Exchanging conformations of a hydroformylation catalyst structurally characterized using two-dimensional vibrational spectroscopy

Article abstract of DOI:10.1021/ic402254q

Catalytic transition-metal complexes often occur in several conformations that exchange rapidly (2H using two-dimensional vibrational spectroscopy, a method that can be applied to any catalyst provided that the exchange between its conformers occurs on a time scale of a few picoseconds or slower. We find that, in one of the conformations, the OC-Rh-CO angle deviates significantly from the canonical value in a trigonal-bipyramidal structure. On the basis of complementary density functional calculations, we ascribe this effect to attractive van der Waals interaction between the CO and the xantphos ligand.

Full text of DOI:10.1021/ic402254q

Article
pubs.acs.org/IC
Exchanging Conformations of a Hydroformylation Catalyst
Structurally Characterized Using Two-Dimensional Vibrational
Spectroscopy
Matthijs R. Panman, Jannie Vos, Vladica Bocokic, Rosalba Bellini, Bas de Bruin, Joost H. N. Reek,
́
and Sander Woutersen*
Van ’t Hoff Institute for Molecular Sciences, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, The Netherlands
S
* Supporting Information
ABSTRACT: Catalytic transition-metal complexes often occur in
several conformations that exchange rapidly (<ms) in solution so
that their spatial structures are difficult to characterize with
conventional methods. Here, we determine specific bond angles in
the two rapidly exchanging solution conformations of the hydro-
formylation catalyst (xantphos)Rh(CO)2H using two-dimensional
vibrational spectroscopy, a method that can be applied to any
catalyst provided that the exchange between its conformers occurs
on a time scale of a few picoseconds or slower. We find that, in one
of the conformations, the OC−Rh−CO angle deviates significantly
from the canonical value in a trigonal-bipyramidal structure. On the
basis of complementary density functional calculations, we ascribe
this effect to attractive van der Waals interaction between the CO
and the xantphos ligand.
excellent tool for studying catalytic reactions,2,8−17 but the
conformational information provided by a conventional IR
INTRODUCTION
■
Many catalytic transition-metal complexes exist in several
spectrum is rather limited. This limitation can be overcome by
conformations that exchange rapidly (on the ms time scale or
using two-dimensional infrared (2D-IR) spectroscopy experi-
faster) under ambient conditions. Generally, these conforma-
ments,18−21 the IR analog of 2D-NMR spectroscopy. In the
tions have very different catalytic activities.1 Hence, to obtain an
pump−probe version of these experiments, one specifically
understanding of how a fluxional transition-metal complex
excites a normal mode of a molecule (more precisely, one
catalyzes a reaction, one ideally would like to characterize all of
induces a transition between the v = 0 and v = 1 eigenstates of
its conformations in detail. It is, however, very difficult to
investigate such rapidly exchanging conformers with conven-
the molecule−bath system22), and by probing the response of
all other normal modes, observes the couplings between the
tional experimental methods. X-ray crystallography generally
excited and the probed modes as cross-peaks in a 2D plot, in a
provides access to only a single conformer. The standard
similar manner as in 2D-NMR (COSY). The 2D-IR cross-peak
method to probe conformations in solution is nuclear magnetic
intensities and polarization dependences provide direct
resonance (NMR). However, if the exchange between the
information about the relative distance and orientation of the
conformations is faster than the NMR time scale (determined
coupled vibrating bonds,18 information that generally cannot be
by the difference between the resonance frequencies of the
exchanging conformers, and typically 1 ms),2 the observed
obtained from the conventional (1D) IR spectrum. Because of
its picosecond time resolution, 2D-IR is a powerful method to
spectrum is an average of the spectra of the different
investigate the structure and dynamics of catalytic transition-
conformers, and the structural information about the individual
metal complexes in solution.7,19,21,23−29
conformations is lost.
In the infrared (IR) spectrum, the difference between the
resonance frequencies is much larger than that in the NMR
spectrum. The difference between the CO-stretching frequen-
Here, we use 2D-IR spectroscopy to study the catalyst
(xantphos)Rh(CO)2H (Figure 1, right), a prototypical member
of a large class of rhodium catalysts used for the hydro-
formylation of terminal alkenes.1,30,31 These catalysts occur in
cies of different conformers of a transition-metal complex is
typically several cm−1 (1011−1012 Hz), which corresponds to a
two conformations, one in which the two P atoms are
equatorial and the CO ligands are equatorial and axial (here
time scale of a few picoseconds. Hence, any exchange taking
place more slowly than this time scale is not averaged out in the
IR spectrum,3−7 and the exchanging conformers appear as well-
separated peaks or sets of peaks. IR spectroscopy is thus an
Received: September 4, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic402254q | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Right: Schematic representation of the two conformers of (xantphos)Rh(CO)2H. The phenyl groups on the P atoms and all H atoms,
except the one coordinated to Rh, have been omitted for clarity. (A) Conventional IR absorption spectrum of (xantphos)Rh(CO)2H. The labels "e"
and "a" refer to the equatorial and axial positions of the CO ligands. This spectrum is also shown above 2D spectrum C and to the right of 2D spectra
C and E. (B, C) 2D-IR spectra of (xantphos)Rh(CO)2H in hexane at 50 bar syngas pressure, for parallel (B) and perpendicular (C) polarizations of
the pump and probe pulses, recorded at a delay of 1.8 ps. Red colors denote ΔA > 0, blue colors ΔA > 0, green ΔA = 0. (D, E) Calculated 2D-IR
spectra obtained from a global least-squares fit, for parallel (D) and perpendicular (E) polarizations of the pump and probe pulses, using θee = 121°,
θea = 108°, and θaH = 170°, as obtained from a global least-squares fit to the data.
mid-infrared pulses with a duration of ∼100 fs at 2000 cm−1 and an
denoted as “ea”), and one in which one P atom is equatorial
energy and bandwidth of 10 μJ and ∼200 cm−1, respectively. Probe
and the other axial, and the CO ligands are both equatorial
and reference pulses are obtained from the mid-IR by reflection off the
(denoted as “ee”) (Figure 1, right). These conformers exchange
front and back surfaces of a wedged BaF2 window. Each beam is
on a 250 ns time scale (estimated by extrapolating low-
approximately 5% of the total intensity of the mid-IR radiation. The
temperature NMR data; see the Supporting Information), so in
remainder of the light is used as a pump beam. Both the probe and the
the IR spectrum, the two conformers are separately visible. We
find that, from the 2D-IR spectrum, we can determine specific
bond angles in both the exchanging conformers.
reference beams are focused through the sample by means of an f =
100 mm off-axis parabolic mirror. The probe is 200 μm in diameter at
the sample. The mid-IR reference beam passes through an area of the
sample not influenced by the pump (the purpose of the reference
beam is to correct for small pulse-to-pulse fluctuations in the infrared
intensity). The changes induced in the sample by the pump pulse are
monitored by the mid-IR probe pulses, at various time delays between
the pump and mid-IR probe pulse. The delay between pump and
probe pulses is controlled with a motorized delay stage. The beams are
recollimated with a second f = 100 mm off-axis parabolic mirror. The
probe and reference are focused into a spectrograph (MS260i,
Newport Oriel) by a third f = 100 mm off-axis parabolic mirror. The
spectrograph images both beams onto a 2 × 32 pixel HgCdTe array
detector (MCT, Infrared Associates). The pump pulses are optically
filtered using a Fabry−Perot interferometer. This results in spectrally
narrow pump pulses with a bandwidth of 10 cm−1 and an energy of
approximately 1 μJ. The center frequency of the light is varied by
adjusting the distance between the parallel mirrors of the Fabry−Perot
filter using a feedback-controlled piezoelectric mount. To provide the
feedback, we project the pump beam after it has passed the sample
onto the probe array of the MCT detector. We achieve this by
overlapping pump and probe on an adjustable CaF2 plate that then
couples the pump into the spectrograph. Since pump and probe now
run collinearly from the plate onward, a bipositional flag is placed such
EXPERIMENTAL SECTION
■
Sample Preparation. The preparation of the (xantphos)Rh-
(CO)2H sample is based on the procedure described by van der Veen
et al.32 All chemicals were purchased from Sigma Aldrich and were
used without further purification. We dissolved 3.87 mg (1.5 × 10−5
mol) of (acetylacetonato)dicarbonylrhodium(I) in 5 mL of degassed
cyclohexane under an inert atmosphere (Ar). A separate solution of
34.72 mg (6.0 × 10−5 mol) of 4,5-bis(diphenylphosphino)-9,9-
dimethylxanthene (xantphos) in 25 mL of degassed cyclohexane
under an inert atmosphere was prepared in a home-built in situ
infrared autoclave. The autoclave was heated to 60 °C and pressurized
to 50 bar with syngas under continuous stirring. The formation of the
(xantphos)Rh(CO)2H complex was followed with FTIR after the
high-presssure injection of the (acetylacetonato)dicarbonylrhodium(I)
solution into the xantphos mixture. A portion of the mixture was
transferred, under pressure, to a modified high-pressure IR cell (HPL-
C-13, Harrick Scientific). The sample was kept between two CaF2
windows separated by a 1 mm spacer.
Two-Dimensional Infrared Spectroscopy Setup. Using an
experimental setup that has been described previously,33 we generate
B
dx.doi.org/10.1021/ic402254q | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
cm−1 modes, and between the 1930, 1975, and 2040 cm−1
modes. Since coupling occurs only within a conformer, we can
conclude from the cross-peaks that the peak sets {1950, 1997
cm−1} and {1930, 1975, 2040 cm−1} each belong to one of the
two conformers. Since the ea conformer should have three
peaks, we also can assign the peak sets in the IR spectrum to
specific conformations: the {1930, 1975, 2040 cm−1} peaks
arise from the ea conformer, the {1950, 1997 cm−1 } peaks
from the ee conformer. The larger line widths of the ea
conformer are probably caused by the contribution of Rh−H
stretching in the (mixed CO + Rh−H stretch) normal
modes.
We can use the 2D-IR spectrum to characterize the structure
of both the conformers, in particular, to determine the OC−
Rh−CO and OC−Rh−H angles θee, θea, and θaH defined in
Figure 1. These angles determine the dependence of the cross-
peak intensities on the angle between the pump and probe
polarizations18 (the contributions of rotational diffusion,
fluxionality,19 and energy transfer18 to the cross-peak are
negligible at our pump−probe delay of 1.8 ps; see the
Supporting Information). To obtain the angles and the
coupling strengths from the 2D-IR spectrum, we globally fit
an excitonic model18 to the data (details are given in the
Supporting Information). It should be noted that, since the
strength of the coupling between the modes is of the same
order as the difference between their frequencies, the data
cannot be analyzed by reading off the cross-peak intensities and
anisotropies directly from the 2D-IR spectrum. If the localized
CO and Rh−H stretching modes are used as the basis set,
the Hamiltonians of the two conformers are given by
that it blocks the probe and reference during the optimization of the
Fabry−Perot etalon and the pump when data are acquired for the
measurement. The delay between pump and probe pulses is controlled
with a motorized delay stage. The pump is optically chopped at half
the laser repetition frequency to provide the pumped and unpumped
shots necessary for the generation of the difference absorption
spectrum. The pump pulses have an intensity envelope that is an
approximately single-sided exponential with a fwhm of 800 fs, as
determined from a cross-correlation measured using two-photon
absorption in InAs. The polarization of the IR pump pulse is set at 45°
with respect to that of the probe pulse using a MgF2 zero-order λ/2
plate. Subsequently, the polarization of the measured spectrum is
selected using a rotating polarizer situated directly after the sample set
at either 0° (parallel spectrum) or 90° (perpendicular spectrum) with
respect to the pump polarization.
RESULTS AND DISCUSSION
■
Figure 1A shows the IR spectrum of (xantphos)Rh(CO)2H in
hexane at 50 bar syngas pressure. The spectrum consists of four
intense peaks at 1950, 1975, 1997, and 2040 cm−1, and a
weaker shoulder at ∼1930 cm−1. We expect five peaks in total
for the two conformers: two CO-stretch modes (symmetric and
antisymmetric) for the ee conformer, and three CO/Rh−H
combinations for the ea conformer. The Rh−H mode
contributes to the spectrum only in the ea conformer, because
this mode gains sufficient absorption intensity to be observable
only if it mixes with the CO-stretch modes (the intrinsic
absorption of the Rh−H-stretch mode is much smaller than
that of the CO-stretch modes32), and the Rh−H/CO mixing is
much stronger in the ea conformer than in the ee conformer,
since in the former one of the CO bonds is parallel to the Rh−
H bond. From the conventional IR spectrum, it is only possible
to determine which of the five peaks belong to which
conformer by means of isotope labeling,34 but this question
can be solved directly by inspection of the 2D-IR spectrum,
shown in Figure 1B,C for parallel (B) and perpendicular (C)
polarizations of the pump and probe pulses. In this 2D graph,
the absorption change ΔA is plotted as a function of the pump
and probe frequencies. On the diagonal, we observe four
positive−negative couplets when νpump is resonant with one of
the four most intense absorption bands. The negative ΔA
feature is due to the v = 0 → 1 bleaching and v = 1 → 0
stimulated emission of the excited vibrational mode, and the
positive ΔA feature is due to v = 1 → 2 excited-state
absorption, which is decreased in frequency with respect to the
v = 0 → 1 frequency as a consequence of the anharmonicity of
the CO-stretch potential. The diagonal peak of the lowest-
frequency mode is too weak to be observable. Coupling
between vibrational modes causes cross-peaks in the 2D-IR
spectrum. Like the peaks on the diagonal, these cross-peaks are
also positive−negative doublets. This can be understood as
follows. When two modes A and B (with v = 0 → 1 frequencies
νA and νB) are coupled, the frequency of the state in which both
modes are in the v = 1 state is < νA + νB by an amount xAB (the
cross-anharmonicity), which is determined by the strength of
the interaction between modes A and B.18 As a consequence,
exciting mode A to the v = 1 state effectively gives rise to a
change of −xAB in the v = 0 → 1 frequency of mode B, leading
to a negative ΔA at νprobe = νB and a positive ΔA at νprobe = νB
− xAB. The frequency shift xAB is often smaller than the line
width, resulting in a cross-peak ΔA signal that (in the νprobe
direction) looks approximately like the derivative of the
absorption band. In the 2D-IR spectrum of (xantphos)Rh-
(CO)2H, cross-peaks are observed between the 1950 and 1997
⎛
⎜
⎜
⎞
⎟
⎟
εH βH,e βH,a
βH,e εe
⎛
⎞
εe
β
e,e
(1)
ee
(1)
⎜
⎜
⎟
⎟
⎠
β
H
=
and Hea
=
e,a
⎜
⎜
⎝
⎟
⎟
⎠
β
εe
⎝ e,e
βH,a
β
εa
e,a
where εi are the stretching frequencies of the Rh−H bond (i =
H) and of the axial and equatorial CO bonds (i = a, e), and βi,j
are the couplings between the stretching modes. The angles
between the CO bonds in each conformer (see Figure 1, right)
enter the calculation through the transition-dipole moment
matrix.18 These angles, the couplings, and site frequencies in
the Hamiltonians, together with the anharmonic shifts (the
difference between the v = 0 → 1 and v = 1 → 2 frequencies of
a local mode), completely determine the 2D-IR spectrum of
each conformer. The experimental 2D-IR spectrum is the sum
of the 2D-IR spectra of the ee and ea isomers weighted with
their relative abundances.
From the fit (shown in Figure 1D,E), we obtain θee = 121°
1°, θea = 108° 3°, and θaH = 170° 50°. The complete list of
parameters obtained from the fit is given in the Supporting
Information. Comparing the coupling strengths obtained from
the fit (see the Supporting Information) to the frequency
splittings in the IR spectrum, we find that, in both conformers,
the contribution of the couplings to the frequency splittings in
the IR spectrum is substantial. In the ee conformer, the local-
mode CO-stretch frequencies are identical because of
symmetry, and the splitting between the IR bands of this
conformer is caused purely by the coupling. Accordingly, the
coupling obtained from the 2D spectrum (22.8 0.1 cm−1) is
exactly half the frequency splitting in the IR spectrum. In the ea
conformer, the local-mode frequencies are all different, and the
observed differences in frequency of the three IR bands are
C
dx.doi.org/10.1021/ic402254q | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
caused partly by the coupling and partly by differences in the
local-mode frequencies. It is interesting to note that, in the ee
conformer, in which the local-mode frequencies are the same
because of symmetry, the angle between the CO ligands might,
in principle, still be determined from the conventional IR
spectrum, but this is not the case for the ea conformer, for
which the number of unknown parameters is too large to derive
the bond angles from the 1D-IR spectrum. They can be
obtained only from the 2D-IR spectrum.
We find that the OC−Rh−CO angle in the ee conformer is
equal to the canonical value of 120°, whereas, in the ea
conformer, the OC−Rh−CO angle deviates significantly from
the 90° expected for an ideal trigonal-bipyramidal structure. To
investigate the origin of this deviation, we performed density
functional calculations (see the Supporting Information), which
we have used previously to investigate similar Rh and Co
catalysts.35,36 We find that geometry optimizations (DFT,
BP86, def2-TZVP) without dispersion corrections (attractive
van der Waals forces) produce structures with OC−Rh−CO
angles close to 90°, deviating significantly from the measured
value of 108°. Optimizations including Grimme’s dispersion
corrections (disp3; DFT-D) lead to a structure (shown in
Figure 2) with a larger OC−Rh−CO angle (101°), which is
find that, in one conformer, there exists significant distortion
from the ideal trigonal-bipyramidal geometry, probably due to
Van der Waals attraction between the CO and xantphos
ligands. Such detailed structural information about the
conformations of fluxional complexes is difficult to obtain
from conventional spectroscopic methods (including 1D-IR
spectroscopy). Similar 2D-IR experiments can be performed to
determine the exchanging conformations of any fluxional
catalyst, provided that the exchange occurs slower than the
∼1 ps time scale corresponding to typical vibrational frequency
splittings, and that the bond-stretching modes are at
frequencies where subpicosecond IR pulses with sufficient
intensity can be generated and detected (∼1000 cm−1 and
higher). Determination of the coordination geometry of
transition-metal complexes in solution is important, as this
geometry to a large extent determines their reactivity.1 In
particular, it is well-documented that the bite angle of a
bidentate ligand is an important parameter,37 since ligands with
large bite angles lead to relatively unstable square-planar
complexes and as such are more reactive than analogues with
small bite angles. For hydroformylation, it has been
demonstrated that large bite-angle ligands give high selectivity
for the linear product. A direct correlation between the relative
stability of the resting state and its reactivity in palladium-
catalyzed allylic substitution reactions has been reported and
used as a basis for a screening protocol.38 As yet, the crucial
information on coordination geometry in catalytic transition-
metal complexes is mainly extracted from X-ray analysis of such
compounds in the solid state, where packing effects can also
play a role. With 2D-IR spectroscopy, we now can obtain such
information also for complexes in solution.
ASSOCIATED CONTENT
■
S
* Supporting Information
Exchange rate of the complex determined from NMR data,
delay dependence of the IR pump−probe signals, structural
parameters obtained from the global least-squares fit to the 2D-
IR spectra, DFT-calculated structures, and corresponding
structural parameters. This material is available free of charge
via the Internet at http://pubs.acs.org.
Figure 2. Calculated structure of the ea conformer of (xantphos)Rh-
(CO)2H. All H atoms, except the one coordinated to Rh, have been
omitted for clarity.
AUTHOR INFORMATION
■
Corresponding Author
*Fax: +31 20 525 6965. Tel: +31 525 7091. E-mail: s.
woutersen@uva.nl.
clearly in better agreement with the 2D-IR measurements.
Hence, the large θea probably arises from van der Waals
attraction between the CO ligands and the xantphos ligand.
One might think that the observed widening of the OC−Rh−
CO angle could be purely the result of steric repulsion, leading
to tilting of the axial CO ligand toward the hydride and away
from the second CO ligand. However, such behavior was not
reproduced with uncorrected DFT methods, which include
repulsive steric interactions but lack attractive dispersion forces.
Only upon including Grimme’s dispersion corrections (disp3)
was a calculated OC−Rh−CO angle of 101° obtained, showing
that attractive van der Waals forces cause the deviation of the
OC−Rh−CO angle from 90°.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Stichting voor Fundamenteel
Onderzoek der Materie (FOM) and the Nederlandse
Organisatie voor Wetenschappelijk Onderzoek (NWO).
REFERENCES
■
(1) van Leeuwen, P. W. N. M. Homogeneous Catalysis: Understanding
the Art; Springer: New York, 2004.
(2) Heaton, B., Ed. Mechanisms in Homogeneous Catalysis: A
Spectroscopic Approach; Wiley-VCH: Weinheim, Germany, 2005.
(3) Harris, C. B.; Shelby, R. M.; Cornelius, P. A. Chem. Phys. Lett.
1978, 57, 8−14.
(4) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Washington,
J.; Kubiak, C. P. Science 1997, 277, 660−663.
CONCLUSION
■
To conclude, we have used 2D vibrational spectroscopy to
unravel the IR spectrum of a hydroformylation catalyst that
occurs in two rapidly exchanging conformations, and to
characterize the spatial structure of these conformations. We
D
dx.doi.org/10.1021/ic402254q | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(5) Grevels, F.-W.; Kerpen, K.; Klotzbucher, W. E.; McClung, R. E.
D.; Russell, G.; Viotte, M.; Schaffner, K. J. Am. Chem. Soc. 1998, 120,
10423−10433.
(6) Ito, T.; Imai, N.; Yamaguchi, T.; Hamaguchi, T.; Londergan, C.
H.; Kubiak, C. P. Angew. Chem. 2004, 116, 1400−1405.
(7) Butler, J. M.; George, M. W.; Schoonover, J. R.; Dattelbaum, D.
M.; Meyer, T. J. Coord. Chem. Rev. 2007, 251, 492−514.
(8) To, T. T.; Heilweil, E. J.; Burkey, T. J. J. Phys. Chem. A 2006, 110,
10669−10673.
(9) To, T. T.; Heilweil, E. J. J. Phys. Chem. A 2007, 111, 8047−8049.
(10) Sawyer, K. R.; Glascoe, E. A.; Cahoon, J. F.; Schlegel, J. P.;
Harris, C. B. Organometallics 2008, 27, 4370−4379.
(11) Easun, T. L.; Alsindi, W. Z.; Deppermanna, N.; Towrie, M.;
Ronayne, K. L.; Sun, X. Z.; Ward, M. D.; George, M. W. Inorg. Chem.
2009, 48, 8759−8770.
the conformers are independent of the (xantphos)Rh(CO)2H
concentration.
̈
(35) Patureau, F. W.; de Boer, S.; Kuil, M.; Meeuwissen, J.; Breuil, P.-
A. R.; Siegler, M. A.; Spek, A. L.; Sandee, A. J.; de Bruin, B.; Reek, J. N.
H. J. Am. Chem. Soc. 2009, 131, 6683−6685.
(36) Suarez, A. I. O.; Jiang, H.; Zhang, X. P.; de Bruin, B. Dalton
Trans. 2011, 40, 5697−5705.
(37) Birkholz, M.-N.; Freixa, Z.; van Leeuwen, P. W. N. M. Chem.
Soc. Rev. 2009, 38, 1099−1118.
(38) Wassenaar, J.; Jansen, E.; van Zeist, W.-J.; Bickelhaupt, F. M.;
Siegler, M. A.; Spek, A. L.; Reek, J. N. H. Nat. Chem. 2010, 2, 417.
(12) Calladine, J. A.; Torres, O.; Anstey, M.; Ball, G. E.; Bergman, R.
G.; Curley, J.; Duckett, S. B.; George, M. W.; Lawes, D. J.; Perutz, R.
N.; Sun, X.-Z.; Vollhardt, K. P. C. Chem. Sci. 2010, 1, 622−630.
(13) Kubis, C.; Ludwig, R.; Sawall, M.; Neymeyr, K.; Borner, A.;
̈
Wiese, K.-D.; Hess, D.; Franke, R.; Selent, D. ChemCatChem 2010, 2,
287−295.
(14) Schneider, J.; Vuong, K. Q.; Calladine, J. A.; Sun, X. Z.;
Whitwood, A. C.; George, M. W.; Perutz, R. N. Inorg. Chem. 2011, 50,
11877−11889.
(15) Lomont, J. P.; Nguyen, S. C.; Zoerb, M. C.; Hill, A.; Schlegel, J.
P.; Harris, C. Organometallics 2012, 31, 3582−3587.
(16) Diebolt, O.; van Leeuwen, P. W. N. M.; Kamer, P. C. J. ACS
Catal. 2012, 2, 2357−2370.
(17) Lomont, J. P.; Nguyen, S. C.; Harris, C. B. J. Am. Chem. Soc.
2013, 135, 11266−11273.
(18) Hamm, P.; Zanni, M. Concepts and Methods of 2D Infrared
Spectroscopy; Cambridge University Press: Cambridge, U.K., 2011.
(19) Cahoon, J. F.; Sawyer, K. R.; Schlegel, J. P.; Harris, C. B. Science
2008, 319, 1820−1823.
(20) Hunt, N. T. Chem. Soc. Rev. 2009, 38, 1837−1848.
(21) Baiz, C. R.; McRobbie, P. L.; Anna, J. M.; Geta, E.; Kubarych, K.
J. Acc. Chem. Res. 2009, 42, 1395−1404.
(22) Mukamel, S. Principles of Nonlinear Optical Spectroscopy; Oxford
University Press: Oxford, U.K., 1995.
(23) Golonzka, O.; Khalil, M.; Demirdoven, N.; Tokmakoff, A. J.
̈
Chem. Phys. 2001, 115, 10814−10828.
(24) Bredenbeck, J.; Helbing, J.; Hamm, P. J. Am. Chem. Soc. 2004,
126, 990−991.
(25) Baiz, C. R.; McCanne, R.; Kubarych, K. J. Appl. Spectrosc. 2010,
64, 1037−1044.
(26) Stewart, A. I.; Wright, J. A.; Greetham, G. M.; Kaziannis, S.;
Santabarbara, S.; Towrie, M.; Parker, A. W.; Pickett, C. J.; Hunt, N. T.
Inorg. Chem. 2010, 49, 9563−9573.
(27) Anna, J. M.; King, J. T.; Kubarych, K. J. Inorg. Chem. 2011, 50,
9273−9283.
(28) Bingaman, J. L.; Kohnhorst, C. L.; Avan Meter, G. A.; McElroy,
B. A.; Rakowski, E. A.; Caplins, B. W.; Gutowski, T. A.; Stromberg, C.
J.; Webster, C. E.; Heilweil, E. J. J. Phys. Chem. A 2012, 116, 7261−
7271.
(29) Messmer, A. T.; Lippert, K. M.; Steinwand, S.; Lerch, E. W.;
Hof, K.; Ley, D.; Gerbig, D.; Hausmann, H.; Schreiner, P. R.;
Bredenbeck, J. Chem.Eur. J. 2012, 18, 14989−14995.
(30) Webb, P. B.; Kunene, T. E.; Cole-Hamilton, D. J. Green Chem.
2005, 7, 373−379.
(31) Franke, R.; Selent, D.; Borner, A. Chem. Rev. 2012, 112, 5675−
̈
5732.
(32) van der Veen, L. A.; Boele, M. D. K.; Bregman, F. R.; Kamer, P.
C. J.; van Leeuwen, P. W. N. M.; Goubitz, K.; Fraanje, J.; Schenk, H.;
Bo, C. J. Am. Chem. Soc. 1998, 120, 11616−11626.
(33) Huerta-Viga, A.; Shaw, D. J.; Woutersen, S. J. Phys. Chem. B
2010, 114, 15212−15220.
(34) Singular-value decomposition of the concentration dependence
of the spectrum would not work, since the relative concentrations of
E
dx.doi.org/10.1021/ic402254q | Inorg. Chem. XXXX, XXX, XXX−XXX