a Re-based transition for a
similar system.40-42 Whereas in the earlier study the absorption
bands were extremely weak (ꢀ < 20 M-1 cm-1), leading to an
(40) (a) Ramsis, M. N. Monatsh. Chem. 1993, 124, 849-855. (b) Wrighton,
M. S.; Morse, D. L.; Gray, H. B.; Ottensen, K. K. J. Am. Chem. Soc.
1976, 98, 1111. (c) Giordano, P. J.; Wrighton, M. S. J. Am. Chem.
Soc. 1979, 101, 2888.
(41) Dahlgren, R. M.; Zink, J. I. Inorg. Chem. 1977, 16, 3154.
(42) (a) Lees, A. J. Comments Inorg. Chem. 1995, 17, 319-346. (b) Kern,
J.-M.; Sauvage, J.-P.; Weidmann, J.-L.; Armaroli, N.; Flamigni, L.;
Ceroni, P.; Balzani, V. Inorg. Chem. 1997, 36, 5329-5338.
1
more LC charge transfer event than the corresponding C3P4-
containing complexes in the region attributed to the polyphos-
phine spacer (Table 7). Presumably, the addition of another
π-bond appears to facilitate the delocalization of electrons which
may lower the π-antibonding orbitals, causing another charge
transfer band to appear at lower energy. Once a RuII or OsII

6048 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
Table 8. Photophysical Data for ReI Complexes
τ,a ns ((10%)
Φa,b
a
λem
,
Re
Ru
Os (×10-3
)
compound nm basede based based ((10%) kr,c s-1
knr,d s-1
ReC2P2e
ReC3P4
ReC4P4
ReC3P4Re 480
ReC4P4Re 530
RuC3P4Re 600
RuC4P4Re 605
OsC3P4Re 580
OsC4P4Re 610
480
480
530
2.1
3.3
6.8
4.5
8.9
f
2.28 1.1 × 106 4.8 × 108
4.42 1.3 × 106 3.0 × 108
1.83 2.7 × 105 1.5 × 108
0.53 1.1 × 105 2.2 × 108
0.32 3.6 × 104 1.1 × 108
2.24 1.8 × 105 8.0 × 107
6.51 6.3 × 105 9.6 × 107
65.06 6.6 × 104 9.5 × 105
41.00 5.3 × 104 1.3 × 106
12.4
10.3
f
1.1
f
983
760
a λex ) 420 nm for the mono- and homobimetallic complexes and
450 nm for the heterobimetallic complexes, in acetonitrile at 295 K.
All solution samples were treated three times with freeze-pump-thaw
cycles before the measurements of emission and quantum yields.
Luminescence lifetimes are obtained from the least-squares fit of single-
exponential decay. b Φem vs Ru(bpy)3(PF6)2 (Φem ) 0.062). c kr ) Φem
τ-1 d knr ) τ-1 - kr. e Measurements were performed on the phase-
.
modulated fluorescence system (500 ps to 200 ns). f Lifetime of the
quenched ReI excited state is shorter than the detection limit of the
system.
metallic OsII complexes containing the same spacers with
cumulenic bridges.3,6 In addition, from a comparison of the
extinction coefficients in the absorption spectra of the mono-
and homobimetallic complexes with ReI, RuII, or OsII centers,
the percentage of chromophores excited can be calculated at
different excitation wavelengths. It was determined that the ReI-
based 1MLCT band is responsible for ∼50% of the light
absorption at the excitation wavelength of 350 nm, while use
of light at 450 nm leads primarily to the excitation (∼80%) of
the RuII- and OsII-based chromophores. The consequences of
these results will be made evident in the later discussion of
energy transfer within this system.
The excited-state emission and lifetime data are collected in
Table 8 along with the resulting calculated values for the
observed quantum yield, Φ, and the radiative and nonradiative
decay rate constants, kr and knr, respectively. The monometallic
complexes ReC2P2e, ReC3P4, and ReC4P4 were found to exhibit
weak luminescence at 480, 480, and 530 nm, respectively. For
comparison, the emission maximum for the pentacarbonyl
starting material, Re(CO)5Cl, occurred at 500 nm.40 The
emission of ReC4P4 was slightly red shifted with respect to the
other two monometallic complexes with shorter spacers and the
Re(CO)5Cl starting material. This observation is a result of the
increased σ-donating ability of the longer spacer, as evidenced
by the previously reported electrochemical and XPS studies of
the RuII and OsII complexes,3a which appears to decrease the
gap between the ground state and the MLCT state. The same
trend was observed for the two homobimetallic species,
ReC3P4Re and ReC4P4Re, at 480 and 530 nm, respectively.
Regarding the lifetimes of the new complexes with the ReI
center, it was observed that as the length of the cumulenic ligand
was increased for the monometallic species, the lifetime of the
ReI exited-state emission also slightly increased: 2.1 ns for
ReC2P2e, 3.3 ns for ReC3P4, and 6.8 ns for ReC4P4. The same
trend was also witnessed for the two homobimetallic species,
ReC3P4Re and ReC4P4Re, 4.5 and 10.9 ns, respectively. As
previously discussed, the longer the rigid cumulenic backbone
of the spacer becomes, the more σ-donation occurs from the
spacer to the metal termini, causing a decrease in the triplet
energy level. Such lowering of the MLCT state may inhibit the
mixing with other higher energy excited states and results in
the slight prolongation of the triplet-state lifetime. For hetero-
metallic systems, the lifetimes from ReI excited state were not
Figure 6. (a) Comparison of the electronic absorption spectra between
(a) ReC3P4 (- - -) and ReC3P4 (‚‚‚) and (b) OsC3P4Re (- - -) and
OsC4P4Re (‚‚‚).
assignment of a spin-forbidden d-d transition,40 here the
absorption bands are 1 or 2 orders of magnitude greater in
intensity, which contradicts the assignment of a forbidden d-d
transition in this system. The parent complex of this series of
compounds, Re(CO)5Cl, was known to luminescent from a d-d
excited state.40b Upon replacement of two CO ligands with
bipyridine or o-phenanthroline, the lowest excited state was
reported as a metal-to-ligand charge transfer (MLCT) state.40,42a
The polyphosphine ligands used in this study, similar to
polypyridyl ligands, most likely possess low-lying ring and
polyene π-antibonding orbitals capable of participation in charge
transfer interactions. Hence, a metal-to-phosphine charge transfer
state is assigned as the emitting excited state based on the
literature study and the observed extinction coefficients.
1
The heterobimetallic compounds exhibited MLCT bands
similar to those observed for the corresponding RuII and OsII
monometallic complexes MCnP4 (M ) Ru, Os; n ) 3, 4). The
1MLCT bands for these heterobimetallic compounds occurred
around the same region (430-440 nm) as the ReI-based MLCT
band, but with stronger extinction coefficients (ꢀ ≈ 103-104
M-1 cm-1). The OsII complexes (OsC3P4Re and OsC4P4Re)
also exhibit an additional metal-to-ligand charge transfer band
that was triplet in nature (3MLCT). This 3MLCT band occurred
between 580 and 600 nm (Figure 6b) with extinction coefficients
ꢀ ≈ 103 M-1 cm-1. This assignment is in agreement with our
earlier observations in the corresponding mono- and homobi-

Cumulene-Bridged Complexes
Inorganic Chemistry, Vol. 39, No. 26, 2000 6049
Figure 7. Representative decay trace and fit for OsC3P4Re.
Figure 8. Representative comparison of the absorption spectrum
(- - -) with the emission spectrum (‚‚‚) of the OsII heterobimetallic
complex OsC3P4Re.
measurable on our current setup, and only lifetimes of the RuII
or OsII chromophores were obtained and are listed in Table 8.
A representative decay trace and fit for OsC3P4Re are shown
in Figure 7. All of the emission decay traces fit to single-
exponential decay curves, leading to the conclusion that the
emissions are predominately from the RuII- or OsII-based
chromophores.
ReI-based chromophores in this system are not as efficient as
the OsII- and RuII-based ones. However, even though the
heterometallic compounds in this study have relatively low kr
values, such values are still an order of magnitude greater than
the corresponding monometallic complexes MCnP4 (M ) Ru,
Os; n ) 3, 4) without the ReI pentacarbonyl unit.
The quantum yields, Φ, were calculated from the steady-
state emission measurements and referenced to a standard
solution of [Ru(bpy)3][PF6]2 (Φ ) 0.062 in acetonitrile).14 The
OsII-containing heterometallic compounds exhibited the largest
Φ (0.065 and 0.041), while those for the ReI monometallic
compounds were an order of magnitude smaller (∼10-3). The
Φ values of ReC3P4Re and ReC4P4Re were another order of
magnitude smaller (∼10-4) than the corresponding monometallic
complexes ReC3P4 and ReC4P4. The radiative decay rate
constant, kr, was calculated for each compound from the
emission quantum yield and the lifetime of the emitting MLCT
state (kr ) Φ/τ). The highly emissive OsII-containing com-
pounds, however, exhibited the smallest kr (∼104) with respect
to the ReI-only compounds (∼105-106), while also displaying
the smallest nonradiative decay constant (knr ) 1/τ - kr) (∼106).
While the low knr is in agreement with the high Φ, the relatively
low kr is in contradiction with what is expected from more
emissive compounds. For complexes with a higher degree of
electronic coupling across the bridging units, there is also the
possibility of the presence of low-lying, unoccupied orbitals,
including the low-energy ligand-based excited states.43,44 There-
fore, the low observed kr may be a result of the deactivation of
the emitting 3MLCT state by mixing with low lying nonemitting
ligand-based excited states (i.e., pπ*(L)). Another possibility
Energy Transfer. The rate constant of energy transfer could
not be calculated for the four heterobimetallic complexes with
ReI/RuII and ReI/OsII donor/acceptor pairs due to the inability
to determine the lifetime of the ReI donor in heterometallic
systems. Therefore, the mechanism by which any possible
energy transfer process occurs could not be identified. However,
these complexes exhibited excited-state behavior that was
indicative of an efficient energy transfer. The emission maxima
for the four heterobimetallic complexes were found between
580 and 610 nm (Table 8), which were assigned as RuII- and
OsII-centered on the basis of comparison of their luminescence
properties with the corresponding monometallic model com-
plexes with the same spacers.3 These emission maxima are red-
shifted significantly from the ReI-based mono- and homobime-
tallic complexes and occur in the region associated with the
RuII and OsII 3MLCT energies. The fact that no ReI-based
emission is detected is a possible consequence of the energy
transfer from the higher MLCT ReI excited state to the lower
RuII- and OsII-based excited states. A representative comparison
of the absorption spectrum with the emission spectrum of the
heterobimetallic OsC3P4Re is shown in Figure 8. It is also noted
that the observed emission maxima from the ReI centers
occurred in the same region and overlapped with the absorption
energy of the corresponding RuII and OsII monometallic
complexes (440-460 nm and 550-615 nm, respectively). In
addition, the driving force, ∆G°, of such energy transfer can
be estimated for the ReI/RuII and ReI/OsII donor/acceptor
combinations. By calculating the difference in the spectroscopic
energies of the donor and acceptor using the equation ∆G° )
1
is the rapid relaxation of the MLCT state to the ground state
before a substantial amount of intersystem crossing can occur
to afford the emitting 3MLCT state. The former would explain
how a kr value could be decreased without increasing knr of the
same system. The ReI-only complexes have the greatest knr,
which is in agreement with the low Φ values. Apparently, the
em(donor) - λem(acceptor),39,44 the driving force was found to
(43) Baba, A. I.; Ensley, H. E.; Schmehl, R. H. Inorg. Chem. 1995, 34,
1198-1207.
λ
be exothermic for the four heterobimetallic complexes. The
values thus obtained were -0.36 eV for RuC3P4Re, -0.44 eV
for OsC3P4Re, -0.25 eV for RuC4P4Re, and -0.46 eV for
OsC4P4Re.
(44) (a) Sauvage, J.-P.; Colin, J.-P.; Chambron, J.-C.; Guillercz, S.; Cudret,
C.; Balzani, V.; Barigelletti, F.; De Cola, L.; Flamigni, L. Chem. ReV.
1994, 94, 993. (b) Barigelletti, F.; Flamigni, L.; Balzani, V.; Collin,
J.-P.; Sauvage, J.-P.; Sour, A.; Constable, E. C.; Thompson, A. M.
W. C. J. Am. Chem. Soc. 1994, 116, 7692. (c) Schlicke, B.; Belser,
P.; De Cola, L.; Sabbioni, E.; Balzani, V. J. Am. Chem. Soc. 1999,
121, 4207.
In the aforementioned electronic absorption study, it was
determined that at 350 nm the ReI center absorbs ca. 50% of

6050 Inorganic Chemistry, Vol. 39, No. 26, 2000
Ortega et al.
the incident light energy, while at 450 nm the ReI chromophore
absorbs only ca. 20%. Excitation using both wavelengths of
heterobimetallic complexes led to practically identical quantum
yield results (ca. 1% difference). This result suggests that
irrespective of which chromophore absorbs the excitation
radiation, the emission is occurring from a single metal center
(i.e., Os(II)* or Ru(II)*) in each heterometallic complex with
CnP4 (n ) 3, 4) spacers.
N-donor bpy ligands by the P-donor rigid spacers. The decrease
in σ-donation from the P-donors with respect to the N-donors
caused the destabilization of the MLCT excited state and the
stabilization of the ground state, resulting in a blue shift to higher
energies for these molecular complexes. A red shift was
observed as the length of the spacer was increased resulting
from the higher amount of π-electrons in the spacer, which gives
rise to an increase in the σ-donation from the phosphine units.
Due to the weak and relatively short excited-state lifetimes
of the ReI components in the heterobimetallic complexes, the
energy transfer rate constant could not be calculated and the
mechanism by which any possible energy transfer occurs could
not be identified. However, on the basis of their excited-state
behavior, these complexes were found to be capable of energy
transfer from the ReI center to the RuII and OsII centers. Only
Concluding Remarks
Several different characterization techniques were successfully
employed in the identification and study of the multicomponent
complexes with cumulenic spacers. The FAB mass spectral
technique, as a relatively soft ionization process, afforded the
characterization of the fragmentation and isotope patterns as
well as the identity of the species. While different fragmentation
did occur, the loss of the counterion PF6, the linkage component
PPh2, and the ligands Cl and CO were observed. The addition
of oxygen atoms was also observed as a characteristic phenom-
enon in the FAB/MS analysis. Furthermore, single-crystal X-ray
structure determination provides a comparison of the metal-to-
metal distances and dihedral angles that exist between the odd-
and even-numbered cumulenic spacers. The metal-to-metal
distance for ReC4P4Re was found to be 1.33 Å longer than in
ReC3P4Re. Due to the resulting four-membered ring upon
complexation to the ReI center, the ligand coordination angle
for both complexes was much smaller than the ideal 120°
expected for sp2-hybridized carbons. In addition, the IR
techniques have also been utilized to describe the electronic
donating ability of the different cumulenic spacers in addition
to confirming the local coordination environment described by
X-ray crystal structure determination for the ReI units.
3
the RuII- and OsII-based MLCT emissions were detected, and
excitation at different wavelengths, where a different percentage
of different metal-based chromophores were excited, resulted
in identical quantum yield values. These observations are
indicative of a highly efficient energy transfer process across
the polyphosphine spacers with cumulenic carbon chains.
Acknowledgment. This work was supported by the National
Science Foundation CAREER award (CHE-9733546). We thank
Dr. John Greaves (UCI Mass Spectral Laboratory) for his
assistance in the FAB/MS analysis.
Supporting Information Available: Atomic coordinates, thermal
parameters, and bond lengths and angles have been deposited at the
Cambridge Crystallographic Data Center (CCDC). Any request to the
CCDC for this material should quote the full literature citation and the
reference number 146268 for ReC3P4Re and 146269 for ReC4P4Re.
Tables listing detailed crystallographic data, atomic parameters, and
bond lengths and bond angles. This material is available free of charge
via the Internet at http://pubs.acs.org.
The photophysical properties of these supramolecular com-
plexes with rigid spacers were explored through the utilization
of several spectroscopic techniques. The electronic absorption
spectra characterized the effects of the substitution of the
IC0006910