First direct observation of a CO-bridged primary photoproduct of [Ru3(CO)12] by picosecond time-resolved IR spectroscopy

Article abstract of DOI:10.1039/b201272c

For the first time, a CO-bridged primary photoproduct was observed for [Ru₃(CO)₁₂] by using picosecond time-resolved IR spectroscopy (ps-TRIR).

Full text of DOI:10.1039/b201272c

First direct observation of a CO-bridged primary photoproduct of
Ru (CO) ] by picosecond time-resolved IR spectroscopy
[
3
12
a
a
b
a
b
Frank W. Vergeer, Franti s˘ ek Hartl,* Pavel Matousek, Derk J. Stufkens and Michael Towrie
a
Institute of Molecular Chemistry, Universiteit van Amsterdam, Nieuwe Achtergracht 166, NL-1018 WV
Amsterdam, The Netherlands. E-mail: hartl@science.uva.nl
Central Laser Facility, CLRC Rutherford Appleton Laboratory, Chilton, Didcot, Oxfordshire, UK
OX110QX
b
Received (in Cambridge, UK) 5th February 2002, Accepted 25th April 2002
First published as an Advance Article on the web 7th May 2002
9
For the first time, a CO-bridged primary photoproduct was
clusters [Os
copy. In this work we present the results of a ps-TRIR study of
the reactive photoisomer of [Ru (CO)12], which represents the
3
(CO)10L] (L = 1,3-diene) with ns-TRIR spectros-
3
observed for [Ru (CO)12] by using picosecond time-resolved
IR spectroscopy (ps-TRIR).
3
first application of this technique on a transition metal cluster.
Excitation at 400 nm followed by IR probing, was used to
investigate if a bridging carbonyl is indeed present in this
product.
Photochemical and photophysical studies of transition metal
carbonyl clusters are of considerable interest, not only because
of the potential of these compounds to act as versatile catalysts
1
or catalyst precursors, for example in fine chemistry, but also
The ground state IR spectrum of [Ru
3
(CO)12]† in heptane at
in view of their challenging application as key components of
more complex supramolecular systems. The photoactivation of
room temperature shows four distinct n(CO) IR bands at 2061,
2
1
2031, 2017 and 2012 cm . Picosecond TRIR spectra‡ were
recorded in this solvent at several pump-probe delays between
0 and 500 ps after excitation into the lowest-energy absorption
band, having predominantly metal–metal bonding to metal–
metal antibonding (s ? s*) character. Fig. 2 shows representa-
tive spectra at six selected time delays.
After excitation, the first spectra display instantaneous
bleaching of the parent n(CO) bands superimposed on a broad,
unresolved transient absorption due to the excited state of
(
thermally stable) cluster compounds may lead to novel reaction
2
types with high selectivity. In supramolecular systems, where
clusters may function as a bridge between a donor and an
acceptor site, the changes in the electronic and structural
properties of a cluster upon excitation may be utilized, for
example, in photoinduced vectorial electron/energy transport
from the donor to the acceptor.
Time-resolved IR spectroscopy (TRIR), where UV-visible
flash photolysis is combined with (ultra)fast IR detection, is a
powerful tool for probing the primary events after photoexcita-
tion. This applies in particular to complexes containing strongly
IR-active ligands like CO, that can act as direct IR probes of the
electron density at the metal centre. As the excited states of
transition metal clusters are usually too short-lived to be studied
with nanosecond (ns) transient absorption or TRIR, faster
spectroscopic techniques are required to characterize the
excited state and to study the formation of the primary
photoproducts. Although several picosecond (ps) TRIR studies
on simple, mononuclear transition metal complexes with e.g.
[Ru
bands were observed upon excitation into the s ? s* transition
of [M
(CO)10] (M = Mn, Re)10 and have been reported to result
3
(CO)12]. Similar broad, featureless transient absorption
2
from the appearance of red-shifted CO stretching frequencies
due to the anharmonic coupling with low-frequency (M–C
stretching and M–C–O bending) vibrational modes, which are
highly activated upon light excitation.
In the first few picoseconds, this broad transient absorption
resolves into distinct product absorption bands (2051, 2021 and
2
1
2007 cm ) as the bands become narrower and slightly shift to
higher frequency. These spectral changes are attributed to
3
4
5
I
M(0) (M = Cr, W), M(II) (M = Ru, Os) and Co(
) centres
vibrational relaxation of the low-frequency vibrational
modes.3
,5,10
Upon further decay (520 ps) another positive shift
have been reported, ps-TRIR studies of polynuclear transition
metal complexes are scarce. In order to investigate whether this
technique could contribute to the unraveling of complex
mechanistic problems in transition metal cluster photochem-
istry, we performed a ps-TRIR study on the model cluster
2
1
21
of ca. 5 cm is observed for the 2021 and 2051 cm product
bands, which is attributed to the formation of a primary
photoproduct. On this time scale also a new n(CO) band grows
2
1
in at 1850 cm . The latter band is assigned to a bridging
[
Ru
The photoreactivity of the triangular clusters [M
Ru, Os) has been studied in great detail over the last few
3
(CO)12].
3
(CO)12] (M
=
6–8
decades. In brief, photoexcitation into the two lowest-energy
absorption bands of these clusters has been assumed to result in
3
formation of a reactive isomer of [M (CO)12] (M = Ru, Os),
which is supposed to be the key intermediate for subsequent
reactions but has never been observed directly. This reactive
7
isomer was proposed to consist of an open cluster core in which
one M–M bond is heterolytically split and one carbonyl has
moved to a bridging position to maintain the charge neutrality at
both M atoms (see Fig. 1).
In contrast to this, two different bridging carbonyl ligands
were recently revealed for a photoproduct of the substituted
3
Fig. 2 TRIR spectra of [Ru (CO)12] in heptane at (-) 0 ps, (1) 3 ps, (:) 20
ps, (∂ ) 40 ps, (/) 80 ps and (5) 500 ps after 400 nm (ca. 150 fs FWHM,
5 mJ per pulse) excitation. The arrows indicate the shift of the band maxima
with increasing time delay following excitation.
Fig. 1 Proposed structure of the open-core primary photoproduct of
3
[M (CO)12] (M = Ru, Os).
1
220
CHEM. COMMUN., 2002, 1220–1221
This journal is © The Royal Society of Chemistry 2002

carbonyl and decays on a similar time scale (20–500 ps) as the
bands in the terminal n(CO) region. Upon decay of the transient
absorption bands parent bleach recovery also takes place. On
early time scales (520 ps), the partial recovery of the parent
[Ru
likely to be formed from [Ru
tion of this species may therefore be due to irradiation into the
tailing higher-energy transition of [Ru (CO)12], which is known
to result in CO loss. Concerning the structure of the
[Ru (CO)11(m-CO)] photoproduct, no unambiguous conclu-
sions can be drawn. The close correspondence with the n(m-CO)
3
(CO)11].8 According to the literature, [Ru
(CO)11(m-CO)] and the observa-
3
(CO)11] is not
3
3
2
1
bleach at 2061 cm is mainly due to increased overlap with the
product absorption band. This is inferred from the observation
3
2
1
that the parent bleach at 2031 cm , for which no such change
in overlap takes place, only shows a minor decrease in signal
strength. The excited state is therefore assumed to be almost
completely converted into the primary photoproduct. On longer
time scales (up to 500 ps), the shape and position of the transient
absorption bands do not change and recovery of the parent
bleach signals on these time scales is accordingly ascribed to the
regeneration of the parent cluster. At 500 ps after the laser pulse
the initially formed transient absorption bands have almost
completely disappeared and two small remaining bands at 2040
2
1 8
3
stretching frequency of [Ru (CO)11] (1840–1860 cm ) sug-
gests cleavage of a M–M bond and formation of a single CO
bridge. Quantum chemical calculations including geometry
optimizations to support these assignments, are in progress.
In conclusion, picosecond TRIR spectroscopy proves to be a
powerful tool in the field of photochemistry of transition metal
carbonyl clusters. For the first time we were able to observe and
characterize the primary photoproduct of [Ru (CO)12]. The
3
appearance of a single n(CO) band in the bridging carbonyl
region supports the formation of a reactive isomer of
2
1
and 2007 cm
indicate the formation of a minor amount
(
< 10%) of a second, longer-lived photoproduct. The in-
3
[Ru (CO)12], as proposed in the literature.
complete bleach recovery supports this latter conclusion.
The nearly complete reversibility of the system under the
experimental conditions implies that the cluster core remains
We acknowledge the financial support from the Council for
Chemical Sciences of the Netherlands Organization for Scien-
tific Research (NWO-CW, project No. 348-032; F. W. V. and F.
H.) and from the European Union (LSF ref. no. USEV13C2/
01).
intact and agrees with the formation of [Ru (CO)11(m-CO)] as
3
the primary photoproduct. As both vibrational relaxation
processes and decay of the excited state take place within a few
picoseconds, the determination of the excited state lifetime from
the decay of the terminal n(CO) bands is hampered. However,
Notes and references
2
1
† [Ru
Aldrich) were used as received.
The ps-TRIR spectra were recorded using the PIRATE set-up at the
3
(CO)12] (Strem Chemicals) and heptane (spectroscopic grade,
the grow in of the bridging n(CO) band at 1850 cm is not
accompanied by a shift to higher frequency and is therefore
assumed not to be influenced by vibrational relaxation proc-
esses. Gaussian curve fitting was therefore performed on this
well separated bridging n(CO) band. Plotting the peak area of
the 1850 cm band for each time delay against time allows the
determination of both the excited state lifetime (3.9 ± 0.9 ps,
Fig. 3(a)), which is assumed to correspond with the grow in of
‡
Central Laser Facility of the Rutherford Appleton Laboratory. Second
harmonic generation of a part of the 800 nm output of a Ti-sapphire
regenerative amplifier (1 kHz, 150 fs, 2 mJ) produced 400 nm pulses for
excitation of the sample. The other part of the 800 nm light was used to
pump an optical parametric amplifier generating tuneable mid IR outputs
2
1
21
(150–200 cm FWHM, 200 fs) by frequency down conversion of the
2
1
signal and idler outputs in an AgGaS crystal. Changes in the IR absorption
the 1850 cm
band, and the lifetime of the primary
2
were recorded by normalising the outputs from a pair of HgCdTe linear
array detectors on a shot-by-shot basis.
photoproduct (56.6 ± 6 ps, Fig. 3(b)). The latter lifetime is in
good agreement with the values obtained from the terminal
n(CO) bands (e.g. t = 52.5 ± 4 ps at 2051 cm21, Fig. 3(c)),
1
2
P. Braunstein and J. Rosé, Comprehensive Organometallic Chemistry,
Pergamon, New York, 1995, vol. 10; D. Mani and H. Vahrenkamp, J.
Mol. Catal., 1985, 29, 305; I. Ojima and Z. Zhang, J. Org. Chem., 1988,
53, 4425; P. Braunstein and J. Rosé, Catalysis by Di- and Polynuclear
Metal Cluster Complexes, Wiley–VCH, New York, 1998.
N. E. Leadbeater, J. Chem. Soc., Dalton Trans., 1995, 2923; N. E.
Leadbeater, J. Organomet. Chem., 1999, 573, 211; J. Nijhoff, M. J.
Bakker, F. Hartl, D. J. Stufkens, W.-F. Fu and R. van Eldik, Inorg.
Chem., 1998, 37, 661; E. W. Ainscough, A. M. Brodie, R. K. Coll, T. G.
Kotch, A. J. Lees, A. J. A. Mair and J. M. Waters, J. Organomet. Chem.,
whose decay after t
d
= 20 ps can be mainly ascribed to
regeneration of the parent cluster. The observed kinetics rule out
CO loss as the primary photoprocess since, assuming that
photoexpelled CO escapes from the solvent cage, the back-
reaction in this case would occur under diffusion control and
would therefore take place on a much longer time scale.
The IR bands of the remaining photoproduct (after 500 ps)
are close to those reported for the unsaturated cluster
1
996, 517, 173; A. J. Arce, A. J. Deeming, Y. De Sanctis, D. M. Speel
and A. Di Trapani, J. Organomet. Chem., 1999, 580, 370.
3
4
5
6
T. P. Dougherty and E. J. Heilweil, Chem. Phys. Lett., 1994, 227, 19.
M. W. George and J. J. Turner, Coord. Chem. Rev., 1998, 177, 201.
T. P. Dougherty and E. J. Heilweil, J. Chem. Phys., 1994, 100, 4006.
A J. Poë and C. V. Sekhar, J. Am. Chem. Soc., 1986, 108, 3673; J. G.
Bentsen and M. S. Wrighton, J. Am. Chem. Soc., 1987, 109, 4518; D. R.
Tyler, M. Altobelli and H. B. Gray, J. Am. Chem. Soc., 1980, 102, 3022;
M. F. Desrossiers, D. A. Wink, R. Trautman, A. E. Friedman and P. C.
Ford, J. Am. Chem. Soc., 1986, 108, 1917; J. G. Bentsen and M. S.
Wrighton, J. Am. Chem. Soc., 1987, 109, 4530; J. A. Dibenedetto, D. W.
Ryba and P. C. Ford, Inorg. Chem., 1989, 28, 3503; P. C. Ford, J.
Organomet. Chem., 1990, 383, 339; N. E. Leadbeater, J. Chem. Soc.,
Dalton Trans., 1995, 2923.
7
8
9
M. F. Desrosiers and P. C. Ford, Organometallics, 1982, 1, 1715; J.
Malito, S. Markiewicz and A. Poë, Inorg. Chem., 1982, 21, 4335.
F.-W. Grevels, W. E. Klotzbücher, J. Schrickel and K. Schaffner, J. Am.
Chem. Soc., 1994, 116, 6229.
M. J. Bakker, F. W. Vergeer, F. Hartl, O. S. Jina and X.-Z. Sun and M.
W. George, Inorg. Chim. Acta, 2000, 300–302, 597.
Fig. 3 Kinetic traces of [Ru
3
(CO)12] in heptane representing (a) the grow in
10 J. C. Owrutsky and A. P. Baronavski, J. Chem. Phys., 1996, 105, 9864;
H. Yang, T. Snee, K. T. Kotz, C. K. Payne and C. B. Harris, J. Am.
Chem. Soc., 2001, 123, 4204.
21
21
of the n(CO) band at 1850 cm (b) the decay of the 1850 cm band and
1
(
c) the decay at 2051 cm2
.
CHEM. COMMUN., 2002, 1220–1221
1221