Sub-picosecond IR study of the reactive intermediate in an alkane C-H bond activation reaction by CpRh(CO)2

Article abstract of DOI:10.1021/om980194f

The reactive intermediate in a cyclohexane C-H bond activation reaction by CpRh(CO)₂ has been identified as the cydohexane solvate η⁵-CpRh(CO)(C₆H₁₂) by sub-picosecond infrared spectroscopy.

Full text of DOI:10.1021/om980194f

Organometallics 1998, 17, 3417-3419
3417
Su b-P icosecon d IR Stu d y of th e Rea ctive In ter m ed ia te in
a n Alk a n e C-H Bon d Activa tion Rea ction by Cp Rh (CO)2
J ohn B. Asbury,† Hirendra N. Ghosh,† J ake S. Yeston,‡ Robert G. Bergman,‡ and
Tianquan Lian*,†
Department of Chemistry, Emory University, Atlanta, Georgia 30322, and Chemical Sciences
Division, Lawrence Berkeley National Laboratory and Department of Chemistry, University of
California, Berkeley, California 94720
Received March 16, 1998
Summary: The reactive intermediate in a cyclohexane
C-H bond activation reaction by CpRh(CO)2 has been
identified as the cyclohexane solvate η5-CpRh(CO)(C6H12)
a
b
c
by sub-picosecond infrared spectroscopy.
The photochemistry of transition-metal complexes,
such as CpM(CO)2 (M ) Rh, Ir), capable of undergoing
oxidative addition with alkane carbon-hydrogen bonds
has been studied intensely.1,2 This type of reaction is
traditionally thought to begin by the initial dissociation
of a ligand from the electronically saturated (18-
electron) metal complex to form an electronically un-
saturated (16-electron) intermediate. It is now well-
established (see below) that this reactive intermediate
rapidly forms a complex (or solvate) with an alkane
molecule, and in a subsequent step, one of the C-H
bonds in this complexed alkane undergoes oxidative
addition at the metal center.1,2 The first steps of this
reaction have been identified by experiments in the gas
phase,3 liquefied noble gases,4 and low-temperature
matrixes.5 However, the detailed mechanism of the
subsequent bond activation step is not well understood,
due to the difficulties in characterizing short-lived
alkane complex intermediates.
Recently, a C-H oxidative addition reaction was
studied by ultrafast time-resolved spectroscopy.6,7 This
involved a complete time-resolved IR spectroscopic study
of the C-H bond activation reaction of Tp*Rh(CO)2 (Tp*
) HB-Pz3*, Pz* ) 3,5-dimethylpyrazolyl), shown in
Figure 1c, in room-temperature alkane solution covering
a time range from a few femtoseconds to milliseconds.6
Several reactive intermediates in the overall C-H bond
activation process were identified, and the correspond-
ing reaction rates of various steps were determined. Two
sequentially formed intermediates were detected, one
of which was an η2-Tp*Rh(CO)(alkane) complex in
which a pyrazolyl-Rh bond has been cleaved.
F igu r e 1. Structures of (a) η5-CpRh(CO)(C6H12, (b) η3-
CpRh(CO)(C6H12), and (c) Tp*Rh(CO)2.
The above results raise the possibility that additional
ligand detachment, rather than simple CO loss, might
also be involved in C-H bond activation involving other
organometallic compounds, especially the much-studied
complexes Cp*M(CO)2 or CpM(CO)2 (Cp* ) C5Me5).8,9
In these systems, such intermediates could be generated
by partial decomplexation of the Cp or Cp* ring in the
initially generated monocarbonyl solvate to give solvates
of “ring-slipped” η3-CpRhCO, as shown in Figure 1b.
So far, only the highly methyl-substituted Cp* dicar-
bonyls have been subjected to ultrafast spectroscopic
investigation. Unfortunately, due to their very low
quantum yields for CO dissociation, the reactive inter-
mediates involved in C-H bond activation reactions
formed from Cp*M(CO)2 have not been observed in
these studies in room-temperature alkane solutions.10,11
The smaller yield was attributed to the generation of
nondissociative excited states which relax back to the
ground state in less than 40 ps. In this communication,
we report the first ultrafast study of the photochemistry
of the related unmethylated complex CpRh(CO)2.
A
higher quantum yield for CO loss in this molecule
compared to the Cp* analogue has enabled us to observe
the reactive intermediate directly and follow the time
course of its reactions up to 1 ns. This molecule serves
as an excellent system in which to study metal-alkane
interactions because its relative simplicity is more
† Emory University.
‡ University of California, Berkeley.
(1) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.
Acc. Chem. Res. 1995, 28, 154-162.
(2) Perutz, R. N. Chem. Soc. Rev. 1993, 22, 361-369.
(3) Wasserman, E. P.; Moore, C. B.; Bergman, R. G. Science 1992,
255, 315-318.
(4) Bengali, A. A.; Arndtsen, B. A.; Burger, P. M.; Schultz, R. H.;
Weiller, B. H.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. Pure Appl.
Chem. 1995, 67, 281-288.
(8) Schultz, R. H.; Bengali, A. A.; Tauber, M. J .; Weiller, B. H.;
Wasserman, E. P.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. J . Am.
Chem. Soc. 1994, 116, 7369-7377.
(5) Rest, A. J .; Whitwell, I.; Graham, W. A. G.; Hoyano, J .; McMas-
ter, A. D. J . Chem. Soc., Dalton Trans. 1987, 1181.
(6) Bromberg, S. E.; Haw, Y.; Asplund, M. C.; Lian, T.; McNamara,
B. K.; Kotz, K. T.; Yeston, J . S.; Wilkens, M.; Frei, H.; Bergman, R.
G.; Harris, C. B. Science 1997, 278, 260-263.
(9) Bengali, A. A.; Bergman, R. G.; Moore, C. B. J . Am. Chem. Soc.
1995, 117, 3879-3880.
(10) Bromberg, S. E.; Lian, T. Q.; Bergman, R. G.; Harris, C. B. J .
Am. Chem. Soc. 1996, 118, 2069-2072.
(7) Lian, T.; Bromberg, S. E.; Yang, H.; Proulx, G.; Bergman, R. G.;
Harris, C. B. J . Am. Chem. Soc. 1996, 118, 3769-3770.
(11) Dougherty, T. P.; Grubbs, W. T.; Heilweil, E. J . J . Phys. Chem.
1994, 98, 9396-9399.
S0276-7333(98)00194-0 CCC: $15.00 © 1998 American Chemical Society
Publication on Web 07/09/1998

3418 Organometallics, Vol. 17, No. 16, 1998
Communications
F igu r e 2. Transient IR spectra of CpRh(CO)2 in cyclo-
hexane after 267 nm photolysis. The spectra at different
delay times have been displaced vertically for comparison.
F igu r e 3. Kinetics of CpRh(CO)2 in cyclohexane measured
at 1963 (Panel a), 1985 (panel b), and 2048 cm-1 (panel c).
amenable than the permethylated system or Tp*Rh-
(CO)2 to high-level electronic structure calculations.12-14
The 1 kHz sub-picosecond tunable IR spectrometer15
used for this study has a 150 fs time resolution and a
tunable IR probe wavelength from 1 to 10 µm. A flowing
cell (250 µm path length) containing CpRh(CO)2 in
cyclohexane was pumped by a 3 µJ photolysis pulse with
a FWHM of ca. 100 fs at 267 nm, and the subsequent
IR spectral changes in the CO stretching mode region
were recorded as a function of time for up to 1 ns. After
passing through the sample, the probe IR pulses are
dispersed in a monochromator to obtain a 2 cm-1
spectral resolution. The pump and probe beam diam-
eters at the sample are 350 and 250 µm, respectively.
The relative polarization of the pump UV and probe IR
beams was set at 54.7° to ensure that the measured
absorbance change was due to population dynamics
only. CpRh(CO)2 was synthesized according to a pub-
lished procedure16 and fully characterized by conven-
tional spectroscopic methods. The sample solution (ca.
10 mM concentration) was kept under dry N2 and
continuously flowed through an airtight cell to ensure
that each laser pulse probed a fresh volume of sample.
Shown in Figure 2 are the transient IR difference
spectra obtained upon irradiation of CpRh(CO)2 in
cyclohexane recorded at 1, 10, 100, and 500 ps after the
267 nm photolysis pulse. The negative peaks at the
frequencies corresponding to the parent CO stretching
bands of 1985 and 2048 cm-1 indicate the depletion of
parent molecules due to photolysis. A single new band
at 1963 cm-1 grows in, suggesting the formation of only
one transient solvate.17 After photolysis, the recovery
of the bleach at the parent CO bands indicates the
partial reformation of the parent molecules. Shown in
Figure 3 are the kinetics measured at the parent
absorption bands (panels b and c) and the new product
band at 1963 cm-1 (panel a). The bleach recovery
kinetics are well fit to a single-exponential recovery time
of 55 ( 5 ps. For the new product band, the kinetics
can be well fit by an exponential rise time of 10 ps and
a small 60 ps decay component. A small solvent
background signal centered around t ) 0 has been
subtracted from all the figures presented here.18
As shown in Figures 2 and 3, about 75% of the
bleached absorptions at 1985 and 2048 cm-1 have
recovered within 200 ps. The corresponding 25% loss
of the parent molecules can be attributed to the CO loss
channel, since no long-lived nondissociative excited
states were present. For CpRh(CO)2 in decalin solution,
the quantum yield for photosubstitution by phos-
phine16,22 was determined earlier to be around 20% with
313 nm photolysis. This yield is consistent with our
estimate of a 25% CO loss channel in our experiments.
The measured 55 ps recovery time of the bleaches is due
to the return of excited parent molecules to the lowest
vibrational level of their ground electronic state; this
time constant contains the contributions of both elec-
tronic relaxation (radiationless decay) as well as vibra-
tional relaxation. Metal carbonyls with excess vibra-
tional energy have been shown earlier to exhibit broad
and featureless IR bands which are red shifted com-
pared to the CO stretching bands of the corresponding
vibrationally cold molecules. These broad features can
be attributed to the anharmonic coupling of CO stretch-
(18) Under the same experimental conditions, a small signal in neat
cyclohexane, which has a peak absorbance change of ca. 5 mOD at t )
0 and decays to negligible values at 5 ps, was observed. This pump-
induced solvent signal, whose origin is so far not clear, appears to be
independent of wavelength in this spectral region. An identical feature
can also be observed in CpRh(CO)2 samples and is subtracted from all
the figures presented here.
(12) Musaev, D. G.; Morokuma, K. J . Am. Chem. Soc. 1995, 117,
799-805.
(13) Couty, M.; Bayse, C. A.; J imenez-Catano, R.; Hall, M. B. J . Phys.
Chem. 1996, 100, 13976.
(14) Saillard, J .-Y.; Hoffmann, R. J . Am. Chem. Soc. 1984, 106, 2006.
(15) Ghosh, H. N.; Asbury, J . B.; Lian, T. J . Phys. Chem. B, in press.
(16) Purwoko, A. A.; Drolet, D. P.; Lees, A. J . J . Organomet. Chem.
1995, 504, 107-113.
(19) Dougherty, T. P.; Heilweil, E. J . Chem. Phys. Lett. 1994, 227,
19-25.
(17) The small feature at 1940 cm-1 has been carefully examined
in a separate experiment that focused on that region. We found that
it was not a reproducible peak within the current signal-to-noise level
of our experiment. Similarly, within the signal-to-noise of the data,
(20) Grubbs, W. T.; Dougherty, T. P.; Heilweil, E. J . Chem. Phys.
Lett. 1994, 227, 480-484.
(21) Beckerle, J . D.; Casassa, M. P.; Cavanagh, R. R.; Heilweil, E.
J .; Stephenson, J . C. Chem. Phys. 1992, 160, 487-497.
(22) Dunwoody, N.; Lees, A. J . Organometallics 1997, 16, 5770.
there are no other new peaks besides the one at 1963 cm-1
.

Communications
Organometallics, Vol. 17, No. 16, 1998 3419
ing modes with highly populated low-frequency modes
in the molecules. As the metal carbonyls dissipate their
excess vibrational energy to the surrounding solvent
molecules, the broad features narrow and shift to higher
energy and the bleach recovers.19 This type of feature
is also present in our transient spectra at 1 and 10 ps
as shown in Figure 2. The vibrational relaxation times
in metal carbonyls of similar size are typically about
10-20 ps for low-frequency modes and over 30 ps for
CO stretching modes.20,21 We conclude that irradiation
of CpRh(CO)2 leads to excited states with a dissociative
channel and a short-lived nondissociative channel. The
lifetime of the latter channel, which undergoes radia-
tionless decay back to the electronic ground state, is
estimated to be less than 10 ps based on the lack of
distinct vibrational peaks associated with this channel
at 10 ps.
spectra for these vibrationally hot nascent species. A
10 ps rise time is typical of the vibrational relaxation
time for low-frequency modes.19 The small 60 ps decay
component measured at 1963 cm-1, which is identical
to the bleach recovery time, can be attributed to the
cooling of hot CpRh(CO)2 molecules, whose broadened
and red-shifted CO stretching bands have a small
absorption in the 1920-1980 cm-1 and 2000-2050 cm-1
region, as shown in Figure 2.
The final C-H bond-activated product, CpRh(CO)(R)-
(H), has a CO stretching mode at around 2024 cm-1 5
.
None of this material is formed on the nanosecond-or-
shorter time scale. We expect that the C-H bond
cleavage should occur on the 10-100 ns time scale,
based on the results of the Tp*Rh(CO)2 study6 and the
estimated activation barrier measured in the Cp*Rh-
(CO)2/liquefied noble gas experiments8 discussed above.
Our ongoing research will investigate the bond activa-
tion step by measuring the kinetics in the nanosecond
to millisecond time scale using step-scan FTIR. The
activation of Si-H bonds, which was reported to occur
in the picosecond time scale for CpMn(CO)3,24 will also
be investigated for the CpRh(CO)2 system.
Our experiment is the first direct observation of the
reactive monocarbonyl-alkane complex, η5-CpRh(CO)-
(C6H12), in room-temperature alkane solution. Our
results are consistent with the idea that this alkane
complex solvate is formed on the pathway to the final
C-H oxidative addition product, as suggested for the
Cp* and Tp* systems. Because we see only one inter-
mediate here, there is a clear difference between the
behavior of this system and that of Tp*Rh(CO)2, where
a second (presumably pyrazolyl-detached) species was
detected. It seems reasonable that ligand detachment
would be more favorable in that system, since bis-
(pyrazolyl)borate complexes have been identified as
stable species in closely related complexes.
The reactive channel leads to a transient that exhibits
only one new metal carbonyl IR stretching band, sug-
gesting that this species is a monocarbonyl intermedi-
ate. In a previous study of the photolysis of CpRh(CO)2
in an Ar matrix,5 a CO stretching band at 1968 cm-1
was assigned to η5-CpRh(CO). The well-known stable
18-electron η5-CpRh(CO)(PR3) complexes typically ab-
sorb near 1952 cm-1 16,22
We therefore believe that the
.
transient species detected in our experiments is the
solvated complex η5-CpRh(CO)(C6H12), shown in Figure
1a. This assignment is consistent with the observation
that there is no further change of its frequency up to 1
ns. The “naked” 16-electron complex is not directly
observed here (nor was it observed in the previous Tp*
study) because the complex formation between CpRh-
(CO) and a surrounding solvent molecule is expected to
be a process with a negligible barrier and should occur
in the hundreds of femtosecond to a few picosecond time
scale at room-temperature solution in which solute-
solvent collision occurs on the 100 fs time scale. For
example, solvate formation between nascent Cr(CO)5
and alkanes and alcohols23 was observed to occur in the
1-2 ps time scale. The formation time of the CpRh-
(CO)(C6H12) band at 1963 cm-1 is well fit by a 10 ps
exponential rise as shown in Figure 3. This formation
time contains contributions from both the formation of
vibrationally hot CpRh(CO)(C6H12) and the cooling of
the hot monocarbonyl. Identifying the spectral features
of the short-lived naked species and separating the time
scale for solvent addition and vibrational cooling is
difficult because of the very broad nature of the IR
Ack n ow led gm en t. The investigators at Emory ac-
knowledge the financial support from the Petroleum
Research Fund, the Emory University Research Com-
mittee, and the National Science Foundation CAREER
award under Grant No. 9733796. The investigators at
LBNL/UC Berkeley were supported by the Director,
Office of Energy Research, Office of Basic Energy
Sciences, Chemical Sciences Division, U.S. Department
of Energy, under Contract No. DE-AC03-76SF00098.
OM980194F
(24) Yang, H.; Kotz, K. T.; Asplund, M. C.; Harris, C. B. J . Am.
Chem. Soc. 1997, 119, 9564.
(23) Simon, J . D.; Xie, X. J . Phys. Chem. 1987, 91, 5538.