PROBING VIBRATIONAL RELAXATION DYNAMICS IN CHARGE-TRANSFER
EXCITED STATES: SYNTHESIS, PHYSICAL, AND PHOTOPHYSICAL
CHARACTERIZATION OF CYANO-SUBSTITUTED POLYPYRIDYL
COMPLEXES OF RU(II)
By
Catherine Emily McCusker

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
CHEMISTRY
2010

ABSTRACT
PROBING VIBRATIONAL RELAXATION DYNAMICS IN CHARGE-TRANSFER
EXCITED STATES: SYNTHESIS, PHYSICAL, AND PHOTOPHYSICAL
CHARACTERIZATION OF CYANO-SUBSTITUTED POLYPYRIDYL
COMPLEXES OF RU(II)

By

Catherine Emily McCusker

Photo-induced charge separation is the physical phenomenon underlying
virtually all schemes geared toward the conversion of light into chemical,
electrical, and/or mechanical energy. Charge separation is typically effected in a
molecular system through charge-transfer excited states, in which photon
absorption causes charge redistribution within the chromophore: maintaining,
amplifying, or, in the least favorable circumstances, destroying the resulting
chemical potential depends on dynamics that occur within the chromophore
immediately following the absorptive event.

In transition metal complexes a

majority of excited state energy is dissipated through non-radiative decay.
Despite the prominent role it plays in the deactivation of charge-transfer excited
states, there is relatively known about the mechanism of non-radiative decay in
charge transfer complexes.
Probing vibrational relaxation in transition metal complexes is not always a
straightforward process.

Often times information about vibrational relaxation

from the initially excited Franck-Condon state to the long lived excited state is
inferred from transient electronic absorption spectroscopy. Infrared spectroscopy
is more direct way to probe the vibrational state of an electronic excited state.

This dissertation investigates the fundamental photophysics of ruthenium
polypyridyl complexes, in particular the non-radiative decay between the initial
excited Franck-Condon state and the long lived excited state. The complexes
studied in this dissertation incorporate cyanide groups as infrared tags in order to
use infrared transient absorption spectroscopy, coupled with visible transient
absorption spectroscopy, to probe the vibrational relaxation dynamics in
ruthenium polypyridyl complexes.
Before ultrafast dynamics can be interpreted it is critical to have a solid
understanding of the properties of both the initial Franck-Condon excited state
and the long lived excited state.

Nanosecond time-resolved spectroscopic

techniques which can be used to probe the long lived excited state of transition
metal complexes are discussed. These techniques are used to characterize the
long lived excited states of a series of cyano-substituted ruthenium(II) bipyridine
and terpyridine complexes.

A combination of ultrafast infrared and visible

absorption spectroscopies are used to probe the excited state dynamics in the
series of cyano-substituted ruthenium(II) bipyridine complexes. By combining the
two techniques it can be conclusively shown that small amplitude changes in the
time resolved electronic absorption spectra on a ~1-10 ps time scale are due to
vibrational relaxation on the lowest energy excited state potential surface. The
large energy difference between the cyano substituted bipyridine ligand and the
unsubstituted bipyridine ligand allows for selectively localizing the initial excited
state on either the cyano-substituted or unsubstituted bipyridine ligands. This
potentially allows the dynamics associated with intramolecular vibrational
redistribution to be decoupled from the dynamics associated with interligand
electron transfer.

Copyright by
Catherine Emily McCusker
2010

ACKNOWLEDGMENTS

I first have to thank the people who have been invaluable in collecting the
data contained in this dissertation. I was fortunate to be able to collaborate with
Professor Tony Vlček for the ultrafast time-resolved infrared absorption
spectroscopy.

Dr. Heinz Frei has been incredibly helpful with step-scan IR,

helping us design our experiment here at MSU, inviting me to his lab to learn the
ins and outs of step-scan IR and collect my first set of data, and helping to
troubleshoot the many instrumental problems I've had over the years. Allison
Brown collected all of the ultrafast time-resolved visible absorption spectroscopy
presented here.

Rick Fehir and Drew Kouzelos have been very patient in

answering all of my calculation questions over the years.
Next I want to thank Jim and the McCusker group. Jim, thank you for taking
me into your group and always being supportive of my interest in teaching.
You've always gone out of your way to help me build my teaching resume and
confidence and I hope to put it to good use. Thanks to all of the McCusker group
members, past and present, especially Lisa, Lindsey, Alli, Drew, Rick and Dong.
You guys have made my time in grad school a once-in-a-lifetime experience and
are one of the (very) few things I'll miss about living in Michigan.
Last, but certainly not least, I have to thank my family. Mom and Dad, you
have always been supportive and encouraging of the scientist in me, even when
it meant collecting bugs and snakes, "rescuing" baby birds and squirrels, and
making homemade sparklers with my chemistry set. I owe the biggest thanks to
my amazingly supportive (and incredibly patient) husband Jason. You've kept
me sane thorough all the ups and downs of grad school and I couldn't have done
this without you.

v

TABLE OF CONTENTS
List of Tables

................................................................................................... x

List of Figures

....................................................................................................xi

Key to Symbols and Abbreviations .....................................................................xxv
Chapter 1:

Introduction to Photophysical Properties of Ruthenium Polypyridyl
Complexes .......................................................................................1
1.1 Introduction..............................................................................1
1.2 Ground State Properties of [Ru(bpy)3]2+ ...................................1
1.3 Long Lived Excited State of [Ru(bpy)3]2+ ..................................5
1.4 Ultrafast Dynamics of [Ru(bpy)3]2+ ............................................7
1.4.1 Intersystem Crossing.......................................................7
1.4.2 Vibrational Relaxation .....................................................8
1.4.3 Localization or Delocalization of the Excited State........10
1.5 Applications of Ruthenium Polypyridyl Complexes ..................12
1.6 Dissertation Outline ..................................................................13
1.7 References ...............................................................................16

Chapter 2:

The Application of Nanosecond Time-Resolved Spectroscopies to
the Study of Inorganic Photophysics and Photochemistry .............20
2.1 Introduction ..............................................................................20
2.2 Time-Resolved Emission Spectroscopy...................................20
2.2.1 Experimental Setup .......................................................20
2.2.2 Examples.......................................................................24
2.3 Time-Resolved Absorption Spectroscopy ................................28
2.3.1 Experimental Setup .......................................................28
2.3.2 Analysis and Interpretation of Time-Resolved ..............29
Absorption Spectra
2.3.3 Examples.......................................................................31
2.4 Time-Resolved Vibrational Spectroscopy ................................35
2.4.1 Experimental Setup .......................................................36
2.4.2 Examples.......................................................................42
2.4.2.1 Time-Resolved Resonance Raman ...................42
2.4.2.2 Step-Scan IR ......................................................44
2.5 Time-Resolved X-ray Spectroscopy.........................................46
2.5.1 Experimental Setup .......................................................46
2.5.2 Examples.......................................................................49
2.6 Concluding Comments .............................................................50
vi

2.7 References ...............................................................................52
Chapter 3:

Synthesis and Spectroscopic Characterization of CyanoSubstituted Bipyridyl Complexes of Ruthenium(II) .........................55
3.1 Introduction ..............................................................................55
3.2 Experimental Section ...............................................................58
3.2.1 General..........................................................................58
3.2.2 Ligand Synthesis ...........................................................58
3.2.3 Synthesis of [Ru(bpy)2(CN-Me-bpy)](PF6)2, (1) ............63
3.2.4 Synthesis of [Ru(bpy)(CN-Me-bpy)2](PF6)2, (2) ............66
3.2.5 Synthesis of [Ru(CN-Me-bpy)3](PF6)2, (3) ....................68
3.2.6 Physical Measurements ................................................69
3.3 Results and Discussion ............................................................74
3.3.1 Synthesis .......................................................................75
3.3.2 Ground State Spectroscopic Properties ........................79
3.3.3 Steady-State and Time-Resolved Emission ..................81
Spectroscopies
3.3.4 Time-Resolved Electronic Absorption Spectroscopy ....86
3.3.5 Nanosecond Step-Scan Time-Resolved Infrared ..........91
Spectroscopy
3.4 Concluding Comments .............................................................96
3.5 References ...............................................................................99

Chapter 4:

Probing Vibrational Relaxation Dynamics With Ultrafast Electronic
and Infrared Absorption Spectroscopy .........................................104
4.1 Introduction ............................................................................104
4.2 Experimental ..........................................................................108
4.2.1 Calculations .................................................................108
4.2.2 Time-Resolved Spectroscopic Measurements ............108
4.3 Results and Discussion ..........................................................109
4.3.1 Time Dependent DFT .................................................109
4.3.2 Time-Resolved Infrared Spectroscopy ........................114
4.3.3 Time-Resolved Electronic Absorption Spectroscopy ..119
4.3.3.1 Low Energy Excitation ......................................120
4.3.2.2 High Energy Excitation .....................................129
4.4 Future Works..........................................................................134
4.4.1 Solvent Effects ............................................................134
4.4.2 ILET and Localization ..................................................137
4.5 Concluding Comments ..........................................................139
4.6 References .............................................................................142

vii

Chapter 5:

Synthesis and Spectroscopic Characterization of CyanoSubstituted Terpyridyl Complexes of Ruthenium(II).....................145
5.1 Introduction ............................................................................145
5.2 Experimental ..........................................................................149
5.2.1 General........................................................................149
5.2.2 Synthesis of [Ru(tpy)(CN-tpy)](PF6)2 ..........................150
5.2.3 Synthesis of [Ru(CN-tpy)2](PF6)2................................152
5.2.4 Physical Measurements ..............................................153
5.3 Results and Discussion ..........................................................156
5.3.1 Synthesis ....................................................................157
5.3.2 Ground State Spectroscopic Properties ......................157
5.3.3 Steady-State and Time-Resolved Emission ................164
Spectroscopies
5.3.4 Time-Resolved Electronic Absorption Spectroscopy ..170
5.4 Concluding Comments and Future Directions .......................173
5.6 References .............................................................................169

Appendix A: Additional Figures for Chapter 3 ...................................................180
Appendix B: Additional Figures for Chapter 5 ....................................................191
Appendix C: Instructions for the Collection Step Scan IR Data on the Bruker
IFS66/S .....................................................................................197
C.1 Instrument Warm Up Procedure...........................................197
C.2 Collection of the Ground State Spectrum .............................197
C.3 Collection of the DC Coupled (Ground State) Step-Scan ....199
Spectra
C.4 Collection of the AC Coupled (Excited State) Step Scan .....200
Spectra
C.5 Data Analysis .......................................................................202
C.6 Instrument Shut Down Procedure ........................................205
C.7 Example Advanced Measurement Settings .........................205
C.8 Example Step Scan Time Resolved Measurements ...........207
Settings

viii

Appendix D: Cartesian Coordinates for Optimized Geometries of CN-Me-bpy
Ruthenium(II) Complexes ............................................................209
D.1 [Ru(bpy)2(CN-Me-bpy)](PF6)2 Optimized Ground State ......209
Geometry
D.2 [Ru(bpy)2(CN-Me-bpy)](PF6)2 Optimized Lowest Energy ....211
Triplet State Geometry
D.3 [Ru(bpy)(CN-Me-bpy)2](PF6)2 Optimized Ground State ......213
Geometry
D.4 [Ru(CN-Me-bpy)3](PF6)2 Optimized Ground State ..............215
Geometry

ix

LIST OF TABLES
Table 3-1: Electrochemical Data for Complexes 1-3 and [Ru(bpy)3](PF6)2 ........80
Table 3-2: Photophysical Data for Complexes 1-3 and [Ru(bpy)3](PF6)2 ...........83
Table 3-3: Spectral Fitting Results for Complexes 1-3 and [Ru(bpy)3](PF6)2 .....84
Table 3-4: Step-Scan IR and DFT Results for Complexes 1-3 ...........................92
Table 5-1: Electrochemical Data for Complexes 1-2 and [Ru(tpy)2](PF6)2 .......160
Table 5-2: Photophysical Data for Complexes 1 and 2 .................................... 166
Table 5-3: Spectral Fitting Results for Terpyridine Complexes .........................167

x

LIST OF FIGURES
Figure 1-1: Simplified molecular orbital diagram showing the interactions
between a π acceptor ligand and metal d orbitals in Oh symmetry.
For a low spin d6 metal center the possible transitions in the
electronic absorption spectrum are shown by the arrows. MLCT
(blue arrow) is a metal to ligand charge transfer transition from a
filled metal orbital to an empty π* ligand orbital. IL (green arrow) is
an intraligand transition from a filled ligand π orbital to an empty π*
ligand. LF (red arrow) is a ligand field transition from a filled metal
d orbital to an empty metal d orbital. For interpretation of the
references to color in this and all other figures, the reader is
referred to the electronic version of this dissertation........................2
Figure 1-2: Electronic absorption spectrum of [Ru(bpy)3](PF6)2 in CH3CN
solution. MLCT is the metal-to-ligand charge transfer absorption
band, LF is the ligand field absorption bands, and IL is the
intraligand π → π* absorption band. .................................................3
Figure 1-3: Cyclic voltammogram of [Ru(bpy)3](PF6)2 in CH3CN with 0.1 M
TBAH as the electrolyte. The spectrum was collected using the
standard three electrode setup with a platinum working electrode,
graphite counter electrode, and a Ag/AgCl reference electrode.
The spectrum is referenced to the ferrocene/ferrocenium redox
couple, which was added as an internal standard. The differential
pulse voltammograms are shown in the insets. ...............................4
Figure 1-4: Potential energy diagram for [Ru(bpy)3]2+. The 1A1 ground state
can be excited with visible light into a 1MLCT state. This initial
1MLCT state rapidly intersystem crosses to form the 3MLCT state,
which is the long lived excited stae. The 3MLCT state can relax
back to the ground state through radiative and non-radiative
decay.
.........................................................................................6
Figure 1-5: Three state model for intersystem crossing in [Ru(bpy)3]2+
developed by Mathies and coworkers. Reproduced from reference
23.
.........................................................................................9

xi

Figure 1-6: Structure
of
the
ruthenium
blue
dimer,
cis,cis[(bpy)2(H2O)RuIIIORuIII(OH2)-(bpy)2]4+.
Reproduced
from
reference 49. ..................................................................................13
Figure 2-1: Example optical layout of a time-resolved emission spectrometer.
A pulsed Nd:YAG laser excites the sample and a fraction of the
emission is collected and focused into a monochromator. The
emission decay is monitored in real time at a single wavelength. A
photodiode trigger synchronized data collection with sample
excitation. ......................................................................................21
Figure 2-2: Plot of the observed value of In(knr) vs. Eem, the emission
maximum in acetonitrile solution. Complexes have the general
formula [Os(bpy')2(L)]2+, where L is one bidentate or two
monodentate ligands and bpy' is bpy (●), bpy-d8 (◦), or phen (▴).
The drawn lines indicate the separate linear correlations for the bpy
and phen series, both with slopes of (1,100 cm-1). Reproduced
from reference 3. ............................................................................23
Figure 2-3: Plot of emission lifetime vs temperature for [Ru(bpy)3](SO4)2 doped
in [Zn(bpy)3](SO4)2 (1:1000). The solid line the result of fitting the
data to a modified Boltzmann type model. The energy level
diagram to the right shows the energy separations and intrinsic
emission lifetimes resulting from the fit.
Reproduced from
reference 5. ....................................................................................25
Figure 2-4: Time-resolved
emission
data
for
[Fe(pyacac)3
(Re(deeb)(CO)3)3](OTf)3 (deeb = 4,4'-diethylester-2,2'-bipyridine).
The emission was monitored at λprobe= 624 nm following excitation
at λpump = 430 nm. The solid red line corresponds to a fit to a
single-exponential decay model with τobs = 2.5 ± 0.1 ns. The inset
shows time-resolved emission data for the corresponding model
complex, [Al(pyacac)3(Re(deeb)(CO)3)3](OTf)3 with τobs= 235 ± 20
ns. Reproduced from reference 2. ................................................26

xii

Figure 2-5: An example optical layout of a transient absorption spectrometer.
A pulsed Nd:YAG laser excites the sample and a pulsed Xe arc
lamp probes the sample. Shutters and a digital delay generator
control the timing of the laser and white light pulses. The intensity
of the probe beam is measured at a single wavelength before and
after the sample is excited. A photodiode trigger synchronizes data
collection with sample excitation. ...................................................27
Figure 2-6: Cartoon showing a theoretical ground state absorption (GS) and
excited state absorption (ES) (left graph). The transient absorption
experiment measures the difference in the extinction coefficients of
these two states (right graph), leading to the possibility of both
positive and negative features in the transient spectrum.
Reproduced from reference 11. .....................................................30
Figure 2-7: Reaction pathways available for the photoexcited Fisher carbene
complex (CO)5W=C(OMe)Ph as determined by time resolved
absorption spectroscopy. The excited state of the carbene can
decay nonradiatively back to the ground state, loose a CO ligand to
form the tetracarbonyl complex, or isomerize from syn (CZ) to anti
(CE). Reproduced from reference 14. ..........................................32
Figure 2-8: Plot of the excited-state differential absorbance of a CH3CN
solution containing [Ru(dmb)3]2+ and [Fe2(OH)(O2CCH3)2(Tp)2]+ at
440 nm following 10 ns excitation at 450 nm. The solid line
indicates a fit to a biexponential kinetic model. Reproduced from
reference 16. ..................................................................................34
Figure 2-9: Example optical layout for two-color time-resolved resonance
Raman spectrometer.
There are two pulsed lasers to
independently pump and probe the sample. A pulse delay
generator controls the time between the pump and probe pulses.
Raman scatter is collected at 90°, dispersed with a spectrograph
and the full spectrum is collected with a CCD array. Reproduced
from reference 19. ..........................................................................37
Figure 2-10: Example optical layout of a time-resolved step-scan FTIR
spectrometer. Reproduced from reference 22. ..............................38

xiii

Figure 2-11: Data collection process in SSIR experiment. Decay traces (I vs t)
are collected at each fixed mirror position δ. After data collection,
interferograms (I vs δ) are generated for every time t, and these
interferograms are used for further data analysis. Reprinted from
reference 23. ..................................................................................38
Figure 2-12: Example of data processing in the SSIR experiment. A single two
dimensional slice of the three dimensional spectra are shown for
simplicity. .......................................................................................40
Figure 2-13: Nanosecond transient resonance Raman spectra of CuIITSPP in
dioxane/ water solvent mixture. Percentages show the increasing
fraction of dioxane in the solvent mixture. Reproduced from
reference 26. ..................................................................................43
Figure 2-14: TRIR spectra of fac-[Re(dppz)(CO)3(py-PTZ)]+* (pictured above)
separated into early and late time average spectra. At early times
(<300 ns, dashed line), the charge separated state dominates. At
later times (>300 ns, solid line), the IL state dominates and the
ground-to-excited-state shifts in ν(CO) decrease. Reproduced from
reference 29. ..................................................................................45
Figure 2-15: Schematic diagram of a time-resolved X-ray diffraction experiment
at the Advanced Photon Source, Argonne National Lab.
Reproduced from reference 30. .....................................................47
Figure 2-16: ES geometries (orange) of [Cu(dmp)(dppe)]+ superimposed on the
GS of the complex (Cu, green; C, black; P, purple; N, blue). The
different views are shown to illustrate the change in rocking
distortion (left) and the displacement of the phenanthroline ligand
from its GS plane (right) upon excitation. Reproduced from
reference 35. ..................................................................................50
Figure 3-1: Synthesis of 4,4'-dichloro-2,2'-bipyridine. (i) Acetic acid and 30%
hydrogen peroxide, refluxed. (ii) Sulfuric acid and fuming nitric
acid, gentle heating. (iii) Acetyl anhydride refluxed in acetic acid
(iv) PCl3 refluxed in acetonitrile ......................................................59

xiv

Figure 3-2: Synthesis of complexes 1-3. (i) bpy or Cl-Me-bpy and LiCl refluxed
in DMF. (ii) bpy or Cl-Me-bpy refluxed in ethanol for 48 hrs. (iii)
Pd2(dba)3, dppf, ZnCN, and zinc dust heated in DMA. See text for
details. .......................................................................................62
Figure 3-3: Electronic absorption spectra of the conversion of [Ru(bpy)2(Cl-Mebpy)]2+ to [Ru(bpy)2(CN-Me-bpy)]2+. The inset indicates the time
following initiation of the reaction that each aliquot was taken. See
text for further details. ....................................................................76
Figure 3-4: Electronic absorption spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1,
blue line), [Ru(bpy)(CN-Me-bpy)2](PF6)2 (2, green line), [Ru(CNMe-bpy)3](PF6)2 (3, red line), and [Ru(bpy)3](PF6)2 (black line) in
CH3CN solutions. ..........................................................................78
Figure 3-5: Cyclic voltammogram of complex 1 in CH3CN with 0.1 M TBAH as
the electrolyte. Spectrum was collected using the standard three
electrode setup with a platinum working electrode, graphite counter
electrode, and a Ag/AgCl reference electrode and is referenced to
the ferrocene/ferrocenium redox couple, which was added as an
internal standard. The differential pulse voltammogram is shown in
the insets. .......................................................................................80
Figure 3-6: Steady-state emission spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1,
blue squares), [Ru(bpy)(CN-Me-bpy)2](PF6)2 (2, green triangles),
and [Ru(CN-Me-bpy)3](PF6)2 (3, red circles). (a) Roomtemperature spectra acquired in deoxygenated CH3CN solution.
The solid lines correspond to fits to an asymmetric double
sigmoidal function – see text for further details. (b) Emission
spectra acquired in a 9:2 butyronitrile/propionitrile glass
at 80 K. .......................................................................................82
Figure 3-7: Nanosecond time-resolved differential absorption spectra acquired
in room-temperature CH3CN solution for (a) [Ru(bpy)2(CN-Mebpy)](PF6)2 (1), (b) [Ru(bpy)(CN-Me-bpy)2](PF6)2 (2), (c) [Ru (CNMe-bpy)3](PF6)2 (3), and (d) [Ru(bpy)3](PF6)2. The individual points
correspond to the amplitudes of fits of the kinetics data to singleexponential decay models; a smoothed solid line has been included
in each plot to guide the eye. .........................................................87

xv

Figure 3-8: (a) Ground state absorption spectrum for [Ru(bpy)2(CN-Mebpy)](PF6)2 (1) in room-temperature CH3CN solution. (b) Oxidative
difference spectra acquired at an applied potential of +1.55 V
versus Ag/AgCl. The inset corresponds to an expanded view of the
low-energy portion of the spectrum. (c) Reductive difference
spectra acquired at an applied potential of -1.05 V versus Ag/AgCl.
(d) Time-resolved differential absorption spectrum of compound 1
following ca. 10 ns excitation at 500 nm. ......................................89
Figure 3-9: Steady-state and time-resolved infrared absorption data for
[Ru(bpy)2(CN-Me-bpy)](PF6)2 (1) in room-temperature CH3NO2
solution. (a) Comparison of the ground-state (blue line) and stepscan infrared differential absorption data acquired at a time delay of
150 ns following ca. 10 ns excitation at 540 nm (red line). The ~40
cm-1 red-shift in the CN stretching frequency reflects the presence
of an electron in the π* orbital of the CN-Me-bpy ligand in the
3MLCT excited state. (b) Nanosecond step-scan infra-red spectra
as a function of time following ca. 10 ns excitation at 540 nm. The
kinetics describing the decrease in amplitude of the excited-state
infrared absorption signal are within experimental error of the timeresolved emission and absorption data for compound 1 and are
therefore assigned to relaxation of the 3MLCT excited state. ........93
Figure 3-10: (a) Highest energy π bonding orbital on CN-Me-bpy in the geometry
optimized ground state of complex 1. (b) Lowest energy π*
antibonding orbital on CN-Me-bpy in the geometry optimized lowest
energy triplet state of complex 1. ...................................................95
Figure 4-1: Potential energy surfaces illustrating the available Franck-Condon
transitions between the optically prepared excited state (ES1) and a
higher lying excited state (ES2) when (a) ES1 is in a excited
vibrational (hot) state and (b) ES1 is in a thermalized vibrational
(cooled) state vibrational (hot) state and (b) ES1 is in a thermalized
vibrational (cooled) state ..............................................................105
Figure 4-2: The series of cyano-substituted ruthenium(II) complexes
investigated in this chapter...........................................................107

xvi

Figure 4-3:

TD-DFT results for complex 1 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory. Acetonitrile
solvent environment was modeled with the polarizable continuum
model (PCM). The red triangles are calculated triplet transitions
the blue squares are calculated singlet transitions, and the black
line is the experimental electronic absorption spectrum in CH3CN.
Pictured orbitals are the major contributions to the selected
transitions. ...................................................................................111

Figure 4-4:

TD-DFT results for complex 2 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory. Acetonitrile
solvent environment was modeled with polarizable continuum
model (PCM). The red triangles are calculated triplet transitions
the blue squares are calculated singlet transitions, and the black
line is the experimental electronic absorption spectrum in CH3CN.
Pictured orbitals are the major contributions to the selected
transitions. ....................................................................................112

Figure 4-5:

TD-DFT results for complex 3 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory. Acetonitrile
solvent environment was modeled with polarizable continuum
model (PCM). The red triangles are calculated triplet transitions
the blue squares are calculated singlet transitions, and the black
line is the experimental electronic absorption spectrum in CH3CN.
Pictured orbitals are the major contributions to the selected
transitions. ....................................................................................113

Figure 4-6: Time resolved infrared spectra collected for complex 1 in a
nitromethane solution. (a) 490 nm excitation wavelength where the
initial excited state is localized on the CN-Me-bpy ligand (b) 400 nm
excitation wavelength where the initial excited state is localized on
the unsubstituted bpy ligand. .......................................................115
Figure 4-7: Time resolved infrared measurements on complex 2. in a
nitromethane solution. (a) 490 nm excitation wavelength where the
initial excited state is localized on the CN-Me-bpy ligand (b) 400 nm
excitation wavelength where the initial excited state is localized on
the unsubstituted bpy ligand. .......................................................116

xvii

Figure 4-8: Time resolved infrared measurements on complex 3. in a
nitromethane solution. (a) 475 nm excitation wavelength where the
initial excited state is localized on the CN-Me-bpy ligand (b) 400 nm
excitation wavelength where the initial excited state is localized on
the CN-Me-bpy bpy ligand. ..........................................................117
Figure 4-9: Kinetic trace of the absorption maximum (2200 cm-1) for complex 1
excited at 400 nm. The red line is the result of a single exponential
fit of the data yielding a time constant of 2.29 ± 0.7 ps ...............117
Figure 4-10: Results of polarized absorption spectroscopy, showing no
anisotropic decay. This shows that the ILET from bpy to CN-Mebpy is much faster than 1 ps. .......................................................118
Figure 4-11: Full spectra for compound 1 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the
nanosecond transient absorption spectrum (blue squares) .........120
Figure 4-12: Full spectra for compound 2 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the
nanosecond transient absorption spectrum (green triangle). The
scatter at 490 pump wavelength was removed for a better
comparison...................................................................................121
Figure 4-13: Full spectra for compound 3 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the
nanosecond transient absorption spectrum (red circle) ...............122
Figure 4-14: Single wavelength kinetics for complex 1 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 400 nm (red), (c) 460
nm (green), and (d) 540 nm (purple). The black line is the result of
a double exponential fit, yielding lifetimes of 0.42 ± 0.05 ps and
9.20 ± 2.5 ps. ..............................................................................123
Figure 4-15: Single wavelength kinetics for complex 2 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 410 nm (red), (c) 460
nm (green), and (d) 540 nm (purple). Black line is the results of a
double exponential fit, yielding lifetimes of 0.39 ± 0.13 ps and 7.16
± 2.1 ps. .....................................................................................124

xviii

Figure 4-16: Single wavelength kinetics for complex 3 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 410 nm (red), (c) 460
nm (green), and (d) 540 nm (purple). Black lines are the results of
double exponential fits, yielding lifetimes of 0.15 ± 0.07 ps and 5.25
± 1.5 ps for 410 nm and 0.10 ± .01 ps and 5.45 ± 0.87 ps for 540
nm.
.....................................................................................125
Figure 4-17: Full spectra for compound 1 excited at 490 nm, probed in 33 fs
steps.
.....................................................................................126
Figure 4-18: Full spectra for compound 2 excited at 490 nm, probed in 33 fs
steps.
.....................................................................................126
Figure 4-19: Full spectra for compound 2 excited at 490 nm, probed in 33 fs
steps.
.....................................................................................127
Figure 4-20: Full spectra for [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1) after 425 nm
excitation, probed in (a) 1 ps steps and (b) 33 fs steps ...............128
Figure 4-21: Full spectra for compound 2 with 425 nm excitation probed in (a) 1
ps intervals and (b) 33 fs intervals. ..............................................129
Figure 4-22: Full spectra for compound 3 with 425 nm excitation probed in (a) 1
ps intervals and (b) 33 fs intervals. ..............................................130
Figure 4-23: Single wavelength kinetics for complex 1 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c)
520 nm (green). The black lines on the 460 nm and 520 nm traces
are the results of biexponential fits, yielding lifetimes of 0.12 ± .03
ps and 0.33 ± 0.13 ps for 460 nm and 0.14 ± .07 ps and 1.46 ± 2.3
ps for 540 nm. ..............................................................................131
Figure 4-24: Single wavelength kinetics for complex 2 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c)
520 nm (green). The black line on the 520 nm trace is the result of
a single exponential fit yielding a lifetime of 7.25 ± 1.3 ps. .........132

xix

Figure 4-25: Single wavelength kinetics for complex 3 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c)
540 nm (green). The black line on the 540 nm trace is the result of
a biexponential fit yielding lifetimes of 0.15 ± 0.08 ps and 0.32 ±
0.27 ps .....................................................................................133
Figure 4-26: Single wavelength trace for [Ru(bpy)2(CN-Me-bpy)](PF6)2 collected
at 450 nm, following excitation at 490 nm. The red line is the result
of a single exponential fit, resulting in a lifetime of
180 ± 30 ps. .................................................................................135
Figure 4-27: Electronic absorption spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1)
in different solvents, showing the solvatochromatic behavior of the
lowest energy MLCT absorption band. The inset shows an
expanded view of the lowest energy absorption maxima.............136
Figure 4-28: A series of [Ru(bpy')2(CN-Me-bpy)](PF6)2 complexes with varying
energy differences between bpy' and CN-Me-bpy. Listed energy
differences are estimated by the difference in the first reduction
potential for the corresponding homoleptic complexes, i.e.
Ered{[Ru(CN-Me-bpy)3]2+} - Ered{[Ru(bpy')3]2+}. ..........................138
Figure 5-1: Examples of different isomers formed with two substituents (A and
B) in a ruthenium tris(bipyridine) complex (top) and a ruthenium
bis(terpyridine) complex (bottom).................................................146
Figure 5-2: Examples of ruthenium(II) complexes with tridentate ligands having
a near octahedral coordination geometry. (a) [Ru(dqp)2]2+ (dqp =
2,6-di(quinolin-8-yl)pyridine)4 (b) [Ru(tripy)2]2+ (tripy =1,1'-(2,6pyridinediyl)bis[1-(2-pyridinyl)methanone])5 ................................148
Figure 5-3: Structures of the cyano-substituted terpyridine ruthenium(II)
complexes discussed in this chapter............................................149
Figure 5-4: Normalized absorption spectra of [Ru(tpy)(CN-tpy)]2+ (1, blue) and
[Ru(CN-tpy)2]2+ (2, red) and [Ru(tpy)2]2+(black) in acetonitrile
solution. Spectra are normalized at the maximum of the MLCT
absorption feature to better show the change in shape of the MLCT
absorption band across the series. ..............................................158

xx

Figure 5-5: Cyclic voltammogram of [Ru(tpy)(CN-tpy)](PF6)2 (1) in CH3CN with
0.1 M TBAPF6 as the electrolyte. Spectrum was collected using
the standard three electrode setup with a platinum working
electrode, graphite counter electrode, and a Ag/AgCl reference
electrode and is referenced to the ferrocene/ferrocenium redox
couple, which was added as an internal standard. The differential
pulse voltammogram is shown in the insets.................................160
Figure 5-6: Electronic absorption spectra of [Ru(CN-tpy)2]2+ (2) in methanol
(red line, λmax= 488 nm), acetonitrile (orange line, λmax= 491 nm),
acetone (green line, λmax= 491 nm), nitromethane (blue line, λmax=
492 nm), DMF (purple line, λmax= 494 nm), and DMSO (black line,
λmax= 495 nm). The inset shows an expanded view of the
absorption maxima. ......................................................................161
Figure 5-7: Plot of MLCT absorption maximum vs solvent dipole moment for
[Ru(CN-tpy)2]2+ in various solvents (see Figure 5-6). The plot
shows a correlation between the absorption maximum and the
solvent dipole moment, suggesting the solvatochromatic response
is caused by the charged excited state being stabilized by solvent
dipole interactions. .......................................................................163
Figure 5-8: Steady-state emission spectra of [Ru(tpy)(CN-tpy)](PF6)2 (1, blue
triangles) and [Ru(CN-tpy)2](PF6)2 (2, red circles) (a) Roomtemperature spectra acquired in deoxygenated CH3CN solution.
The solid lines correspond to fits to an asymmetric double
sigmoidal function – see text for further details. (b) Emission
spectra acquired in a 9:2 butyronitrile/propionitrile glass
at 80 K .....................................................................................165
Figure 5-9: Time resolved emission decay traces for [Ru(tpy)(CN-tpy)](PF6)2
(1, blue triangle) and [Ru(CN-tpy)2](PF6)2 (2, red circle). Black
lines are single exponential fits of the emission decay. ...............168
Figure 5-10: Schematic representation of the effect of electron accepting (A) and
electron donating (D) substituents on the energy of the HOMO
(π M) and LUMO (π*L) orbitals in [Ru(tpy')2]2+ complexes. Purple
arrows represent ligand field transitions and green arrows
represent metal-to-ligand charge transfer transitions. Reproduced
from reference 29 .........................................................................169

xxi

Figure 5-11: Nanosecond time-resolved differential absorption spectra acquired
in room-temperature CH3CN solution for (a) [Ru(tpy)(CNtpy)](PF6)2 (1) and (b) [Ru(CN-tpy)2](PF6)2 (2). The individual
points correspond to the amplitudes of fits of the kinetics data to
single-exponential decay models; a smoothed solid line has been
included in both plots to guide the eye. ........................................171
Figure 5-12: A series of ruthenium(II) terpyridyl incorporating an increasing
number of electron withdrawing cyanide groups. .........................172
Figure A-1: 1H-NMR of [Ru(bpy)2(CN-Me-bpy)](PF6)2 in CD3CN. Inset shows
an expanded view of the aromatic region. ...................................180
Figure A-2: Electronic absorption spectra of the conversion of [Ru(bpy)(Cl-Mebpy)2]2+ to [Ru(bpy)(CN-Me-bpy)2]2+. The inset indicates the time
following initiation of the reaction that each aliquot was taken. See
text for further details. ..................................................................181
Figure A-3: 1H-NMR of [Ru(bpy)(CN-Me-bpy)2](PF6)2 in CD3CN. Inset shows
an expanded view of the aromatic region. ...................................182
Figure A-4: Cyclic voltammogram of [Ru(bpy)(CN-Me-bpy)2](PF6)2 in CH3CN
with 0.1 M TBAH as the electrolyte. Spectrum was collected using
the standard three electrode setup with a platinum working
electrode, graphite counter electrode, and a Ag/AgCl reference
electrode and is referenced to the ferrocene/ferrocenium redox
couple, which was added as an internal standard. The differential
pulse voltammogram is shown in the insets.................................183
Figure A-5: Electronic absorption spectra of the conversion of [Ru(Cl-Mebpy)3]2+ to [Ru(CN-Me-bpy)3]2+. The inset indicates the time
following initiation of the reaction that each aliquot was taken. See
text for further details ...................................................................184
Figure A-6: 1H-NMR of [Ru(CN-Me-bpy)3](PF6)2 in CD3CN. Inset shows an
expanded view of the aromatic region. ........................................185

xxii

Figure A-7: Cyclic voltammogram of [Ru(bpy)3](PF6)2 in CH3CN with 0.1 M
TBAH as the electrolyte. Spectrum was collected using the
standard three electrode setup with a platinum working electrode,
graphite counter electrode, and a Ag/AgCl reference electrode and
is referenced to the ferrocene/ferrocenium redox couple, which was
added as an internal standard.
The differential pulse
voltammogram is shown in the insets. .........................................186
Figure A-8: (a) Ground state absorption spectrum for [Ru(bpy)(CN-Mebpy)2](PF6)2 (2) in room-temperature CH3CN solution. (b)
Oxidative difference spectra acquired at an applied potential of
+1.68 V versus Ag/AgCl. The inset corresponds to an expanded
view of the low-energy portion of the spectrum. (c) Reductive
difference spectra acquired at an applied potential of -0.980 V
versus Ag/AgCl. (d) Time-resolved differential absorption spectrum
of compound 1 following ca. 10 ns excitation at 500 nm. ............187
Figure A-9: (a) Ground state absorption spectrum for [Ru(CN-Me-bpy)3](PF6)2
(3) in room-temperature CH3CN solution. (b) Oxidative difference
spectra acquired at an applied potential of +1.85 V versus Ag/AgCl.
The inset corresponds to an expanded view of the low-energy
portion of the spectrum. (c) Reductive difference spectra acquired
at an applied potential of -0.900 V versus Ag/AgCl. (d) Timeresolved differential absorption spectrum of compound 1 following
ca. 10 ns excitation at 500 nm. ....................................................188
Figure A-10: Steady-state and time-resolved infra-red absorption data for 2 in
CH3NO2 solution. (a) Comparison of the ground-state (blue line)
and step-scan infrared differential absorption data acquired at a
time delay of 150 ns following ca. 10 ns excitation at 540 nm (red
line). The ~40 cm-1 red-shift in the CN stretching frequency reflects
the presence of an electron in the π* orbital of the CN-Me-bpy
ligand in the 3MLCT excited state. (b) Nanosecond step-scan
infrared spectra as a function of time following ca. 10 ns excitation
at 540 nm. ...................................................................................189

xxiii

Figure A-11: Steady-state and time-resolved infra-red absorption data for 3 in
CH3NO2 solution. Comparison of the ground-state (blue line) and
step-scan infrared differential absorption data acquired at a time
delay of 150 ns following ca. 10 ns excitation at 500 nm (red line).
The ~40 cm-1 red-shift in the CN stretching frequency reflects the
presence of an electron in the π* orbital of the CN-Me-bpy ligand in
the 3MLCT excited state. .............................................................190
Figure B-1: 1H NMR spectrum of [Ru(tpy)(CN-tpy)](PF6)2 in CD3CN. ...........191
Figure B-2: Infrared spectrum of [Ru(tpy)(CN-tpy)](PF6)2 (1) in a KBr
pellet
.....................................................................................192
Figure B-3: 1H NMR spectrum of [Ru(CN-tpy)2]2+ in CD3CN. ........................193
Figure B-4: Infrared spectrum of [Ru(CN-tpy)2](NO3)1.5(PF6)0.5 (2) in a KBr
pellet
.....................................................................................194
Figure B-5: Cyclic voltammogram of [Ru(CN-tpy)2]2+ (2) in CH3CN with 0.1 M
TBAH as the electrolyte. Spectrum was collected using the
standard three electrode setup with a platinum working electrode,
graphite counter electrode, and a Ag/AgCl reference electrode and
is referenced to the ferrocene/ferrocenium redox couple, which was
added as an internal standard.
The differential pulse
voltammogram is shown in the insets ..........................................195
Figure B-6: Electronic absorption spectra of [Ru(tpy)(CN-tpy)](PF6)2 in
acetonitrile (red line, λmax= 488 nm), acetone (orange line, λmax=
488 nm), nitromethane (green line, λmax= 490 nm), DMSO (blue
line, λmax= 492 nm), and pyridine (purple line, λmax= 492 nm).
The inset shows an expanded view of the absorption
maxima. .....................................................................................196

xxiv

KEY TO ABBREVIATIONS AND SYMBOLS
ΔQ ..........................Nuclear displacement of excited state relative to the ground
state
Δν̅0,1/2 .....................Spectral bandwidth
ωM ........................Average vibrational mode
Φ .............................Quantum yield
acac ........................Acetylacetoneate
bpy ..........................2,2'-bipyridine
CCD ........................Charge coupled device
Cl-Me-bpy ...............4,4'-dichloro-5,5'-dimethyl-2,2'-bipyridine
Cl-tpy ......................4-chloro-2,2':6',2"-terpyridine
CN-Me-bpy .............4,4'-dicyano-5,5'-dimethyl-2,2'-bipyridine
CN-tpy .....................4-cyano-2,2':6',2"-terpyridine
CS ...........................Charge separated state
CV ...........................Cyclic voltammetry
DPV ........................Differential pulse voltammetry
E0 ............................Zero point energy difference
IC ............................Internal conversion
IL .............................Intraligand transition
ILET ........................Interligand electron transfer
IRF ..........................Instrument response function
ISC ..........................Intersystem crossing
IVR ..........................Intramolecular vibrational redistribution
LF ............................Ligand field
LMCT ......................Ligand-to-metal charge transfer
MLCT ......................Metal-to-ligand charge transfer
MMCT .....................Metal-to-metal charge transfer
OPO ........................Optical parametric oscillator

xxv

phen ........................1,10-phenanthroline
PMT ........................Photomultiplier tube
SM ...........................Huang-Rhys factor
SSIR .......................Step-scan infrared
TBAPF6 ...................Tetrabutylammonium hexafluorophosphate
tpy ...........................2,2':6',2"-terpyridine
TR2 .........................Transient resonance Raman
TR3 .........................Time-resolved resonance Raman
TRIR .......................Time-resolved infrared
VR ...........................Vibrational relaxation

xxvi

Chapter 1: Introduction to Photophysical Properties of Ruthenium
Polypyridyl Complexes

1.1 Introduction
This dissertation investigates the fundamental photophysics of ruthenium
polypyridyl complexes, in particular the non-radiative decay between the initial
excited Franck-Condon state and the long lived excited state. The complexes
studied in this dissertation incorporate cyanide groups as infrared tags in order to
use infrared transient absorption spectroscopy, coupled with visible transient
absorption spectroscopy, to probe the vibrational relaxation dynamics in
ruthenium polypyridyl complexes. This chapter highlights what is known in the
literature about the excited state dynamics of [Ru(bpy)3]2+ as a model for all
ruthenium(II) polypyridyl complexes with emphasis on topics relevant to the work
in this dissertation.

Examples of applications for ruthenium(II) polypyridyl

complexes are also highlighted.

1.2 Ground State Properties of [Ru(bpy)3]2+
The discovery of emission from [Ru(bpy)3]2+, the first example of charge
transfer emission, over 50 years ago was the beginning of research interest in
the photophysics of ruthenium polypyridyl complexes.1

In addition to being

(relatively) highly emissive the long lived excited states and reversible oxidation
and reduction chemistry makes this class of compounds useful in many different
applications

and

drives

research

into

understanding

their

fundamental

photophysical behavior.2
In the ground state electronic absorption spectrum of ruthenium(II)
complexes, there are generally three types of transitions that can be observed.
These transitions are defined by the nature of the initial and final orbitals involved

1

in the transitions, as shown in the molecular orbital diagram in Figure 1-1. The
intraligand (IL) transition is a transition between two ligand based orbitals, the
ligand field (LF) transition (also referred to as a d-d or metal centered transition)
is a transition between two metal based orbitals, and the metal-to-ligand charge
transfer (MLCT) transition is a transition from a metal based orbital to a ligand

d"*
L!*
L!*
MLCT

d!
d"

LF

d!

L"

IL
L!

L!

L"
M

ML 6

6L

Figure 1-1: Simplified molecular orbital diagram showing the interactions
between a π acceptor ligand and metal d orbitals in Oh symmetry. For a low spin
d6 metal center the possible transitions in the electronic absorption spectrum are
shown by the arrows. MLCT (blue arrow) is a metal to ligand charge transfer
transition from a filled metal orbital to an empty π* ligand orbital. IL (green arrow)
is an intraligand transition from a filled ligand π orbital to an empty π* ligand. LF
(red arrow) is a ligand field transition from a filled metal d orbital to an empty
metal d orbital. For interpretation of the references to color in this and all other
figures, the reader is referred to the electronic version of this dissertation.

2

based orbital.

The ligand field strength, redox properties of the metal and

ligand(s), as well as the intrinsic properties of the ligand(s) are the three major
factors in determining the relative energies of these three transitions.

The

electronic absorption spectrum of [Ru(bpy)3](PF6)2, the prototypical ruthenium
polypyridyl complex, is shown in Figure 1-2. The intense absorption at 285 nm
has been assigned as an intraligand transition by consideration of the extinction
coefficient and comparison to the absorption spectrum of the protonated
bipyridine ligand3 and the small shoulders on the IL band at approximately 320
nm and 350 nm have been tentatively assigned as ligand field absorptions.4

Wavelength (nm)
250

300

400

500

600

IL

80

Molar Absortivity
3 -1
-1
(x10 M cm )

350

60

40

20

MLCT

LF
LF

0
40

35

30

25
3

20

-1

Energy (x10 cm )

Figure 1-2: Electronic absorption spectrum of [Ru(bpy)3](PF6)2 in CH3CN
solution. MLCT is the metal-to-ligand charge transfer absorption band, LF is the
ligand field absorption bands, and IL is the intraligand π → π* absorption band.

3

The absorption band at 450 nm has been assigned as the metal-to-ligand
charge transfer absorption manifold based on extinction coefficient and
electrochemical data. An MLCT transition can be thought of as the simultaneous
oxidation of the metal center and reduction of the ligand.

Therefore, to a

reasonable approximation, the energy of a metal-to-ligand charge transfer
absorption is the sum of the energy required to oxidize the metal and reduce the
ligand i.e. E(MLCT) ≈

E(M

M+)

+

E(L

L -)

.5 Figure 1-3 shows the

cyclic voltammogram of [Ru(bpy)3](PF6)2; the RuII/III oxidation can be seen at
+0.91 V and the first bpy0/- reduction is at -1.70 V versus the ferrocene/
ferrocenium

redox

couple.

According

the

2 !A

approximation

above,

-1.6

-2.4

-2.0

this

1.2 1.0 0.8 0.6

1.0

0.0

-1.0
Potential (V)

-2.0

Figure 1-3: Cyclic voltammogram of [Ru(bpy)3](PF6)2 in CH3CN with 0.1 M
TBAH as the electrolyte. The spectrum was collected using the standard three
electrode setup with a platinum working electrode, graphite counter electrode,
and a Ag/AgCl reference electrode. The spectrum is referenced to the ferrocene/
ferrocenium redox couple, which was added as an internal standard. The
differential pulse voltammograms are shown in the insets.

4

electrochemical data predicts that the metal-to-ligand charge transfer transition in
[Ru(bpy)3]2+ should occur at 2.61 eV (475 nm) which is fairly close to the
experimentally observed 450 nm, especially considering this model does not
account for solvation energy or electron correlation effects.5
While there are rare examples of ruthenium(II) polypyridyl-like complexes
where the lowest energy excited state is a ligand field state or an intraligand
state,6-8 a vast majority of these complexes have photophysical properties which
are comparable to the [Ru(bpy)3]2+ prototype. Because of this, lessons learned
from studying [Ru(bpy)3]2+ can be applied to a wide range of systems.
1.3 Long Lived Excited State of [Ru(bpy)3]2+
The long lived emissive state in [Ru(bpy)3](PF6)2 has been identified as being
3MLCT

in nature9 and over the years, many different spectroscopic techniques

have been used to characterize the 3MLCT state of [Ru(bpy)3]2+.

Variable

temperature emission experiments have shown that the lowest energy 3MLCT is
actually split into a set of three closely spaced (~80 cm-1) states.10-12 These
states have differing amounts of mixing with higher energy singlet states and
therefore different ground state recovery kinetics. The dynamics observed at
room temperature represent a Boltzmann average of the three states, but at low
temperature the energies and dynamics of the three individual states can be
resolved.
In addition to emission spectroscopy many other spectroscopic techniques
have been used to observe the long lived excited state in [Ru(bpy)3]2+ and other
related molecules. Results of resonance Raman,13,14 time-resolved infrared,15
time-resolved electron paramagnetic resonance,16 and other techniques all show
that the emissive state in ruthenium(II) polypyridyl complexes is localized on a
single ligand rather than being evenly delocalized over all three ligands.

5

Upon absorption of a photon, [Ru(bpy)3]2+ is excited into a 1MLCT state,
which rapidly relaxes into a long lived 3MLCT state via a combination of
intersystem crossing (ISC) and vibrational relaxation (Figure 1-4). This long lived
excited state decays back to the ground state via radiative and non radiative
decay. Ground state recovery in most ruthenium polypyridyl complexes lies in
the Marcus inverted region and non-radiative decay can be modeled with the
energy gap law.17-19

1

MLCT
3

MLCT

Energy

ISC
Vibrational
Relaxation
1

A1

h"

h"!

Nuclear Coordinate (!Q)

Figure 1-4: Potential energy diagram for [Ru(bpy)3]2+. The 1A1 ground state can
be excited with visible light into a 1MLCT state. This initial 1MLCT state rapidly
intersystem crosses to form the 3MLCT state, which is the long lived excited
state. The 3MLCT state can relax back to the ground state through radiative and
non-radiative decay.

6

1.4 Ultrafast Dynamics of [Ru(bpy)3]2+
The advent of ultrafast spectroscopic techniques has allowed for studying
shorter timescale dynamics; i.e. the relaxation from the initially excited state to
the long lived excited state. Three general questions or themes investigated are:
What is the timescale of intersystem crossing between the 1MLCT state and the
3MLCT

state? What is the timescale and mechanism of vibrational relaxation on

the long lived 3MLCT surface?

Are the initial and long lived excited states

delocalized over the entire molecule or localized on a single ligand?

1.4.1 Intersystem Crossing
Differential absorption spectroscopy has been used perviously in our group to
infer a formation time of approximately 100 fs for the 3MLCT state,20 but there is
not necessarily a difference between the absorption spectrum of a 1MLCT state
and a 3MLCT state, so this technique is not ideal for differentiating between
singlet and triplet states.

Fluorescence upconversion is an experimental

technique that is uniquely suited to monitor intersystem crossing.

In the

upconversion experiment, sum frequency generation is used to mix the emitted
light with a gate pulse to create a unique frequency of light to be detected. This
is a non-linear process, so it is selective for spin allowed fluorescence because of
the large kr value (and high photon flux) and discriminates against spin forbidden
phosphorescence because of the small kr value (and low photon flux).
Because of this, the 3MLCT phosphorescence, which dominates the steady
state emission, can be excluded in the fluorescence upconversion experiment.
The 1MLCT fluorescence, which is undetectable in the steady state can be
selectively monitored. Okada and coworkers used this technique in 2002 on
[Ru(bpy)3]2+ and saw a 40±15 fs decay of the 1MLCT emission and a 500 fs to 1
ps decay component that they tentatively assigned to emission from a

7

vibrationally excited 3MLCT state.21 In 2006, Chergui and coworkers also used
fluorescence upconversion to study the intersystem crossing dynamics of
[Ru(bpy)3]2+ and observed a 15±10 fs decay of the 1MLCT fluorescence but saw
no signature for vibrational relaxation.22 Using femtosecond stimulated Raman
spectroscopy coupled with fluorescence upconversion, Mathies and coworkers
have developed a three state model to describe the early time dynamics in
[Ru(bpy)3]2+ (Figure 1-5).23 This model state that the initially excited 1MLCT
state very rapidly (τ < 30 fs) decays to a lower energy 1MLCT state which is
stabilized by excited state solvent interactions. This lower energy 1MLCT state
then intersystem crosses to the long lived 3MLCT state with a time constant of
approximately 110 fs.
Together, all of these results show that despite its spin forbidden nature,
intersystem crossing occurs very rapidly in ruthenium polypyridyl systems. The
classic rules developed through the study of organic photophysics state that
vibrational relaxation is the fastest relaxation process, followed by internal
conversion, and intersystem crossing is the slowest relaxation process (kVR >
kIC > kISC). The results above, along with similar results for other transition metal
systems,24-27 show that the classic rules don't necessarily apply to inorganic
systems and inorganic photophysics are fundamentally different than organic
photophysics.

1.4.2 Vibrational Relaxation
The study of the vibrational relaxation dynamics in ruthenium(II) polypyridyl
complexes is the primary focus of the work in this dissertation. In the literature
previous studies on [Ru(bpy)3]2+ and other similar complexes has yielded some
conflicting results regarding vibrational relaxation in these complexes.

8

Figure 1-5: Three state model for intersystem crossing in [Ru(bpy)3]2+
developed by Mathies and coworkers. Reproduced from reference 23.

As briefly mentioned above, previous work in our group using ultrafast
differential absorption spectra of [Ru(bpy)3](PF6)2 showed that by 500 fs after a
480 nm excitation, the absorption spectrum was identical to the spectrum of the
thermalized 3MLCT state obtained by nanosecond transient absorption
spectroscopy.20

This suggests a formation time of approximately 100 fs for

thermalized 3MLCT state, which includes contributions from intersystem crossing
and vibrational relaxation. The work of Chergui and Mathies, discussed above,

9

also shows very rapid formation of the thermalized 3MLCT state, again
suggesting that vibrational relaxation in the 3MLCT state is occurring on the
same ultrafast timescale as intersystem crossing from the initially excited 1MLCT
state.22,23
In contrast to the results above, previous work from our group on the related
[Ru(dmb)3](PF6)2 (dmb = 4,4'-dimethyl-2,2'-bipyridine) showed an additional
decay signal (τ = 5 ± 0.5 ps) in the transient absorption spectra when exciting the
molecule at 400 nm (compared to 480 nm for the previous [Ru(bpy)3](PF6)2
study).28

This picosecond decay component was assigned as vibrational

relaxation on the 3MLCT surface. Browne, McGarvey, and coworkers recently
used picosecond resonance Raman spectroscopy to monitor the growth of
vibrations of the bipyridine radical anion in the thermalized 3MLCT state.29 In
contrast to the work of Mathies, Browne and McGarvey see the thermalized
triplet grow in over ~20 ps. Vibrational relaxation in many different transition
metal systems occurs on a picosecond timescale,25,30-34 which suggests that the
vibrational relaxation in ruthenium polypyridyl complexes most likely takes place
on a similar timescale. The conflicting conclusions in the literature highlight the
need for further study.

1.4.3 Localization or Delocalization of the Excited State
A metal-to-ligand charge transfer excited state can be describe as a localized
state,

i.e.,

{[Ru3+(bpy)2(bpy-)]2+}*

or

as

a

delocalized

state,

i.e.,

{[Ru3+(bpy-1/3)3]2+}*. As discussed in Chapter 1.3, there is a consensus that on
longer (nanosecond) timescales the long lived 3MLCT state of ruthenium
bipyridyl complexes is localized on a single ligand, even in homoleptic complexes
such as [Ru(bpy)3]2+. The question remains, is the initial Franck-Condon state
localized and, if it is, is that localization maintained throughout the relaxation

10

process?
Using Stark effect spectroscopy, Oh and Boxer showed that there was an 8.8
D change in the dipole moment of [Ru(bpy)3]2+ upon excitation. As the ground
state is D3 symmetry and has no permanent dipole this result means that the
initial excited state has a significant dipole moment and therefore must be
localized rather than being evenly delocalized over the three bipyridine ligands.35
Ground state resonance Raman experiments36 and variable solvent absorption
spectroscopy37 among other experiments also confirm that that the initial excited
state is localized on a single ligand in [Ru(bpy)3]2+.
Ultrafast polarized absorption measurements show a very rapid decrease in
the anisotropic signal from the excited state. Work from our group showed that
the timescale of this anisotropy decay was dependent on solvent and theorized
that it was due to the transition from a delocalized state to a localized state that is
driven by solvent interactions.38 Hammarström and coworkers suggest that the
excited state very rapidly (< 300 fs) randomizes, but it is unclear whether this
randomization is due to the formation of a delocalized state or interligand
electron transfer.39

Recently, Chergui and coworkers have used molecular

dynamics calculations to study the lowest energy triplet state of [Ru(bpy)3]2+. In
the gas phase calculations, the triplet state remained delocalized over all three
ligands for the entire simulation.

However, in calculations done in a water

dielectric continuum the triplet state localized very quickly and oscillated between
being localized on one or two bipyridine ligands.40
As of now there is no clear consensus in the literature about the localized or
delocalized nature of the excited state of [Ru(bpy)3]2+. There is an abundant
amount of experimental evidence to show that the long lived excited state is
localized on a single bipyridine ligand.

There is also strong experimental

evidence to show that the initial Franck-Condon state is also localized on a single

11

bipyridine ligand. What is still unclear is the evolution of the initial excited state
into the long lived excited state; anisotropy absorption measurements and
molecular dynamics simulations suggest that the initial localization of the excited
state isn't maintained. This randomization could be due to the formation of a
delocalized state that will randomly re-localize on a single ligand or it could be a
result of very rapid intramolecular transfer. The idea of excited state localization
vs delocalization and techniques to differentiate between delocalization and rapid
interligand electron transfer will be discussed further in Chapter 4.

1.5 Applications of Ruthenium Polypyridyl Complexes
One reason why ruthenium polypyridyl complexes are so highly studied are
their applications in many areas of chemistry.41 Dye sensitized solar cells often
use ruthenium polypyridyl complexes as dye molecules.42-47 The dye molecule
absorbs light and is excited into a metal-to-ligand charge transfer excited state. If
this excited state is above the conduction band edge, it can inject an electron into
the conduction band.

In these systems vibrational relaxation competes with

electron injection; in cases where the long lived excited state lies above the
conduction band vibrational relaxation represents a loss of potential (or driving
force for electron injection) for each electron and in cases where the long lived
excited state lies below the conduction band edge the electron injection must
outcompete the excited state deactivation in order to get a high injection yield.
As shown previously, ruthenium polypyridyl complexes exhibit reversible
oxidation and reduction behavior.

Because of this ruthenium polypyridyl

complexes are also often used as redox or photoredox catalysts. One of the
most famous is the ruthenium blue dimer developed by Meyer and coworkers in
the 1980's, which was the first example of a molecular catalyst able to split water.
48,49

The blue dimer is an oxo bridged ruthenium(III) dimer with the formula

12

cis,cis-[(bpy)2(H2O)RuIIIORuIII(OH2)-(bpy)2]4+ (Figure 1-6). The blue dimer can
electrochemically catalyze the 4-electron and 4-proton splitting of water in the
presence of cerium(IV) as a sacrificial oxidant.

Figure
1-6:
Structure
of
the
ruthenium
blue
dimer,
[(bpy)2(H2O)RuIIIORuIII(OH2)-(bpy)2]4+. Reproduced from reference 49

cis,cis-

The work of MacMillian and coworkers is another example of using ruthenium
polypyridyl complexes as catalysts.50

In their proposed catalytic cycle

photoexcited [Ru(bpy)3]2+ acts as both an oxidant and a reductant in two discrete
single electron transfer steps, creating two reactive radical species. These two
reactive radicals then combine to form new a carbon-carbon bonds, This is
different from most carbon-carbon bond forming reactions, which require a two
electron oxidation and/or reduction.

1.6 Dissertation Outline
The work described in this dissertation is primarily focused on studying
vibrational relaxation in ruthenium(II) polypyridyl complexes. This involves the
synthesis and characterization of series both of cyano substituted bipyridine and
13

cyano substituted terpyridine ligands followed by an ultrafast transient absorption
study of the bipyridine complexes. An outline of this dissertation is below:
• Chapter 2 is a discussion of the different nanosecond time-resolved
spectroscopic techniques commonly used to study photophysical and
photochemical processes in transition metal complexes.

These techniques

include time-resolved emission, transient absorption, time-resolved infrared
methods, time-resolved electron paramagnetic resonance, and time-resolved
X-ray spectroscopy. The experimental requirements for each technique and
examples from the literature of how each is used are covered.
• Chapter 3 discusses the synthesis and characterization of a series of cyanosubstituted bipyridine complexes. The properties of the ground state and long
lived excited state are investigated by spectroscopic and computational
methods.
• Chapter 4 focuses on the ultrafast dynamics of the cyano-substituted
bipyridine complexes introduced in Chapter 3.

Both visible and infrared

transient absorption spectroscopy are used to monitor the vibrational relaxation
process between the initially excited state and the long lived excited state as a
function of excitation wavelength. Time dependent DFT calculations are used
to identify the initial excited state for each of the excitation wavelengths used.
Possible future applications of these complexes to study excited state
delocalization will also be discussed.
• Chapter 5 reports studies on a series of cyano-substituted terpyridine
complexes. The properties of the ground state and long lived excited state are
investigated by spectroscopic methods and possible future directions on these
systems are discussed.

14

REFERENCES

15

1.7 References
(1) Paris, J. P.; Brandt, W. W. J. Am. Chem. Soc. 1959, 81, 5001-5002.
(2)

Campagna, S.; Puntoriero, F.; Nastasi, F.; Bergamini, G.; Balzani, V. Top.
Curr. Chem. 2007, 280, 117-214.

(3)

Lytle, F. E.; Hercules, D. M. J. Am. Chem. Soc. 1969, 91, 253-257.

(4)

Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; Von
Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85-277.

(5)

Vlček, A. A.; Dodsworth, E. S.; Pietro, W. J.; Lever, A. B. P. Inorg. Chem.
1995, 34, 1906-1913.

(6)

Liu, Y.; De, N., A; Reiff, O.; Ziessel, R.; Schanze, F. S. J. Phys. Chem. A.
2003, 107, 3476-3485.

(7)

Kozlov, D. V.; Tyson, D. S.; Goze, C.; Ziessel, R.; Castellano, F. N. Inorg.
Chem. 2004, 43, 6083-6092.

(8)

Murali, M.; Palaniandavar, M. Dalton Trans. 2006, 730-743.

(9)

Klassen, D. M.; Crosby, G. A. J. Chem. Phys. 1968, 48, 1853–1838.

(10) Harrigan, R. W.; Crosby, G. A. J. Chem. Phys. 1973, 59, 3468.
(11) Yersin, H.; Gallhuber, E. J. Am. Chem. Soc. 1984, 106, 6582-6586.
(12) Yersin, H.; Humbs, W.; Strasser, J. Coord. Chem. Rev. 1997, 159, 325-358.
(13) Bradley, P. G.; Kress, N.; Hornberger, B. A.; Dallinger, R. F.; Woodruff, W.
H. J. Am. Chem. Soc. 1981, 103, 7441-7446.
(14) Mabrouk, P. A.; Wrighton, M. S. Inorg. Chem. 1986, 25, 526–531.
(15) Omberg, K. M.; Schoonover, J. R.; Treadway, J. A.; Leasure, R. M.; Dyer,
R. B.; Meyer, T. J. J. Am. Chem. Soc. 1997, 119, 7013–7018.
(16) Morris, D. E.; Hanck, K. W.; DeArmond, M. K. J. Am. Chem. Soc. 1983,
105, 3032-3038.
(17) Strouse, G. F.; Schoonover, J. R.; Duesing, R.; Boyde, S.; Jones, W. E.;
Meyer, T. J. Inorg. Chem. 1995, 34, 473-487.

16

(18) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys. Chem.
1986, 90, 3722-3734.
(19) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J. Chem.
Soc. 1997, 119, 8253–8268.
(20) Damrauer, N. H.; Cerullo, G.; Yeh, A.; Boussie, T. R.; Shank, C. V.;
McCusker, J. K. Science. 1997, 275, 54–57.
(21) Bhasikuttan, A. C.; Suzuki, M.; Nakashima, S.; Okada, T. J. Am. Chem.
Soc. 2002, 124, 8398-8405.
(22) Cannizzo, A.; van Mourik, F.; Gawelda, W.; Zgrablic, G.; Bressler, C.;
Chergui, M. Angew. Chem. Int. Ed. Engl. 2006, 45, 3174-3176.
(23) Yoon, S.; Kukura, P.; Stuart, C. M.; Mathies, R. A. Molecular Physics. 2006,
104, 1275-1282.
(24) Monat, J. E.; McCusker, J. K. J. Am. Chem. Soc. 2000, 122, 4092–4097.
(25) Juban, E. A.; McCusker, J. K. J. Am. Chem. Soc. 2005, 127, 6857–6865.
(26) Forster, L. S. Coord. Chem. Rev. 2006, 250, 2023-2033.
(27) Schrauben, J. N.; Dillman, K. L.; Beck, W. F.; McCusker, J. K. Chem. Sci.
2010, 1, 405.
(28) Damrauer, N. H.; McCusker, J. K. J. Phys. Chem. A. 1999, 103, 8440–
8446.
(29) Henry, W.; Coates, C. G.; Brady, C.; Ronayne, K. L.; Matousek, P.; Towrie,
M.; Botchway, S. W.; Parker, A. W.; Vos, J. G.; Browne, W. R.; McGarvey,
J. J. J. Phys. Chem. A. 2008, 112, 4537-4544.
(30) King, J. C.; Zhang, J. Z.; Schwartz, B. J.; Harris, C. B. J. Chem. Phys. 1993,
99, 7595.
(31) Liard, D. J.; Busby, M.; Matousek, P.; Towrie, M.; Vlček, A. A. J. Phys.
Chem. A. 2004, 108, 2363-2369.
(32) Záliš, S.; Busby, M.; Kotrba, T.; Matousek, P.; Towrie, M.; Vlček, A. A.
Inorg. Chem. 2004, 43, 1723–1734.

17

(33) Shaw, G. B.; Styers-Barnett, D. J.; Gannon, E. Z.; Granger, J. C.;
Papanikolas, J. M. J. Phys. Chem. A. 2004, 108, 4998-5006.
(34) Maçôas, E. M.; Kananavicius, R.; Myllyperkiö, P.; Pettersson, M.; Kunttu, H.
J. Phys. Chem. A. 2007, 111, 2054–2061.
(35) Oh, D. H.; Boxer, S. G. J. Am. Chem. Soc. 1989, 111, 1130-1131.
(36) Webb, M. A.; Knorr, F. J.; McHale, J. L. J. Raman Spectrosc. 2001, 32,
481-485.
(37) Kober, E. M.; Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1984, 23,
2098-2104.
(38) Yeh, A. T.; Shank, C. V.; McCusker, J. K. Science. 2000, 289, 935–938.
(39) Wallin, S.; Davidsson, J.; Modin, J.; Hammarström, L. J. Phys. Chem. A.
2005, 109, 4697-4704.
(40) Moret, M. E.; Tavernelli, I.; Chergui, M.; Rothlisberger, U. Chem. Eur. J.
2010, 16, 5889-5894.
(41) Kalyanasundaram, K. Coord. Chem. Rev. 1982, 46, 159-244.
(42) OʼRegan, B.; Grätzel, M. Nature. 1991, 353, 737-740.
(43) Hagfeldt, A.; Grätzel, M. Acc. Chem. Res. 2000, 33, 269-277.
(44) Gratzel, M. Inorg. Chem. 2005, 44, 6841-6851.
(45) Grabulosa, A.; Beley, M.; Gros, P. C.; Cazzanti, S.; Caramori, S.; Bignozzi,
C. A. Inorg. Chem. 2009,
(46) Chen, C. Y.; Wang, M.; Li, J. Y.; Pootrakulchote, N.; Alibabaei, L.; Ngoc-le,
C. H.; Decoppet, J. D.; Tsai, J. H.; Grätzel, C.; Wu, C. G.; Zakeeruddin, S.
M.; Grätzel, M. ACS Nano. 2009,
(47) Kim, J. J.; Lim, K.; Choi, H.; Fan, S.; Kang, M. S.; Gao, G.; Kang, H. S.; Ko,
J. Inorg. Chem. 2010,
(48) Gersten, S. W.; Samuels, G. J.; Meyer, T. J. J. Am. Chem. Soc. 1982, 104,
4029-4030.

18

(49) Concepcion, J. J.; Jurss, J. W.; Brennaman, M. K.; Hoertz, P. G.;
Patrocinio, A. O.; Murakami Iha, N. Y.; Templeton, J. L.; Meyer, T. J. Acc.
Chem. Res. 2009, 42, 1954-1965.
(50) Nicewicz, D. A.; MacMillan, D. W. C. Science. 2008, 322, 77-80.

19

Chapter 2: The Application of Nanosecond Time-Resolved Spectroscopies
to the Study of Inorganic Photophysics and Photochemistry

2.1 Introduction
Nanosecond time-resolved spectroscopic methods are becoming more
accessible to the average inorganic chemist.

Unlike ultrafast spectroscopic

techniques, nanosecond systems are generally reasonably affordable and do not
require extensive training and expertise to operate. This chapter discusses timeresolved nanosecond spectroscopic techniques that are most commonly used to
investigate photochemical and photophysical processes in inorganic systems.
The necessary equipment and experimental requirements for each technique will
be discussed and examples of how they are used in the literature will be given.

2.2 Time-Resolved Emission Spectroscopy
2.2.1 Experimental Setup
Time-resolved emission is the simplest of the nanosecond time-resolved
experiments.

The time-resolved emission experiment consists of three main

parts; an excitation source, monochromator, and detector. The excitation source
is generally a pulsed laser, such as a Q-switched Nd:YAG laser (Figure 2-1).
This can either be used alone to excite the sample with a limited number of
fundamental and harmonic frequencies (1064 nm, 532 nm, 355 nm, or 266 nm
for Nd:YAG) or it can be coupled to an optical parametric oscillator (OPO) to
generate a continuous range of frequencies across the UV, visible, and/or
infrared. In some cases, flashlamps, which excite the sample with a temporally
short polychromatic light pulses, are also used as an excitation source. This
practice was far more common before pulsed lasers and OPO's were as widely
available as they are today.

20

Photodiode
Trigger
Sample

Nd:YAG
Q-switched
laser

Oscilloscope

Monochromator
PMT

Figure 2-1: Example optical layout of a time-resolved emission spectrometer. A
pulsed Nd:YAG laser excites the sample and a fraction of the emission is
collected and focused into a monochromator. The emission decay is monitored
in real time at a single wavelength. A photodiode trigger synchronized data
collection with sample excitation.

The sample is excited with the pulsed laser source. While the peak power of
a nanosecond laser pulse is much less than in picosecond or femtosecond laser
pulses, two photon absorption of the sample is still a possibility in some
circumstances. For this experiment, as well as the the others discussed in this
chapter, two photon processes can be easily avoided by checking that the
emission intensity responds linearly with changes in pump energy before each
experiment. When the sample is excited, emission occurs non-coherently in all
directions, so lenses are used to capture a fraction of the emitted light at 90° to
the excitation beam to minimize interference and scatter from the excitation
beam.

The collected emission is focused into a monochromator and the

emission decay at a single wavelength is monitored and recorded in real time. A
trigger is used to synchronize the sample excitation and data collection

21

In the emission experiment, a signal is measured against a dark
background. The ideal detector for emission will be able to reliably measure a
small signal with a minimal amount of background noise. The detector that is
most commonly used in the emission experiment is the photomultiplier tube
(PMT). Alternatively, an array detector, such as a charge coupled device (CCD),
and spectrograph can be used in order to probe the entire emission spectrum as
a function of time rather than probing the emission decay at a single wavelength.
While there may be an advantage to probing the entire emission spectrum rather
than a single wavelength, the higher cost of the array detector leads to the PMT
being far more commonly used.
The temporal resolution of any time-resolved experiment is determined by the
temporal width of the excitation pulse, the bandwidth or response time of the
detector and the bandwidth of any other electronics used to read and/or record
the detector signal.

These factors combined are the instrument response

function (IRF) for a particular experiment. In practice, the ~10 ns laser excitation
pulse is generally what limits the temporal resolution of the time resolved
emission experiment.
When choosing a sample for a time-resolved emission experiment, a few
factors should be considered. One clear requirement is the sample must be
emissive and lifetime must be long enough (at least 3 times the IRF) to be
accurately measured. Other experimental techniques are available to measure
emission on picosecond and femtosecond timescales, but they will not be
discussed here. The sample should also be optically dilute, an absorbance of
0.1 to 0.2 at the excitation wavelength, to avoid intramolecular interactions and
possible self quenching.

Because emission is measured against zero

background, it is an incredibly sensitive technique.

This can be a great

advantage with quenched or weakly emitting systems, but also makes the

22

experiment vulnerable to emissive impurities, which can dominate the spectrum
even when present in trace amounts that are undetectable by other
characterization methods.

This can be a particular problem with covalently

attached donor-acceptor complexes, where any complex dissociation will result
in a (relatively) highly emissive, unquenched fluorophore.1,2

17

19
18
15
17

33

ln(k nr x 1s)

32

15

14 11

16

13

31
30

29

10
8
12
7
28 9

5

27
25
26
24
23

13

3

6
4

2
1

22
20
21

11

13

15

17
-1

19

-3

Eem(cm x10 )

Figure 2-2: Plot of the observed value of In(knr) vs. Eem, the emission maximum
in acetonitrile solution. Complexes have the general formula [Os(bpy')2(L)]2+,
where L is one bidentate or two monodentate ligands and bpy' is bpy ( ), bpy-d8
( ), or phen (▴). The drawn lines indicate the separate linear correlations for the
bpy and phen series, both with slopes of (1,100 cm-1). Reproduced from
reference 3.

23

2.2.2 Examples
Coupled with quantum yield measurements, the radiative and non-radiative
decay contributions to the observed lifetime (τobs) can be calculated using
Equations 2-1 and 2-2. Unlike using the quantum yield (Φ) or lifetime alone,
!=

kr
kobs

kobs = kr + knr

2-1

2-2

trends in kr and knr across a series of molecules can give insight into changes in
transition moment dipole or excited state geometry distortion.4 In 1986 Meyer
and coworkers measured the emission lifetimes and quantum yields across a
large series of osmium(II) polypyridyl complexes. Using this data, they were able
to show, for the first time, that non-radiative decay from the long-lived excited
state back to the ground state in transition metal complexes follows the energy
gap law (Figure 2-2).3
Measuring the emission lifetime of a system as a function of temperature and
fitting the data to a Boltzmann model will give the energy difference between two
or more states in thermal equilibrium; fitting to an Arrhenius type model can give
activation energy of a quenching reaction.

Variable temperature emission

lifetime measurements (Figure 2-3) were a critical piece of evidence that
Harrigan and Crosby used to identify that the the emissive 3MLCT state in
[Ru(bpy)3]2+ is actually a set of three closely spaced triplet states.5 The same
experimental technique has since been applied to additional ruthenium and other
d6 transition metal complexes in order to study the energy splitting within the
emissive 3MLCT state and what effect it has on the photophysical properties of
the complexes.6-9

24

Lifetime (!sec)

200

"##(!sec)
0.6

Calculated
Observed

160
-1

79 cm

120
80

19

10

-1

217

0

20

10

30

40

10 cm

50

60

Temperature (K)

Figure 2-3: Plot of emission lifetime vs temperature for [Ru(bpy)3](SO4)2 doped
in [Zn(bpy)3](SO4)2 (1:1000). The solid line the result of fitting the data to a
modified Boltzmann type model. The energy level diagram to the right shows the
energy separations and intrinsic emission lifetimes resulting from the fit.
Reproduced from reference 5.

Time-resolved emission is a valuable tool for identifying and quantifying
quenching of an emissive state by electron or energy transfer. Emission lifetimes
alone are not sufficient to distinguish between an energy or electron transfer
quenching processes but by comparison to an appropriate non-quenched model
complex the rate of the quenching process can be determined. An example from
our group compares a series of iron(III)-rhenium(I) donor-acceptor complexes of
the general formula [Fe(pyacac)3(Re(bpy')(CO)3)3]3+ (pyacac = 3-(4-pyridyl)acetylacetonate)

with

their

analogous

aluminum(III)-rhenium(I)

control

complexes.2 In the aluminum complexes, the emissive state of the rhenium(I) is
unquenched because there is no possibility of energy or electron transfer from
the ReI to the AlIII. Emission lifetime measurements on these control complexes

25

50

Intensity (au)

Intensity (cps)

40
30
20
10
0
0

400

800
1200
Time (ns)

1600

Time (ns)

Figure
2-4:
Time-resolved
emission
data
for
[Fe(pyacac)3
(Re(deeb)(CO)3)3](OTf)3 (deeb = 4,4'-diethylester-2,2'-bipyridine). The emission
was monitored at λprobe= 624 nm following excitation at λpump = 430 nm. The
solid red line corresponds to a fit to a single-exponential decay model with τobs =
2.5 ± 0.1 ns. The inset shows time-resolved emission data for the corresponding
model complex, [Al(pyacac)3(Re(deeb)(CO)3)3](OTf)3 with τobs= 235 ± 20 ns.
Reproduced from reference 2.

give the intrinsic radiative and non-radiative decay rates of the emissive rhenium
center in the absence of any quenching kinetics (Figure 2-4).

In the iron

complexes the emissive state of the ReI is quenched by the FeIII center. The
observed emission decay rates of the iron(III)-rhenium(I) complexes were
collected by time correlated single photon counting emission spectroscopy
(TCSPC) due to their short lifetimes.

The kinetics in the iron(III)-rhenium(I)

complexes are a combination of the intrinsic decay kinetics of the rhenium center
and the quenching kinetics introduced by replacing aluminum with iron, i.e. kobs =

26

kr + knr + kq.

By using the values of kr and knr from the aluminum model

complexes, the rate of quenching in the iron(III)-rhenium(I) complexes can be
calculated.
The idea of using time-resolved emission to monitor quenching reactions has
been used by the Gray group, among others, as a method to probe protein
folding dynamics.10 This works by attaching a fluorophore and quencher onto a
protein such that when the protein is unfolded, the two are far apart and the
fluorophore is unquenched, but when the protein folds, the fluorophore and
quencher are close to each other and the fluorophore is quenched. Measuring
the the emission lifetime and intensity of the fluorophore over time then gives
kinetic information about protein folding.

Pulsed
Xe Arc
Lamp

Shutter

Monochromator
Sample
PMT

Photodiode
Trigger
Digital Delay
Generator

Shutter

Nd:YAG
Q-switched
laser

Oscilloscope

Figure 2-5: An example optical layout of a transient absorption spectrometer. A
pulsed Nd:YAG laser excites the sample and a pulsed Xe arc lamp probes the
sample. Shutters and a digital delay generator control the timing of the laser and
white light pulses. The intensity of the probe beam is measured at a single
wavelength before and after the sample is excited. A photodiode trigger
synchronizes data collection with sample excitation.

27

2.3 Time-Resolved Absorption Spectroscopy
2.3.1 Experimental Setup
Time-resolved absorption spectroscopy, also referred to as transient
absorption or flash photolysis, can be used with nearly any sample unlike timeresolved emission, which can only be used on emissive samples. Much of the
transient absorption experimental setup is similar to that of the time-resolved
emission. An example of a typical time-resolved absorption spectrometer can be
seen in Figure 2-5.

Just like in the time-resolved emission experiment, the

sample is excited with a pulsed laser, either alone or coupled with an OPO. A
white light source, often a xenon arc lamp, is used as the probe beam.
Generally, but not always, the arc lamp is pulsed. This dramatically increases
the lamp output, especially in the UV region of the spectrum, but it limits the
frequency of data collection. The typical Q-switched Nd:YAG laser operates at
10 Hz whereas the typical arc lamp pulser operates at 1 Hz. Therefore, in this
example, only 1 out of every 10 laser pulses can be used for data collection.
Shutters are used to block the unused pulses from hitting the sample, and a
digital delay generator is used to synchronize the laser Q-switch, arc lamp pulser,
and shutters.
The white light probe beam passes through the sample at 90° to the pump
beam to minimize scatter and is focused into a monochromator (or spectrograph
for full spectra) and the change in intensity of the white light is monitored in real
time at a single wavelength. A photodiode trigger is used to synchronize data
collection and recording to the laser excitation pulse. Detection for time-resolved
absorption spectroscopy is more challenging than in time-resolved emission.
Instead of measuring a small signal against a dark background, transient
absorption measures small differences in large signals.

While this makes

transient absorption far less sensitive that time-resolved emission, it also makes

28

the experiment much less susceptible to trace impurities. As with time-resolved
emission, the spectra can be collected in single wavelength traces by using a
monochromator and PMT or other similar detector or the entire spectrum an be
collected as a function of time using a spectrograph and array detector such as a
CCD. With single wavelength decay traces from a PMT detector, a full spectrum
can be constructed by plotting the initial amplitude of each decay trace, as
determined by a single exponential fit, versus wavelength. This only works when
all the experimental conditions are kept constant (vide infra).

2.3.2 Analysis and Interpretation of Time-Resolved Absorption Spectra
In transient absorption spectroscopy, the intensity of the transmitted probe
beam is measured without the pump beam present to give baseline (ground
state) transmittance of the sample, I0. Then the intensity of the probe beam is
measured as a function of time after sample excitation to give the transient
transmittance of the sample, It.

In wavelength regions where the sample is

emissive, an emission decay trace, Iem, is collected by blocking the probe beam
and exciting the sample with the pump beam, as in the time-resolved emission
experiment.

The change in absorbance for the sample is calculated from

Equation 2-3.
! A = " log

(I t " I em )
I0

2-3

The magnitude and sign of the ΔA signal depends on Equation 2-4, which is
! A= ! ! · b ·[c] gs · "ex

2-4

derived from Beer's Law, where ΔA is the change in absorbance of the sample
before and after excitation, Δε is the difference between the ground state and
excited state extinction coefficients, b is the pathlength, [c]gs is the ground state

29

concentration, and ηex is the fraction of ground state molecules that are excited
into the excited state.

The ground state sample concentration and the cell

pathlength are both constant over the course of an experiment. In order to get
any chemical information from a transient absorption spectrum the ηex must also
stay constant over the course of the experiment.

Many factors go into ηex.

Some of the most important are the extinction coefficient of the sample at the
excitation wavelength, the power of the excitation beam, and the overlap
between the pump and probe beams. The extinction coefficient of the sample at
the excitation wavelength represents the probability that a molecule will absorb a
photon and become excited. This is a fundamental property of the molecule, so
while it plays a role in determining ηex, it cannot play a role in the ηex varying
over the course of an experiment. The power of the excitation beam and the
overlap between the pump and probe beams are the major contributers to
variation in ηex so it critical to keep these two constant during the course of an
experiment.

!ES >#!GS
##########

ES

!

ES - GS

!##########= #! GS
ES

"!

GS

Wavelength

Wavelength

!##ES<#!GS
########

Figure 2-6: Cartoon showing a theoretical ground state absorption (GS) and
excited state absorption (ES) (left graph). The transient absorption experiment
measures the difference in the extinction coefficients of these two states (right
graph), leading to the possibility of both positive and negative features in the
transient spectrum. Reproduced from reference 11.

30

All other things being constant, changes in the magnitude and sign of ΔA
across a spectrum reflect changes in Δε.

Figure 2-6 shows an illustrative

example of how changes in relative extinction coefficients of the ground and
excited states of a sample effect the observed transient absorption. Areas of the
spectrum where the extinction coefficient of the excited state is greater than that
of the ground state the transient absorption spectrum will have a net absorption.
Areas of the spectrum where the ground state extinction coefficient is larger than
the excited state extinction coefficient transient absorption spectrum will have a
net negative feature, or bleach.

Areas of the spectrum where the extinction

coefficents of the ground state and excited state are exactly the same the
transient absorption spectrum will be zero (an isosbestic point).
If the excited state that is being probed by transient absorption is charge
transfer in nature, then spectroelectrochemistry is a useful tool for the qualitative
prediction or interpretation of the charge transfer differential absorption spectrum.
For a metal complex [M(L)6]n+, a charge transfer transition can be written as
[M(n+1)+(L)5(L-)]n+ for a metal-to-ligand charge transfer transition (MLCT) or
[M(n-1)+(L)5(L+)]n+ for a ligand-to-metal charge transfer transition (LMCT).

In

both of these cases, the excited state absorption qualitatively will be a
superposition of the absorption spectrum of the oxidized (or reduced) metal and
the reduced (or oxidized) ligand. This technique can be used to identify features
in the transient absorption spectrum12 or as a method to determine the nature of
the long lived excited state.13 Further discussion and examples of this can be
found in Chapter 3.

2.3.3 Examples
Time-resolved absorption spectroscopy is a more versatile experiment than
than time-resolved emission. Time-resolved absorption can be used to detect

31

"dark states", non emissive states, as well as products of photochemical
reactions.

Steiner and coworkers use transient absorption as well as time-

resolved infrared absorption spectroscopy to monitor isomerization and ligand
loss in Fisher carbene complexes with the formula (CO)5W=C(OMe)Ph.14 By
monitoring the transient absorption spectra of the carbene complex in the
presence of CO or other coordinating ligand the authors were able to determine
that CO loss and isomerization are two competing processes out of the excited
carbene rather than occurring stepwise as was previously thought (Figure 2-7).

3

!-co

5

!1.7 %

3

!

iso
!30 %

3

4

4

L = solvent (n-hexane)
nitrogen, water,
acetonitrile, olefin

3
5

irr

irreversible decay
for L = olefin

3
5

Figure 2-7: Reaction pathways available for the photoexcited Fisher carbene
complex (CO)5W=C(OMe)Ph as determined by time resolved absorption
spectroscopy. The excited state of the carbene can decay non-radiatively back
to the ground state, loose a CO ligand to form the tetracarbonyl complex, or
isomerize from syn (CZ) to anti (CE). Reproduced from reference 14.

32

The experiments showed that CO loss occurred from a higher energy ligand field
state. That ligand field state relaxes rapidly to the lowest energy excited state
which is charge transfer in nature.

This rapid deactivation if the ligand field

states explains the low yield for CO loss. The lowest energy charge transfer
state is the state which undergoes the isomerization reaction.
Time-resolved absorption spectroscopy can be especially useful in identifying
electron transfer reactions and detecting charge separated intermediates.
Bernhardt, Hauser, and coworkers have used nanosecond and femtosecond
transient absorption spectroscopy to study the metal-to-metal charge transfer
(MMCT) excited state of a cyano-bridged, mixed-valence dimer of the formula
[LCoIIINCFeII(CN)5]-, where L is a pentadentate macrocyclic pentaamine
ligand.15 In the ground state, both the cobalt(III) and iron(II) centers are in the
low spin state.

Photoexcitation generates the low spin MMCT excited state,

CoII(l.s.)-FeIII(l.s.), which is too short lived to be detected on the nanosecond
timescale. The MMCT state can undergo intersystem crossing to form CoII(h.s.)FeIII(l.s.), which relaxes back to the low spin ground state on a nanosecond
timescale.
In an energy transfer reaction (Equation 2-5) the donor returns to the ground
state and the acceptor is in an excited state. In an electron transfer reaction

D* + A

D* + A

Energy
Transfer

Electron
Transfer

D + A*

+

-

2-5

2-6

D +A

(Equation 2-6) the donor is oxidized and the acceptor is reduced. Time-resolved
absorption spectroscopy of an energy transfer reaction will show the complete
decay of the donor excited state absorption features and the (possible) growth of

33

acceptor excited state absorption features. In the case of an electron transfer,
reaction time-resolved absorption spectroscopy will show only partial recovery of
the donor excited state absorption features, as shown in Figure 2-8. Previous
work from our group has used nanosecond transient absorption spectroscopy to
differentiate between an electron and energy transfer quenching mechanism in a
bimolecular quenching reaction between ruthenium(II) polypyridyl complexes and
iron(III) dimers.16 Figure 2-8 shows the change in absorbance versus time at

0.02

Change in Absorbance (440 nm)

0.0

-0.02

-0.04

-0.06

-0.08

-0.10

-0.12
0.0

1.0

2.0

3.0

4.0

Time (!sec)

Figure 2-8: Plot of the excited-state differential absorbance of a CH3CN solution
containing [Ru(dmb)3]2+ and [Fe2(OH)(O2CCH3)2(Tp)2]+ at 440 nm following 10
ns excitation at 450 nm. The solid line indicates a fit to a biexponential kinetic
model. Reproduced from reference 16.
34

440 nm which is probing the loss of the MLCT absorption in the excited state.
This trace shows the partial recovery of the MLCT bleach which indicates the
quenching is an electron transfer reaction.

2.4 Time-Resolved Vibrational Spectroscopy
Two major techniques are used for nanosecond time-resolved vibrational
spectroscopy: time-resolved resonance Raman spectroscopy (TR3) and timeresolved infrared spectroscopy (TRIR). Both techniques have their advantages
and disadvantages. TR3 uses visible light to both excite and resonantly scatter
off of the excited state. Consequently, only those vibrational modes which are
vibronically coupled to the excited electronic transition will be enhanced. This
vastly simplifies the spectrum and makes interpretation much easier, but it only
shows a small subset of the entire excited state vibrational spectrum. Because
visible light is used as the probe, emission from the sample can interfere with the
data collection. This can make collecting TR3 data on highly emissive molecules
a challenge.
Time-resolved infrared spectroscopy uses visible light to excite the sample
and probes the excited state with a broadband infrared source. In this way, the
entire infrared allowed excited vibrational spectrum can be measured. This could
make interpretation of the spectra difficult, especially with heteroleptic
complexes. However, it does allow for probing vibrations that are not directly
coupled to the excited state, like the CO stretch in [Re(bpy)(L)(CO)3]+
complexes. Solvents are not completely transparent in the infrared which poses
another challenge in the collection and interpretation of TRIR spectra. Very thin
pathlength cells (0.5 or 0.2 mm are most common) are used to minimize solvent
absorptions. As a consequence of the thinner cell, much higher concentrations
of sample (~10-2-10-3 M for TRIR vs ~10-5 M for TR3) are necessary, which can

35

cause difficulty for samples with low solubility. There are several recent reviews
of time-resolved vibrational spectroscopy that discuss applications of both TR3
and TRIR17-21.

2.4.1 Experimental Setup
The time-resolved resonance Raman experiment uses visible excitation and
probe beams. There are two different ways the experiment can be done: the
one-color, transient resonance Raman (TR2), experiment or the two-color, timeresolved resonance Raman (TR3), experiment. In the TR2 experiment, a single
laser pulse is used to both excite the molecule and resonantly scatter off of that
excited state. In order for this technique to succeed, both the ground state and
the excited state have to absorb at the same wavelength.

For ruthenium(II)

polypyridyl complexes, a 355 nm pulse is generally used because both ground
state and bipyridine radical anion absorb strongly at that wavelength. Also, since
a single laser pulse is acting as both the pump and probe, there is no way to vary
the time delay between the pump and probe. Consequently, the TR2 experiment
only shows a snapshot of the excited state and doesn't give any information
about excited state dynamics.
In the TR3 experiment, two different laser pulses, from two different lasers,
are used for the pump and probe beams. An example of a TR3 spectrometer can
be seen in Figure 2-9. The major advantage of the TR3 experiment is the ability
to independently vary the pump and probe wavelengths to best suit the sample
being studied. Also, independent pump and probe pulses mean the time delay
between pump and probe pulses can be varied to get a true time resolved
spectrum and look at excited state dynamics. The clear disadvantage to TR3 is
that it requires two lasers, which adds considerable cost and complexity to the

36

Figure 2-9: Example optical layout for two-color time-resolved resonance Raman
spectrometer. There are two pulsed lasers to independently pump and probe the
sample. A pulse delay generator controls the time between the pump and probe
pulses. Raman scatter is collected at 90°, dispersed with a spectrograph and the
full spectrum is collected with a CCD array. Reproduced from reference 19.

experiment. For both the TR2 and TR3 methods, the scattered light is collected,
generally at 90° to minimize scatter, and focused into a monochromator. Using
double or triple monochromator will exclude more Rayleigh scatter and other
stray light. It will also increase the experimental resolution and allow for the
detection of lower frequency vibrations. Depending on the probe wavelength
used, notch filters can also be used to exclude Rayleigh scatter. The scattered
light is then dispersed onto an array detector such as a CCD or photodiode to
record the full resonance Raman spectrum.

37

MCT
Detector
Germanium
Filter

Laser
Quartz
Window

L3
L2

Fixed Mirror

Sample

Moving
Mirror

L1
M2
M1

IR Source

Figure 2-10: Example optical layout of a time-resolved step-scan FTIR
spectrometer. Reproduced from reference 22.

I
t

!

Figure 2-11: Data collection process in SSIR experiment. Decay traces (I vs t)
are collected at each fixed mirror position δ. After data collection, interferograms
(I vs δ) are generated for every time t, and these interferograms are used for
further data analysis. Reprinted from reference 23.

38

Nanosecond time resolved infrared spectroscopy generally uses a step-scan
IR (SSIR) spectrometer, which can be seen in Figure 2-10.

The step-scan IR

spectrometer is a modification of the standard FTIR instrument24.

In the

standard rapid scan FTIR instrument, such as the reactIR, the moving mirror in
the interferometer is constantly moving. The time resolution is determined by
how fast the mirror can move back and forth through all the positions (ms-μs
resolution). In the step-scan IR instrument, the moving mirror steps to a fixed
position and a decay trace is collected while the mirror is held in that position.
The mirror then steps to the next position and another decay trace is collected.
This is shown graphically in Figure 2-11. Using this method, the time resolution
of the experiment is limited by the temporal width of the excitation pulse and the
bandwidth of the detector and any other electronics used in data collection (~ns
resolution).
The excitation source is generally a pulsed laser source, such as a Qswitched Nd:YAG, and the probe beam is generated by a broadband source (Hg
lamp for the far IR, globar for the mid IR, or tungsten lamp for the near IR) and
Michelson interferometer. The angle between the pump and probe beams is
kept as small as practically possible (approaching collinear) to maximize the
volume of overlap between the pump and probe beams in the sample. The
probe beam is then detected by the appropriate detector such as MCT
photovoltaic for far to mid infrared or an InGaAs photodiode for the near infrared.
The detector is separated from the sample chamber by a germanium window to
block visible light such as laser scatter from reaching the detector and saturating
or possibly damaging it. Detectors used for this experiment will generally have
both a DC and AC coupled output. The DC coupled output is used to measure
the ground state spectrum of the sample. The AC coupled output contains only

39

(a)

Intensity (V)

0.5
0.0
-0.5

Fourier
Transform
500

400

300 200 100
Mirror position
(b)

3

Intensity (a.u.)

0

2
1
0

Change in Absorbance

2400

2000
-1
Energy (cm )

1600

!A = -log DC+AC
DC

(c)

0.03
0.02
0.01

0.00
2400 2300 2200 2100 2000 1900
-1

Energy (cm )

Figure 2-12: Example of data processing in the SSIR experiment. The
interferogram (a) is Fourier transformed to generate the single channel spectrum
(b). The ΔA spectrum (c) is calculated from the AC and DC coupled single
channel spectra. A single two dimensional slice of the three dimensional spectra
has been shown for simplicity.

40

the portion of the signal which changes over time, so it is used to measure the
excited state spectrum of the sample. Because the excited state signal has a
much smaller amplitude than the ground state, the AC signal is generally
amplified before it is recorded.
As mentioned above, the moving mirror is stepped to discrete positions, and
decay traces (I vs t) are collected while the moving mirror is held those fixed
positions. The number of positions will depend on the spectral range and the
resolution of the experiment. After decay traces are collected at each mirror
position δ interferograms, plots of I vs δ, are generated for each time t and the
final ΔA spectra are then calculated from these interferograms (Figure 2-12).
Each 3D interferogram (I vs δ vs t) is converted to a 3D single channel spectrum
(I vs cm-1 vs t) via a Fourier transform. It is important to note that in order to
avoid introducing artifacts when generating the single channel spectrum, the
spectrum must return to zero at both ends. Often, filters are used to isolate the
spectral region of interest and minimize the necessary data collection time. The
3D ΔA spectrum is calculated from the ground state (DC coupled) spectrum and
the excited state (AC coupled) by Equation 2-725, after accounting for any
amplification of the AC signal.
!A = " log

AC + DC

2-7

DC

Depending on the number of spectra averaged per mirror position and the
resolution and spectral window of the experiment, collecting a single full
spectrum can take thousands or tens of thousands of laser shots over an hour or
more of collection time. This means that the photochemical or photophysical
process being monitored must be repeatable and reproducible over that period of
time and that number of laser shots. Monitoring photochemical reactions by

41

step-scan IR is thus challenging, but flowing the solution to ensure that each
each laser shot is exciting fresh sample is one way to avoid problems with
photstabliity and reproducibility.

A more detailed description of the data

acquisition process for the Bruker IFS 66/S can be found in Appendix C.

2.4.2 Examples
2.4.2.1 Time-Resolved Resonance Raman
Time-resolved resonance Raman spectroscopy was one of the many pieces
of evidence that showed that the long lived excited state of [Ru(bpy)3]2+ was
localized on a single bpy ligand on the nanosecond timescale.26

Dallinger,

Woodruff, and coworkers compared the TR2 spectra of [RuII(bpy)3]2+ collected in
a one-color experiment, exciting at 355 nm, to the ground state resonance
Raman spectra of the free bipyridine radical anion and [RuIII(bpy)3]3+. They
found that the excited state spectrum of [RuII(bpy)3]2+ was essentially a
superposition of the bpy- and the [RuIII(bpy)3]3+ resonance raman spectra, which
supports describing the excited state as {[RuIII(bpy)2(bpy-)]2+}*.
compared

the

TR3

(en=ethylenediamine).

spectra

of

[RuII(bpy)3]2+

and

They also

[RuII(bpy)2(en)]2+

The shift between the ground state and excited state

bipyridine stretches was the same in both complexes, which is also evidence for
a localized excited state. If the excited state was delocalized, the shift between
the ground state and the excited state for [RuII(bpy)2(en)]2+ would be larger than
for [RuII(bpy)3]2+ because the excited electron would be delocalized over two
rather than three bipyridine ligands.
Kim, Yoon, and coworkers used TR2 to study copper(II) porphyrins in
aqueous and non-aqueous environments.27

In the excited state, copper(II)

porphyrins can form an exciplex with solvent molecules, i.e. the solvent

42

1536
1556

1436

1338
1360

1233

1079
1124

80%
70%
60%
50%
40%
30%
20%

1000

1556

1360

1238

1087
1228

10%
0%
ground
1200

1400

1600

1800

Raman Shift (cm )

Figure 2-13: Nanosecond transient resonance Raman spectra of CuIITSPP in
dioxane/ water solvent mixture. Percentages show the increasing fraction of
dioxane in the solvent mixture. Reproduced from reference 27

coordinates to the copper center to form a five-coordinate complex.

This

exciplex formation leads to a rapid deactivation of the excited state. Figure 2-13
shows the transient resonance Raman spectrum of CuIITSPP (TSPP =
tetrakis(p-sulfonatophenyl)porphyrin) collected in mixtures of water and dioxane.
In these spectra two sets of peaks are seen: one for the four-coordinate species
and one for the five-coordinate species.

43

Monitoring the ratio between the

intensities of the four- and five-coordinate species as a function of solvent
mixture allows the authors to infer relative concentrations of five-coordinate
species, and therefore relative binding energies of the different solvents to the
copper center.

2.4.2.2 Step-Scan IR
One major advantage of TRIR over TR3 is that all (infrared allowed) excited
state vibrations will appear in the spectrum rather than only the subset of
vibrations which are Raman allowed and are resonantly enhanced by coupling to
the excited electronic transition. Schoonover, Meyer, and coworkers have used
this idea to build on the work of Dallinger, Woodruff, and coworkers mentioned
above; using TRIR to assign the full mid-infrared excited state vibrational spectra
of

[Ru(bpy)3](PF6)2

and

[Ru(phen)3](PF6)2.22,28

They

compared

the

nanosecond step-scan IR spectra of [RuII(L)3]2+ to electrochemically generated
[RuIII(L)3]3+ and [RuII(L)2(L-)]+ (L = 2,2'-bipyridine or 1,10'-phenanthroline). They
also

used

heteroleptic

[Os(phen)(DAS)2]2+

(4-Etpy

complexes
=

[Re(bpy)(CO)3(4-Etpy)]+

4-ethyl-pyridine

and

DAS

=

and
1,2-bis

(diphenylarsino)ethane) as comparisons as well because the excited state is
forced to be localized to a single ligand in these examples. Like the resonance
Raman experiment discussed above, these experiments showed that the long
lived excited state in [Ru(bpy)3]2+ and [Ru(phen)3]2+ are both localized on a
single ligand. In the case of [Ru(bpy)3]2+, it was also possible to determine
which vibrational mode(s) were contributing to each observed peak

by

comparison to the previous normal coordinate analysis of the ground state IR
and resonance Raman spectra of [Ru(bpy)3]2+.29
Dattelbaum, Schoonover, and coworkers used time-resolved step-scan IR to
study a series of rhenium(I) carbonyl complexes such as the one pictured in

44

Figure 2-11 that have low energy charge separated (CS) and intraligand (IL)
states in competition with the 3MLCT state.30 This is one example of a study
which would not be possible with TR3 because the CO vibrations are not directly
coupled to a charge transfer excitation and are therefore not enhanced in

0.010
late times

Delta Absorbance

early times
0.005

0.000

-0.005

-0.010
1900

2000

1950
-1

2050

Wavenumber (cm )

Figure 2-14: TRIR spectra of fac-[Re(dppz)(CO)3(py-PTZ)]+* (pictured above)
separated into early and late time average spectra. At early times (<300 ns,
dashed line), the charge separated state dominates. At later times (>300 ns, solid
line), the IL state dominates and the ground-to-excited-state shifts in ν(CO)
decrease. Reproduced from reference 30

45

resonance Raman or TR3 spectroscopy. While the CO vibrations are not directly
coupled to the excited state, they are very sensitive to the electron density at the
rhenium center which makes them a very useful indirect reporter of excited state
dynamics.
Figure 2-14 shows the early and late time TRIR spectra of fac[ReI(CO)3(dppz)(py-PTZ)]+ (dppz = dipyrido-[3,2-a:2′,3′-c]phenazine and py-PTZ
= 10-(4-picolyl)phenothiazine). The early time spectra shows the CO vibrations
in the CS state, where the excited electron is transfered to the phenothazine.
Because of the methylene linkage, there is little overlap between the
phenothiazine and the rhenium center and the rhenium is best described as ReII.
The later time spectrum shows the CO vibrations in the IL state of the phenazine
ligand. Because the phenazine ligand is directly bound to the rhenium center,
there is overlap and mixing between the phenazine and rhenium orbitals and to a
small extent the IL state is also delocalized onto the rhenium center, making it
not fully oxidized to ReII. This is reflected in the smaller change in CO stretching
frequency vs ground state than is seen in the CS state.

2.5 Time-Resolved X-ray Spectroscopy
2.5.1 Experimental Setup
Unlike the other techniques discussed in this chapter, there is not a
commercially

available

bench-top

instrument

for

time-resolved

X-ray

spectroscopy. Like the other techniques discussed, the excitation source for the
time resolved X-ray experiment is usually a pulsed laser source. The difficulty in
the time resolved X-ray experiment lies in the probe source. In order to achieve
short data averaging times and reasonable time resolution, the high brightness
and temporally short pulses of a synchrotron X-ray source are necessary. An

46

example of a time resolved X-ray spectrometer can be seen in Figure 2-15.
There are two different methods of collecting time-resolved X-ray
spectroscopy. The diffraction of X-rays through a crystal is described by Bragg's
Law (Equation 2-8), where n is an integer, λ is the wavelength, d is the space
n! = 2d sin"

2-8

between layers in the crystal lattice, and θ is the angle of the incoming X-rays. In
the monochromatic method, λ is constant and θ is varied by rotating the crystal.
In the polychromatic (Laue) method the crystal is kept in a fixed position and
probed with a polychromatic X-ray beam.

Laser
Laser Beam

CCD Detector

Mirror

Pin diode X-Ray Detector
X-Ray Beam

Luminescence
Detector

Photodiode
Chopper

He/Ne Laser

Laser trigger&
Synchronization
Circuits

Figure 2-15: Schematic diagram of a time-resolved X-ray diffraction experiment
at the Advanced Photon Source, Argonne National Lab. Reproduced from
reference 31.

47

The monochromatic method is the same as method that is used in laboratory
single crystal diffractometer and is most often used in small molecule
spectroscopy. A monochromatic X-ray beam probes the sample crystal and the
crystal is rotated to collect a sufficient number of reflections. In order to do this
experiment in a time resolved manner, a pump probe technique, similar to other
experiments discussed previously, is used. Synchrotron X-ray sources have very
high repetition rates (100's of kHz).

In order to most efficiently use the

synchrotron source, the excitation source is a high repetition rate laser like the
Nd vanadate laser, which can operate at a 20 kHz repetition rate with sufficient
power to generate a significant excited state population in the crystal. Even with
the Nd vanadate laser, there is still a large mismatch between the repetition rate
of the laser excitation and the synchrotron source. Consequently shutters or
choppers are used to select the X-ray pulses and synchronize them with the
laser excitation. The data collection process is similar to that in SSIR. At each
position the crystal is excited with the laser pump, probed with the X-ray pulse,
and the diffracted X-rays are collected by CCD detector. Then, the crystal is
probed without laser excitation as a background. The crystal then rotates to the
next position and the process repeats. Much like SSIR, this experiment requires
that the structural change being monitored is reversible and reproducible over
many pump-probe cycles. Crystal integrity creates an additional difficulty in time
resolve X-ray spectroscopy. Repeated structural changes can put strain on the
crystal lattice and cause the long range order of the crystal to break down.
The Laue method is an alternate method that is most often used for
macromolecular structures such as proteins or other biological molecules. This
technique uses a polychromatic X-ray beam and keeps the crystal stationary
during data collection. This results in an overall shorter collection time, which
minimizes damage to fragile biological crystals. The Laue method is also a good

48

option for irreversible structural changes or photochemical reactions because a
complete set of reflections can be collected from a single shot of the pump and
probe beams. There are several recent reviews that cover both small molecule
and macromolecular time resolved X-ray spectroscopy.31-34

2.5.2 Examples
Moffat and coworkers used the Laue method to monitor the structural
changes that accompany substrate binding in myoglobin.35 When crystals of a
carbon monoxide complex of myoglobin, where the CO is bound axially to the
iron heme center, are photoexcited the CO dissociates from the iron center.
Over time, the CO ligand re-coordinates to the iron heme center and the
structural changes that occur as the CO re-coordinates are monitored by X-ray
probe. The structure was probed at six different time delays (4 ns, 1 μs, 7.5 μs,
50.5 μs, 350 μs, and 1.9 ms). The most prominent features seen in the electron
density difference plots was the loss of CO at the iron center, which recombines
in ~100 μs, and the iron moving out of the heme plane and towards the axially
coordinated histidine ligand when the CO ligand dissociates. The iron returns to
the plane of the heme ring as the CO ligand re-coordinates.
Coppens and coworkers used time resolved X-ray spectroscopy to observe
structural changes that occur upon photexcitation in a copper(I) phenanthroline
complex, [Cu(dmp)(dppe)]+ (dmp = 2,9-dimethyl-1,10-phenanthroline and dppe
= 1,2-bis(diphenylphosphino)-ethane).36

In the ground state, four coordinate

copper(I) complexes favor tetrahedral geometries. In an MLCT excited state the
copper(I) is (at least partially) oxidized to copper(II), which tend to favor square
planar geometries. This means that once a copper(I) complex is excited into an
MLCT excited state it will distort and flatten to a more square planar geometry.
Figure 2-16 shows an overlay of the ground state and distorted excited state

49

structures in [Cu(dmp)(dppe)]+. The structures show the rocking and the out of
plane bending of the phenanthroline ligand in the excited state.

Cu1
Cu2

Figure 2-16: ES geometries (orange) of [Cu(dmp)(dppe)]+ superimposed on the
GS of the complex (Cu, green; C, black; P, purple; N, blue). The different views
are shown to illustrate the change in rocking distortion (left) and the displacement
of the phenanthroline ligand from its GS plane (right) upon excitation.
Reproduced from reference 36

2.6 Concluding Comments
Nanosecond time resolved spectroscopy can be a valuable and versatile tool
for any inorganic chemist.

Unlike ultrafast spectroscopic techniques, the

nanosecond spectroscopic techniques do not require specialized training and
expertise. Commercial benchtop spectrometers are available for many of the
techniques

discussed,

with

the

exception

of

the

time-resolved

X-ray

spectroscopy. Nanosecond time resolved spectroscopic techniques can be used
to investigate a wide range of inorganic systems and give insight to many
chemical questions.

50

REFERENCES

51

2.7 References
(1)

Soler, M.; McCusker, J. K. J. Am. Chem. Soc. 2008, 130, 4708-4724.

(2)

Knight, T. E.; Guo, D.; Claude, J. P.; McCusker, J. K. Inorg. Chem. 2008,
47, 7249-7261.

(3)

Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys. Chem.
1986, 90, 3722-3734.

(4)

Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J. Chem.
Soc. 1997, 119, 8253–8268.

(5)

Harrigan, R. W.; Crosby, G. A. J. Chem. Phys. 1973, 59, 3468.

(6)

Elfring, W. H.; Crosby, G. A. J. Am. Chem. Soc. 1981, 103, 2683-2687.

(7)

Yersin, H.; Gallhuber, E. J. Am. Chem. Soc. 1984, 106, 6582-6586.

(8)

Yersin, H.; Humbs, W.; Strasser, J. Coord. Chem. Rev. 1997, 159, 325-358.

(9)

Hofbeck, T.; Yersin, H. Inorg. Chem. 2010, 49, 9290-9299.

(10) Lyubovitsky, J. G.; Gray, H. B.; Winkler, J. R. J. Am. Chem. Soc. 2002, 124,
5481-5485.
(11) Schrauben, J. N. Ph.D. Dissertation, Michigan State University, East
Lansing, MI, 2010
(12) Damrauer, N. H.; McCusker, J. K. J. Phys. Chem. A. 1999, 103, 8440–
8446.
(13) Berger, R. M.; McMillin, D. R. Inorg. Chem. 1988, 27, 4245–4249.
(14) Gut, H.-P.; Welte, N.; Link, U.; Fischer, H.; Steiner, U. E. Organometallics.
2000, 19, 2354-2364.
(15) Macpherson, B. P.; Bernhardt, P. V.; Hauser, A.; Pages, S.; Vauthey, E.
Inorg. Chem. 2005, 44, 5530-5536.
(16) Weldon, B. T.; Wheeler, D. E.; Kirby, J. P.; McCusker, J. K. Inorg. Chem.
2001, 40, 6802–6812.

52

(17) Schoonover, J. R.; Bignozzi, C. A.; Meyer, T. J. Coord. Chem. Rev. 1997,
165, 239–266.
(18) Schoonover, J. R.; Strouse, G. F. Chem. Rev. 1998, 98, 1335–1355.
(19) Phillips, D. L.; Kwok, W. M.; Ma, C. An Introduction to Time-Resolved
Resonance Raman Spectroscopy and Its Application to Reactive
Intermediates; Wiley: 2007; 123-182.
(20) Butler, J. M.; George, M. W.; Schoonover, J. R.; Dattelbaum, D. M.; Meyer,
T. J. Coord. Chem. Rev. 2007, 251, 492-514.
(21) Browne, W. R.; McGarvey, J. J. Coord. Chem. Rev. 2007, 251, 454–473.
(22) Omberg, K. M.; Schoonover, J. R.; Treadway, J. A.; Leasure, R. M.; Dyer,
R. B.; Meyer, T. J. J. Am. Chem. Soc. 1997, 119, 7013–7018.
(23) Johnson, T. J.; Zachmann, G. Introduction to Step-Scan FTIR; Bruker
Optics: Billerica, MA, 2000; 1-95.
(24) Uhmann, W.; Becker, A.; Taran, C.; Siebert, F. Appl. Spectrosc. 1991, 45,
390–397.
(25) Sun, H.; Frei, H. J. Phys. Chem. B. 1997, 101, 205–209.
(26) Bradley, P. G.; Kress, N.; Hornberger, B. A.; Dallinger, R. F.; Woodruff, W.
H. J. Am. Chem. Soc. 1981, 103, 7441-7446.
(27) Jeoung, S. C.; Kim, D.; Cho, D. W.; Yoon, M. J. Phys. Chem. 1996, 100,
3075-3083.
(28) Omberg, K. M.; Schoonover, J. R.; Bernhard, S.; Moss, J. A.; Treadway, J.
A.; Kober, E. M.; Dyer, R. B.; Meyer, T. J. Inorg. Chem. 1998, 37, 3505–
3508.
(29) Mallick, P. K.; Danzer, G. D.; Strommen, D. P.; Kincaid, J. R. J. Phys.
Chem. 1988, 92, 5628–5634.
(30) Dattelbaum, D. M.; Omberg, K. M.; Hay, P. J.; Gebhart, N. L.; Martin, R. L.;
Schoonover, J. R.; Meyer, T. J. J. Phys. Chem. A. 2004, 108, 3527–3536.
(31) Coppens, P.; Vorontsov, I. I.; Graber, T.; Gembicky, M.; Kovalevsky, A. Y.
Acta Crystallogr. A. 2005, 61, 162-172.

53

(32) Moffat, K. Acta Crystallogr. A. 1998, 54, 833-841.
(33) Schlichting, I. Acc. Chem. Res. 2000, 33, 532-538.
(34) Coppens, P.; Benedict, J.; Messerschmidt, M.; Novozhilova, I.; Graber, T.;
Chen, Y. S.; Vorontsov, I.; Scheins, S.; Zheng, S. L. Acta Crystallogr. A.
2010, 66, 179-188.
(35) Srajer, V.; Teng, T. Y.; Ursby, T.; Pradervand, C.; Ren, Z.; Adachi, S.;
Schildkamp, W.; Bourgeois, D.; Wulff, M.; Moffat, K. Science. 1996, 274,
1726-1729.
(36) Vorontsov, I. I.; Graber, T.; Kovalevsky, A. Y.; Novozhilova, I. V.; Gembicky,
M.; Chen, Y. S.; Coppens, P. J. Am. Chem. Soc. 2009, 131, 6566-6573.

54

Chapter 3: Synthesis and Spectroscopic Characterization of CyanoSubstituted Bipyridyl Complexes of Ruthenium(II)

3.1 Introduction
Photo-induced charge separation is the physical phenomenon underlying
virtually all schemes geared toward the conversion of light into chemical,
electrical, and/or mechanical energy. For example, in photosynthesis absorption
of light by either the light harvesting complex or the reaction center itself sets the
stage for the creation of a trans-membrane potential gradient: this ultimately
provides the chemical energy for ATP synthase.1,2 In materials science, where
semiconductors are the functional element of the vast majority of commercial
photovoltaic devices, irradiation leads to the formation of electron-hole pairs that
establish an electric potential. The well-known Grätzel cell, which has arguably
helped to redefine research on photovoltaics over the past two decades,3-6
essentially combines molecular and semiconductor photophysics to create a
functional device with overall conversion efficiencies that now exceed 11%7,8. In
all of these cases, absorption of light and the resulting redistribution of energy
within the chromophoric system represent the primary steps for solar energy
conversion.
Charge separation is typically effected in a molecular system through chargetransfer excited states, in which photon absorption causes charge redistribution
within the chromophore: maintaining, amplifying, or, in the least favorable
circumstances, destroying the resulting chemical potential depends on dynamics
that occur within the chromophore immediately following the absorptive event.
Transition metal complexes having intense charge-transfer features have long
been exploited to achieve the goal of creating reactive excited states for
photochemical energy conversion.9,10 Formation of the desired reactive state

55

then relies on a sequence of events: excitation into an upper level having
significant radiative coupling to the ground state, followed by relaxation to the
lower energy state(s). A host of dynamic processes are potentially involved in
this process, among them intersystem crossing (ISC), internal conversion (IC),
vibrational relaxation (VR) and/or intramolecular vibrational redistribution (IVR),
as well as solvation dynamics. In general, all of these fundamental processes
can be categorized as either radiative or non-radiative in nature. With regard to
intramolecular dynamics, with the notable exception of certain highly emissive
compounds,11-13 energy dissipation in the vast majority of transition metal-based
charge-transfer systems occurs primarily through vibrational relaxation. Despite
being the dominant contribution to excited-state relaxation – and in practical
terms an important competing process for the efficient conversion of absorbed
energy into more productive reaction pathways – there is relatively little known
concerning the mechanism(s) by which vibrational relaxation occurs.
Probing vibrational relaxation in transition metal complexes is not always a
straightforward process.

Often times information about vibrational relaxation

from the initially excited Franck-Condon state to the long lived excited state is
inferred from transient electronic absorption spectroscopy.14 The number and
energy of the possible Franck-Condon transitions changes as the molecule goes
from a vibrationally hot excited state to a thermalized excited state, which can
lead to changes in bandshape and/or peak position in the electronic absorption
spectra. These changes are usually seen as a red shifting and narrowing of an
absorption feature over time, but the nature of the change depends on the
relative displacement and slope of the two excited state potential energy
surfaces. Infrared spectroscopy is more direct way to probe the vibrational state
of an electronic excited state. In infrared spectroscopy, vibrational relaxation
generally manifests itself as a narrowing and blue shifting of the excited state

56

vibrational peak.15,16 This is because the Franck-Condon excitation creates a
distribution of vibrationally excited states which all relax to a single νʼ = 0
vibrational state which causes the spectrum to narrow and the anharmonic
nature of the potential surface means the vibrational spacing is smaller at the top
of the potential well and larger at the bottom, leading to a blue shift.

Browne,

McGarvey, and coworkers have used time-resolved resonance Raman to monitor
the formation of the thermalized 3MLCT state in [Ru(bpy)3]2+ by monitoring the
growth of ring stretching and bending vibrations assigned to the bipyridyl radical
anion.17,18
While the results of Browne, McGarvey, and coworkers are significant
because it is the first time that vibrational relaxation has been directly probed in a
ruthenium(II) polypyridyl complex, using vibrations in the fingerprint region is less
than ideal.

The disadvantage of using these vibrations to probe vibrational

relaxation in the excited state is that there are many overlapping vibrational
modes in this region of the spectrum, which can mask the subtle changes in the
peak shape and energy that accompany vibrational relaxation. In the fingerprint
region interference from solvent vibrations can also cause significant
interference.

In the case of heteroleptic complexes, the complexity of the

vibrational spectrum in the fingerprint region dramatically increases. This makes
even distinguishing the vibrations of the different ligands difficult at best.19 An
alternative is to use an infrared tag to probe the vibrational state of an excited
chromophore. The ideal infrared tag is a functional group which has a vibration
which is spectrally isolated, is not overlapping with other vibrations of the
molecule or solvent, has a significant change in frequency from ground to excited
state, and is located where the charge transfer excited state is localized. Many
studies have used the CO stretch in metal carbonyl complexes as an infrared
tag, however many of these complexes are not photostable and the the CO

57

stretch is used to monitor photodisassociation reactions.20-22 In cases where no
photochemistry is occurring, such as in Re(bpy)(L)(CO)3 type complexes, the CO
is an ancillary ligand, and is not directly coupled to the bpy MLCT excited state23.
The cyano group is a less frequently used alternative to the carbonyl group. The
CN stretching frequency is sensitive to local environment and well isolated from
other molecular and solvent vibrations, so cyano substituted amino acids have
been used to probe secondary structure and environment in proteins.24 The
terminal CN in metal dimers can be used to monitor electron transfer
reactions.25-27 Vlček and coworkers used the CN stretch of 4-cyano-pyridine to
assign the nature of the charge transfer transitions in W(CO)5(4-CN-py).15
Combining the features of the charge-transfer character of RuII-polypyridyls
with the vibrational properties of 4-cyanopyridine takes advantage of benefits of
both systems. Specifically, we felt that incorporating CN groups into the bipyridyl
ligand would introduce a vibrationally well-isolated infrared tag into the
chromophore. Moreover, placing the CN group at the 4 and/or 4ʼ positions of the
rings would result in direct coupling of the CN group to the π* system of the
bipyridine ring; the presence of an electron in the π* orbital should result in a
significant shift in the CN stretching frequency, thereby allowing us an
unambiguous probe for vibrational processes associated with the charge-transfer
state. This chapter describes the synthesis and spectroscopic characterization of
such a system and provides the foundation for the detailed study of vibrational
relaxation dynamics in a prototypical MLCT chromophore which is discussed in
Chapter 4.

3.2 Experimental Section
3.2.1 General.
All chemicals and solvents were obtained from Fisher or Aldrich Chemical Co.

58

and used without further purification unless otherwise stated.

RuCl3•xH2O,

tris(benzylidineacetone)dipalladium(0) (Pd2(dba)3), and diphenylphosphinoferrocene (dppf) were purchased from Strem Chemicals, Inc. and C18 reversed
phase silica gel was purchased from SiliCycle. NMR spectra were collected on
Varian Inova-300 (300 MHz) or Varian UnityPlus-500 (500 MHz) spectrometers.
Ground state infrared spectra (4000-400 cm-1) were measured as KBr pellets
using a Mattson Galaxy series 3000 FTIR spectrophotometer. Mass spectra
were obtained through the Michigan State University Mass Spectrometry Facility.
Elemental analyses were obtained through the analytical facilities at Michigan
State University or from Columbia Analytics.

Solvent included in calculated

elemental analysis percentages was included to give the best fit based on
solvents identified by NMR and IR.

NO2
i

N

N

N
N
O O

N
N
O O

5,5'-dimethylbipyridine

dmb-N-oxide

Cl

NO2

ii

Cl

Cl

NO2-dmb-N-oxide

iii

Cl

iv

N

N
N
O O

N

Cl-Me-bpy

Cl-dmb-N-oxide

Figure 3-1: Synthesis of 4,4'-dichloro-2,2'-bipyridine. (i) Acetic acid and 30%
hydrogen peroxide, refluxed. (ii) Sulfuric acid and fuming nitric acid, gentle
heating. (iii) Acetyl anhydride refluxed in acetic acid (iv) PCl3 refluxed in
acetonitrile
59

3.2.2 Ligand Synthesis
5,5ʼ-dimethyl-2,2ʼ-bipyridine-N,Nʼ-dioxide
dmb-N-oxide28
5,5'-dimethyl-2,2'-bipyridine (5.0 g) was dissolved in 35 mL of glacial acetic
acid and 7.0 mL of 30% hydrogen peroxide were slowly added. After heating the
solution to 70-80 °C for 4 hours, an additional 4.0 mL of 30% hydrogen peroxide
were added and the reaction mixture was heated overnight. The product was
isolated by pouring the cooled solution over approximately 400 mL of acetone
and reducing the volume under vacuum until a white precipitate formed. The
white precipitate was collected by vacuum filtration, rinsed sparingly with acetone
and ether, and then dried further under vacuum. Yield: 4.76 g (81%). 1H NMR
(300 MHz, D2O): δ 8.19 (s, 2H), 7.57 (d, J=8.0 Hz, 2H), 7.14 (d, J=8.0 Hz, 2H),
2.35 (s, 6H).

5,5ʼ-dimethyl-4,4ʼ-dinitro-2,2ʼ-bipyridine-N,Nʼ-dioxide
NO2-dmb-N-oxide 29,30
4.5 g of dmb-N-oxide was dissolved in 21 mL of sulfuric acid and 9.0 mL of
fuming nitric acid were slowly added to the solution. The solution was gently
heated to 90-100° C for 4 hours. During the course of the reaction the mixture
became yellow and a brown gas formed above the solution. The solution was
cooled to room temperature and slowly poured over ice formed by mixing 75 mL
of water with excess liquid nitrogen. Additional liquid nitrogen was added, while
stirring, to form a yellow slush. As the slush melted, a yellow precipitate was
collected by vacuum filtration, rinsed thoroughly with water and ether, and then
dried under vacuum. Yield: 3.75 g (59%). 1H NMR (500 MHz, DMSO-d6): δ 8.65
(s, 2H), 8.53 (s, 2H), 2.58 (s, 6H).

60

4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine-N,Nʼ-dioxide
Cl-dmb-N-oxide 31
NO2-dmb-N-oxide (3.75 g) was suspended in 55.0 mL of glacial acetic acid
and 45.0 mL of acetyl chloride. The suspension was refluxed for two hours,
during which time the solution became a clear yellow color. The solution was
cooled to room temperature and slowly poured over approximately 500 mL of ice.
The resulting clear yellow solution was neutralized with concentrated sodium
hydroxide, resulting in the formation of a white precipitate. The white solid was
collected by vacuum filtration, rinsed with water and ether, and dried under
vacuum.

Yield: 3.13 g (90%). 1H NMR (300 MHz, DMSO-d6): δ 8.45 (s, 2H),

7.86 (s, 2H), 2.31 (s, 6H).

4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine
Cl-Me-bpy 32
Cl-Me-bpy-N-oxide (3.13 g) was suspended in 460 mL of dry acetonitrile and
21.0 mL of phosphorus trichloride were slowly added to the mixture. The
suspension was refluxed under nitrogen for four hours, after which the solution
became a clear yellow color. The solution was cooled to room temperature and
slowly poured over approximately 500 mL of ice.

The solution was made basic

(pH >11) with the addition of concentrated sodium hydroxide resulting in the
formation of a white precipitate.

The white solid was collected by vacuum

filtration, rinsed with water and dried under vacuum.

Concentrating the filtrate

yielded additional product. Yield: 2.57 g (92%). 1H NMR (500 MHz, CDCl3): δ
8.47 (s, 2H), 8.37 (s, 2H), 2.42 (s, 6H). 13C NMR (500 MHz, CDCl3): δ 154.51,
150.85, 145.45, 132.44, 121.46, 16.94.

61

Cl

N
N

ii

Ru(bpy)2Cl2

N

N

iii

Ru
N

N
Ru

N
N

N

N
N

Cl

CN

[Ru(bpy)2(CN-Me-bpy)]2+
(1)

i

CN

Cl
Cl

Ru(DMSO)4Cl2

NC

Cl

N
N

ii

iii

N

N
Ru

N

N
N

Cl

CN

N
N

Ru
N

N
N

NC

Cl

CN

CN

Cl

[Ru(CN-Me-bpy)3]2+
(3)

i

CN

Cl
Cl

ii
Ru(Cl-Me-bpy)2Cl2

CN

N

N

N

iii

Ru
N

NC

Cl

N

N

N
Ru

N

N
N

Cl

CN

N

N
N

CN

[Ru(bpy)(CN-Me-bpy)2]2+
(2)

Figure 3-2: Synthesis of complexes 1-3. (i) bpy or Cl-Me-bpy and LiCl refluxed
in DMF. (ii) bpy or Cl-Me-bpy refluxed in ethanol for 48 hrs. (iii) Pd2(dba)3, dppf,
ZnCN, and zinc dust heated in DMA. See text for details.

62

3.2.3 Synthesis of [Ru(bpy)2(CN-Me-bpy)](PF6)2, (1)
Dichlorotetrakis(dimethyl sulfoxide)ruthenium(II)
Ru(DMSO)4Cl2
Ru(DMSO)4Cl2 was prepared by a modified version of the reported literature
method.33 Dimethyl sulfoxide (25.0 mL) was bubble degassed with nitrogen for
15 minutes, after which time 1.50 g of RuCl3•xH2O were added to the solution.
The reaction was gently heated under nitrogen for 30-45 minutes until the dark
black-red solution became dark yellow-orange in color. The solution was cooled
to room temperature, poured over 100 mL of acetone, and cooled in the freezer
overnight to precipitate the product.

The yellow microcrystalline solid was

collected by vacuum filtration, rinsed once with acetone, rinsed three times with
ether, and dried under vacuum.

Yield: 2.20 g.

Anal. Calcd (Found)

for

C8H24Cl2O4S4Ru•0.25 (CH3)2SO: C, 20.26 (20.14); H, 5.10 (5.23).

Bis(2,2ʼ-bipyridine)dichlororuthenium(II)
Ru(bpy)2Cl2
Ru(bpy)2Cl2 was prepared from a modified version of the reported literature
method, using Ru(DMSO)4Cl2 rather than RuCl3•xH2O.34 Freshly distilled DMF
(35.0 mL) was bubble degassed with nitrogen for 15 minutes and 1.60 g of
Ru(DMSO)4Cl2, 1.10 g of 2,2'-bipyridine, and 7.30 g of LiCl were added to the
solution. The solution was shielded from light using aluminum foil and heated to
reflux under nitrogen for approximately 4 hours; the progress of the reaction was
monitored

using

thin

layer

chromatography

(silica,

10%

acetone

in

dichloromethane). The dark purple reaction mixture was then cooled slightly
after which the warm solution was poured over approximately 400 mL of acetone
and cooled overnight in the freezer to precipitate the product.

The dark

precipitate was collected by vacuum filtration and rinsed with water to remove

63

excess LiCl and a side product of [Ru(bpy)3]2+ until the filtrate was colorless. The
remaining purple-green solid was rinsed three times with ether and dried under
vacuum. Yield: 0.989 g (60%). Anal. Calcd (Found) for C20H16Cl2N4Ru•1.0 H2O:
C, 47.82 (47.71); H, 3.61 (3.56); N, 11.15 (10.90). 1H NMR (500 MHz, DMSOd6): δ 9.97 (d, J=5.0 Hz, 2H), 8.65 (d, J=8.1 Hz, 2H), 8.49 (d, J=8.0 Hz, 2H), 8.07
(dt, J=7.8, 1.26 Hz, 2H), 7.77 (dt, J=6.6,1.0 Hz, 2H), 7.68 (dt, J=7.8, 1.1 Hz, 2H),
7.52 (d, J=5.6 Hz, 2H), 7.10 (dt J=6.6, 1.1 Hz, 2H).

Bis(2,2ʼ-bipyridine)(4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II)
hexafluorophosphate
[Ru(bpy)2(Cl-Me-bpy)](PF6)2
Ethanol (30.0 mL) was bubble degassed with nitrogen for 15 minutes and
0.59 g of Ru(bpy)2Cl2 and 0.33 g of Cl-Me-bpy were added to the solution. The
solution was shielded from light and heated to reflux overnight under nitrogen.
The solution was cooled to room temperature and excess NaPF6 dissolved in 50
mL of water was added to the solution to precipitate a red powder.

The

precipitate was collected by vacuum filtration, rinsed well with water and ether,
and dried under vacuum. The solid was recrystallized once by acetonitrile/ether
diffusion.

Yield:

0.849

g

(73%).

Anal.

Calcd

(Found)

for

C32H27Cl2F12N6P2Ru•0.5 H2O: C, 39.81 (39.60); H, 2.82 (2.80); N, 8.70 (8.63).
1H

NMR (500 MHz, CD3CN): δ 8.51 (s, 2H), 8.48 (t, J=8.5 Hz, 4H), 8.05 (m,

J=7.9 Hz, 4H), 7.75 (d, J=5.6 Hz, 2H), 7.65 (d, J=5.6 Hz, 2H), 7.51 (s, 2H), 7.41
(dt, J=6.6, 1.3 Hz, 2H), 7.37 (dt J=6.7, 1.3 Hz, 2H), 2.20 (s, 6H). 13C NMR (500
MHz, CD3CN): δ 158.38, 158.33, 156.32, 153.86, 153.18, 152.98, 146.90,
139.19, 139.17, 138.52, 128.96, 128.77, 125.78, 125.65, 125.64, 17.71. MS [ESI
(CH3CN), m/z (rel. int.)]: 333.1 (100) [M-2PF6]2+, 811.2 (68) [M-PF6]1+.

64

Bis(2,2ʼ-bipyridine)(4,4ʼ-dicyano-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II)
hexafluorophosphate
[Ru(bpy)2(CN-Me-bpy)](PF6)2, (1)
This synthesis is a modified version of the reported synthesis for 4ʼcyano-2,2ʼ:6ʼ,2”-terpyridine

complexes

of

ruthenium.35

Anhydrous

dimethylacetamide (DMA) was degassed using freeze-pump-thaw techniques
prior to use.

[Ru(bpy)2(Cl-Me-bpy)](PF6)2 (0.500 g), zinc cyanide (0.740 g),

Pd2(dba)3 (0.045 g), dppf (0.060 g), zinc dust (0.200 g), and DMA (80.0 mL)
were combined in an inert atmosphere glovebox. The solution was transfered
from the glovebox to a Schlenk line, shielded from light, and heated slowly under
nitrogen. The progress of the reaction was monitored by electronic absorption
spectroscopy. After the reaction was complete (i.e., when no further changes in
the absorption spectrum were noted ca. 1 hour) the solution was cooled to room
temperature and filtered through celite.

The filtrate was evaporated to near

dryness under vacuum and the dark red-orange residue was dissolved in
acetonitrile and precipitated with ether. The resulting orange solid was collected
by vacuum filtration and washed with ether. The solid was recrystallized once by
acetonitrile/ether diffusion and then purified twice by column chromatography
using

neutral

alumina

and

C-18

reverse

phase

silica

gel

with

7:1

acetonitrile:aqueous KNO3 as the eluent for both columns. After the column
purification the product was recrystallized a final time by acetonitrile/ether
diffusion to remove excess nitrate salts from the column eluent. Yield: 0.309 g
(63%). Anal. Calcd (Found) for C34H26F12N8P2Ru: C, 43.55 (43.40); H, 2.79
(2.85); N, 11.95 (11.84); Ru, 10.78 (10.11). 1H NMR (500 MHz, CD3CN): δ 8.69
(s, 2H), 8.49 (t, J=7.8 Hz, 4H), 8.08 (m, J=7.9, 1.5 Hz, 4H), 7.78 (s, 2H), 7.64 (dd,
J=12.6, 5.5 Hz, 4H), 7.41 (m, J=7.08, 1.3 Hz, 4H), 2.38 (s, 6H). 13C NMR (500
MHz, CD3CN): δ 156.88, 156.56, 154.54, 153.52, 151.95, 151.54, 140.52,

65

138.49, 138.41, 127.90, 127.60, 126.21, 124.57, 124.54, 121.22, 114.80, 17.12.
MS [ESI (CH3CN), m/z (rel. int.)]: 324.0 (95) [M-2PF6]2+, 793.1 (100) [M-PF6]1+.
MS [HR-ESI (CH3CN)] m/z

793.0973 [M-PF6]1+, calcd. (C34H26N8F6PRu)

793.0966. FTIR (selected frequencies in KBr pellet, cm-1) 3117(w), 2234(m),
1605(m), 1467(s), 1447(s), 1241(m), 839(vs), 762(s), 557(vs). Electronic
absorption (CH3CN) λ, nm (ε, M-1 cm-1): 285 (63,800) 318 (36,700), 420
(13,400) 479 (14,000).

3.2.4 Synthesis of [Ru(bpy)(CN-Me-bpy)2](PF6)2, (2)
This molecule was synthesized in the same manner as [Ru(bpy)2(CN-Mebpy)](PF6)2, starting from the appropriate Cl-Me-bpy ruthenium complex.

Bis(4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine)dichlororuthenium(II)
Ru(Cl-Me-bpy)2Cl2
Freshly distilled DMF (20.0 mL) was bubble degassed for 15 minutes and
0.810 g of Ru(DMSO)4Cl2, 0.850 g of Cl-Me-bpy, and 8.00 g of LiCl were added
to the solution. The solution was shielded from light with aluminum foil and
heated to reflux under nitrogen for 2 hours. The dark purple reaction mixture was
then cooled slightly after which the warm solution was poured over approximately
400 mL of acetone and cooled overnight in the freezer to precipitate the product.
The dark precipitate was collected by vacuum filtration and rinsed with water to
remove excess LiCl and a side product of [Ru(Cl-Me-bpy)3]2+ until the filtrate was
colorless.

The remaining dark purple solid was rinsed with ether and dried

under vacuum. The product was recrystallized once by dichloromethane/ether
diffusion.

Yield: 0.747 g (66%). 1H NMR (500 MHz, DMSO-d6): δ 9.72 (s, 2H),

8.89 (s, 2H), 8.75 (s, 2H), 7.53 (s, 2H), 2.53 (s, 6H), 2.10 (s, 6H).

66

(2,2ʼ-bipyridine)bis(4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II)
hexafluorophosphate
[Ru(bpy)(Cl-Me-bpy)2](PF6)2
30.0 mL of ethanol were bubble degassed with nitrogen for 15 min and 0.500
g of Ru(Cl-Me-bpy)2Cl2 and 0.120 g of 2,2'-bipyridine were added to the solution.
The reaction was shielded from light and heated to reflux, under nitrogen,
overnight. The solution was cooled to room temperature and excess NaPF6
dissolved in 50.0 mL of water was added to precipitate a red powder.

The

precipitate was collected by vacuum filtration, rinsed well with water and ether,
and dried under vacuum. The red powder was recrystallized once by acetonitrile/
ether

diffusion.

Yield:

0.681

g

(88%).

Anal.

Calcd

(Found)

for

C34H28Cl4F12N6P2Ru•0.50 H2O: C, 38.43 (38.74); H, 2.75 (2.87); N, 7.91 (8.13).
1H

NMR (500 MHz, CD3CN): δ 8.51 (s, 2H), 8.49 (s, 2H), 8.47 (d, J=7.9 Hz 2H),

8.06 (dt, J=7.9, 1.3 Hz, 2H), 7.70 (d, J=5.7 Hz, 2H), 7.55 (s, 2H), 7.41 (s, 2H),
7.39 (dt, J=6.6, 1.3 Hz, 2H), 2.25 (s, 6H), 2.18 (s, 6H). 13C NMR (500 MHz,
CD3CN): δ 158.43, 156.42, 156.40, 154.03, 153.84, 153.23, 147.07, 147.04,
139.34, 138.59, 138.56, 128.91, 125.89, 125.80, 17.84, 17.67. MS: [ESI, m/z (rel.
int.)]: 382.1 (100) [M-2PF6]2+, 909.1 (67) [M-PF6]1+.

(2,2ʼ-bipyridine)bis(4,4ʼ-dicyano-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II)
hexafluorophosphate
[Ru(bpy)(CN-Me-bpy)2](PF6)2, (2)
[Ru(bpy)(CN-Me-bpy)2](PF6)2 was synthesized in the same manner as
[Ru(bpy)2(CN-Me-bpy)](PF6)2 using 0.300 g of [Ru(bpy)(Cl-Me-bpy)2](PF6)2,
0.400 g of zinc cyanide, 0.051 g

of Pd2(dba)3, 0.063 g of dppf, 0.300 g of zinc

dust, and 60.0 mL of DMA. Yield: 0.116 g (40%). Anal. Calcd (Found) for
C38H28F12N10P2Ru•1.5 H2O: C, 43.77 (43.89); H, 2.99 (2.72); N, 13.43 (13.20);

67

Ru, 9.69 (8.94). 1H NMR (500 MHz, CD3CN): δ 8.70 (s, 2H), 8.68(s, 2H), 8.50 (d,
J=7.9 Hz, 2H), 8.12 (dt, J=7.9 Hz, 1.5, 2H), 7.68 (t, J=0.7 Hz, 2H), 7.64 (t, J=0.7
Hz, 2H) 7.58 (d, J=5.3 Hz, 2H), 7.42 (dt, J=6.7, 1.3 Hz, 2H), 2.41 (s, 6H), 2.37 (s,
6H). 13C NMR (500 MHz, CD3CN): δ 157.90, 155.74, 155.45, 155.10, 154.80,
153.21, 142.25, 142.15, 140.34, 129.26, 127.77, 127.76, 126.22, 123.36, 123.27,
116.04, 116.02, 18.65, 18.44. MS [ESI (CH3CN), m/z (rel. int.)]: 363.1 (100)
[M-2PF6]2+, 871.1 (48)

[M-PF6]1+. MS [HR-ESI (CH3CN)] m/z 871.1166 [M-

PF6]1+, calcd (C38H28N10F6PRu) 871.1184. FTIR (selected frequencies in KBr
pellet, cm-1) 3117(w), 2920(w), 2236(s), 1607(m), 1478(s), 1384(s), 1241(m),
841(vs), 765(m), 562(vs). Electronic absorption (CH3CN) λ, nm (ε, M-1 cm-1):
284 (37,000) 317 (55,200), 367 (12,800), 440 (12,600), 477 (16,700).

3.2.5 Synthesis of [Ru(CN-Me-bpy)3](PF6)2, (3).
This molecule was synthesized in the same manner as [Ru(bpy)2(CN-Mebpy)](PF6)2, starting from the appropriate Cl-Me-bpy ruthenium complex.
Tris(4,4ʼ-dichloro-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II) hexafluorophosphate
[Ru(Cl-Me-bpy)3](PF6)2
Ethanol (30 mL) was bubble degassed with nitrogen for 15 minutes and 0.260
g of Ru(DMSO)4Cl2 and 0.530 g of Cl-Me-bpy were added to the solution. The
reaction was shielded from light and heated to reflux for 48 hours under nitrogen.
The solution was cooled to room temperature and excess NaPF6, dissolved in
50.0 mL of water, was added to precipitate a red powder. The precipitate was
collected by vacuum filtration, rinsed well with water and ether, and dried under
vacuum. The solid was recrystallized once by acetonitrile/ether diffusion. Yield:
0.385 g (63%). Anal. Calcd (Found) for C36H30Cl6F12N6P2Ru: C, 37.59 (37.17);
H, 2.63 (2.62); N, 7.31 (7.25). 1H NMR (500 MHz, CD3CN): δ 8.49 (s, 2H), 7.48
(s, 2H), 2.23 (s, 6H). 13C NMR (500 MHz, CD3CN): δ 156.33, 153.88, 147.11,

68

138.57, 125.91, 17.68. MS [ESI, m/z (rel. int.)]: 430.1 (100) [M-2PF6]2+, 1005.1
(55) [M-PF6]1+.

Tris(4,4ʼ-dicyano-5,5ʼ-dimethyl-2,2ʼ-bipyridine)ruthenium(II) hexafluorophosphate
[Ru(CN-Me-bpy)3](PF6)2, (3)
[Ru(CN-Me-bpy)3](PF6)2

was

synthesized

in

the

same

manner

as

[Ru(bpy)2(CN-Me-bpy)](PF6)2 using 0.630 g of [Ru(Cl-Me-bpy)3](PF6)2, 1.90 g of
zinc cyanide, 0.160 g of Pd2(dba)3, 0.180 g of dppf, 0.500 g of zinc dust, and 125
mL

of

DMA.

Yield:

0.117

g

(20%).

Anal.

Calcd

(Found)

for

C42H30F12N12P2Ru•1.75 CH3CN: C, 46.89 (47.29); H, 3.05 (2.77); N, 16.52
(16.11); Ru, 8.67 (8.29). 1H NMR (500 MHz, CD3CN): δ 8.71 (s, 2H), 7.59 (s,
2H), 2.41 (s, 6H). 13C NMR (500 MHz, CD3CN): δ 155.33, 154.99, 142.50,
128.01, 123.83, 115.95, 18.51. MS [ESI (CH3CN), m/z (rel. int.)]: 402.1(100)
[M-2PF6]2+, 949.1 (10)

[M-PF6]1+. MS [HR-ESI (CH3CN)] m/z 949.1413 [M-

PF6]1+, calcd (C42H30N12F6PRu) 949.1402. FTIR (selected frequencies in KBr
pellet, cm-1) 3050(w), 2235(m), 1631(m), 1477(s), 1385(s), 1242(m), 842(vs),
558(vs). Electronic absorption (CH3CN) λ, nm (ε, M-1 cm-1): 315 (78,800), 458
(23,000).

3.2.6 Physical Measurements
Electrochemistry and Spectroelectrochemistry.
Electrochemical measurements were carried out in a Ar-filled drybox
(Vacuum Atmospheres) using a CHI 630B electrochemical analyzer. A standard
three-electrode arrangement was used consisting of a Pt working electrode, a
graphite counter electrode, and a Ag/AgCl reference electrode (Cypress
Systems). Measurements were carried out in spectrophotometric grade CH3CN,
which was freeze-pump-thaw degassed before use, and using 0.1 M

69

tetrabutylammonium

hexafluorophosphate

(TBAPF6)

as

the

supporting

electrolyte. Data were acquired by cyclic voltammetry (CV) and differential pulse
voltammetry (DPV); the scan rate for the CV measurements was 50 mV/s and
the scan rate and pulse width for the DPV measurements were 20 mV/s and 50
mV, respectively.
comparable.

Values for E1/2 obtained by the two techniques were

All oxidation and reduction waves were reversible over several

consecutive scans.

Potentials are listed versus the ferrocene/ferrocenium

couple, which was used as an internal standard, and quoted as E1/2 values as
calculated from the DPV peak potentials.36
UV-visible spectroelectrochemical experiments were performed in a 1 mm
pathlength spectroelectrochemical cell (CH Instruments) with Ag/AgCl reference
electrode inside an Ar filled glovebox.

Measurements were carried out in

spectrophotometric grade CH3CN, which was freeze-pump-thaw degassed
before use, and using 0.1 M TBAPF6 as the supporting electrolyte. Samples
were dissolved in the electrolyte solution to give an absorbance of 0.3-0.5 at the
maximum of the main absorption band in the visible region. Difference spectra
were collected on a SI440 CCD spectrometer in ca. 30 second intervals as the
samples were oxidized or reduced.

Oxidative spectra were collected at a

potential 100 mV more positive than E1/2ox; reductive spectra were collected at a
potential 100 mV more negative than E1/2red1 or halfway between E1/2red1 and
E1/2red2, whichever was less reducing.

Electronic Steady-State and Time-Resolved Spectroscopies.
All spectra were collected in spectrophotometric grade CH3CN unless
otherwise noted. For steady state emission and time-resolved measurements,
the solvent was freeze-pump-thaw degassed before use. Electronic absorption
spectra for all compounds were acquired using a Cary 50 spectrophotometer.

70

Steady-state emission spectra were acquired using a Spex Fluoromax
fluorimeter and corrected for instrumental response using a NIST standard of
spectral irradiance (Optronic Laboratories, Inc., OL220 M tungsten quartz lamp).
Samples were prepared in an inert atmosphere glovebox in 1 cm quartz cuvettes
and measured under optically dilute conditions (o.d. 0.1-0.2). The resulting
emission spectra were fit with an asymmetric double sigmoidal function using
IGOR pro.

This function has no mathematical significance but it is able to

accurately reproduce the shape of the entire emission curve and thereby capture
the small area (< 10%) that lies outside of the detector range thereby providing
for a more accurate estimate of the integrated spectrum.

Relative radiative

quantum yields (Φr) were determined using [Ru(bpy)3](PF6)2 as a standard
(Φstd = 0.095 in degassed CH3CN37). Quantum yields were calculated using
Equation 3-1,where Φunk is the relative radiative quantum yield of the sample,
Iunk and Istd are the integrated areas of the corrected emission spectra of the
! unk = ! std

I unk
Aunk

A std
I std

"unk 2
"std

3-1

sample and standard respectively, Aunk and Astd are the absorbances of the
sample and the standard at the excitation wavelength, and ηunk and ηstd are the
indexes of refraction of the respective solvents (taken to be equal to the neat
solvents in both cases). Low-temperature emission spectra were collected using
a Janis SVT-100 optical cryostat as described previously.38 Measurements were
taken at 80 K in a 9:2 mixture of butyronitrile and propionitrile, both of which were
freeze-pump-thaw degassed before use. Estimates of the zero point energy gap
(E0), Huang-Rhys factor (SM), energy of the average vibrational mode coupling
the ground and excited states ( ωM), and spectral bandwidth (Δν̅0,1/2) were
determined by a single mode fit of the steady-state emission spectra to Equation

71

3-2 as described by Claude and Meyer.39 The correction of Parker and Rees
was applied to all spectra when converting from wavelength to energy units.40
5

I (¯ =
!)
!M=0

E0 " ! M #M
E0

exp " 4(ln 2)

3

3-2

!
S MM
!M !

! " E 0 + ! M #M
¯
$ ! 0 ,1/ 2
¯

2

Nanosecond time-resolved emission and transient absorption experiments
were carried out using a Nd:YAG laser spectrometer that has been described
previously.41,42 Time-resolved emission was collected on the same samples
used to acquire the room-temperature steady-state emission spectra (vide
supra). Samples for time-resolved absorption measurements were prepared with
an absorbance in the range of 0.3-0.5 at the excitation wavelength (500 nm for 1
and 2 and 480 nm for 3) and sealed under an Ar atmosphere in 1 cm quartz
cuvettes. Data corresponds to a 15 shot average (0.2 Hz) of the signal and
baseline as well as background sample emission with 1-3 mJ of power at the
sample.

The baseline and emission were subsequently subtracted from the

signal and the data analyzed using a program of local origin. All data were
checked for linearity with respect to pump power. In addition, absorption spectra
were measured before and after all time-resolved absorption experiments to
ensure the integrity of the sample. Laser power was periodically monitored to
ensure constant pump power over the course of the experiment. Data acquired
at each probe wavelength was fit to a single exponential kinetic model; the
amplitudes for each of these fits were plotted to produce the differential
absorption spectra reported herein.

72

Nanosecond Time-Resolved Infrared Spectroscopy.
Nanosecond time-resolved step-scan infrared (SSIR) absorption spectra were
measured on a step-scan modified Bruker IFS66 FTIR spectrometer with a
standard globar source and nitrogen purge.

Nitromethane was used as the

solvent in these experiments; emission measurements on complex 1 in CH3CN
and CH3NO2 showed no significant differences in the emission profile or lifetime.
The compounds were dissolved in dried CH3NO2 to give a ground state infrared
absorption of 0.2-0.6 for the ν̅(CN) band (10-20 mM).

All solutions were

deoxygenated by bubbling with nitrogen for 15 min. Spectra were measured in a
demountable CaF2 cell with a 0.5 mm or 1 mm Teflon spacer (Specac). Samples
were excited using an OPOTEK Vibrant Nd:YAG laser (~3-4 mJ/pulse, 10 Hz).
Excitation wavelengths within the lowest energy visible absorption feature were
chosen such that the absorbance was less than 2 in order to ensure uniform
excitation of the sample. An AC/DC-coupled photovoltaic Kolmar Technologies
mercury cadmium telluride (PV MCT) detector with a 20 MHz preamplifier was
used to sample the transmitted IR probe beam. The detectorʼs AC signal was
further amplified (25X) with a 350 MHz fast preamplifier (Stanford Research
SR445A) before being directed to a 100/200 MHz PAD82a transient digitizer
board.

The interferogram response before and after laser excitation was

collected in 10 ns time slices, with 30 laser shots averaged at each mirror
position.

For each scan, folding limits of 2600 and 1000 cm-1 at 4 cm-1

resolution resulted in 1332 mirror positions.

The DC signal was collected

separately and used to check for sample decomposition as well as for phase
correction of the AC signal. Bruker Instrumentsʼ Opus 5.5 software was used to
process the recorded data. Differential absorbance spectra were calculated from
the AC and DC single channel spectra as described previously.43

The

differential excited state absorption spectra reported herein represent an average

73

of 4-8 scans, ground-state spectra correspond to an average of 30 (rapid) scans.
Calculations.
Calculations on complex 1 were performed using the Gaussian 03 software
package.44 Geometry optimizations were done on both the ground state and the
lowest energy triplet state using a spin unrestricted formalism at the B3LYP/
LANL2DZ level of theory.45,46

No symmetry restrictions were placed on the

geometry optimizations. The effects of the acetonitrile solvent environment were
included by using the polarizable continuum model (PCM).

Frequency

calculations were performed on both the singlet and triplet optimized structures to
ensure that these geometries corresponded to global minima.

No imaginary

frequencies were obtained for either of the optimized geometries. The highest
energy CN-Me-bpy π bonding orbital of the ground state optimized geometry and
the lowest energy CN-Me-bpy π* antibonding orbital of the lowest energy triplet
optimized geometry were visualized with GaussView. The cartesian coordinates
for the optimized geometries can be found in Appendix D.

3.3 Results and Discussion
I have synthesized a series of cyano substituted ruthenium polypyridyl
complexes with the general form [Ru(bpy)3-n(CN-Me-bpy)n](PF6)2,as shown in
Figure 3-2.

These complexes have been designed to combine the well

documented charge transfer properties of the ruthenium polypyridyl complexes
with the cyanide infrared tag in order to study the vibrational relaxation process in
charge transfer excited states. This chapter describes the synthesis of the three
complexes as well as the spectroscopic characterization of the long lived 3MLCT
state; this study will provide the necessary foundation for the detailed study of
vibrational relaxation dynamics in a MLCT chromophore in Chapter 4.

74

3.3.1 Synthesis
The synthesis of ruthenium complexes is well established, versatile, and can
be used with bipyridine ligands with electron withdrawing groups such as such as
nitro, trifluoromethyl, and ethyl ester.47-51 However, initial attempts to synthesize
ruthenium complexes of 4,4'-dicyano-2,2'-bipyridine (CN-bpy) using a variety of
reaction conditions led to only trace amounts of the desired products even with
extended (weeks to months) reaction time. The fact that examples of ruthenium
complexes containing cyano substituted bipyridine ligands exist in the literature
suggests that [Ru(bpy)3-n(CN-bpy)n]2+ (n=1-3) complexes are thermodynamically
stable and kinetic factors are likely responsible for the low yields of the CN-bpy
complexes.52,53 If this is true, a change in reaction mechanism could beneficially
alter the reaction kinetics, and (potentially) increase product yield.
One alternate pathway to form ruthenium complexes of CN-bpy is to carry out
the cyanide functionalization after the ligand is bound to the metal center. There
are many examples of such transformations in the literature, circumventing
difficult purifications, insoluble intermediates, and/or kinetically unfavorable
reactions.54-59

In particular palladium catalysts have proven effective in

converting aryl halides to cyanides in high yields;60,61 Hanan, Campagna, and
coworkers have demonstrated that this reaction can also be used on ruthenium
complexes.35

Starting from the chloro-substituted ruthenium bipyridine

complexes, I was able to synthesize ruthenium complexes of 4,4'-dicyano-2,2'bipyridine, proving that the desired complexes are indeed thermodynamically
stable. Unfortunately, these complexes were highly susceptible to hydrolysis to
form the corresponding amide, especially during column chromatography as well
as subsequent to reduction during electrochemical measurements. In order to
increase the basicity of the nitrogens and (hopefully) increase the stability of the
complexes, electron donating methyl groups were added to the 5 and 5' positions

75

of the bipyridine ring. This approach proved to be successful, yielding the final
[Ru(bpy)3-n(CN-Me-bpy)n]2+ (n=1-3) complexes, shown in Figure 3-2, that could
withstand

column

chromatography

as

well

as

exhibiting

reversible

electrochemistry.

Wavelength (nm)
300

Normalized Absorbance

6

350

400

500

Start
15 min
20 min
28 min
30 min
35 min
40 min

5
4
3

600

45 min
50 min
57 min
60 min
63 min
67 min

2
1
0

35

30

25

Energy (x10

3

20

15

-1

cm )

Figure 3-3: Electronic absorption spectra of the conversion of [Ru(bpy)2(Cl-Mebpy)]2+ to [Ru(bpy)2(CN-Me-bpy)]2+. The inset indicates the time following
initiation of the reaction that each aliquot was taken. See text for further details.

The synthesis of the desired complexes took place in three parts. First, the
Cl-Me-bpy ligand was synthesized via the nitro-substituted intermediate using the
literature procedure reported for 4,4-dichloro-2,2'-bipyridine.28-32 The ruthenium
complexes of the Cl-Me-bpy ligand were then synthesized using well-established
literature methods. For the final step of synthesizing the CN-Me-bpy ruthenium
complexes, the procedure Hanan, Campagna, and coworkers reported for

76

[Ru(tpy)(CN-tpy)]2+ and [Ru(CN-tpy)2]2+(CN-tpy = 4'-cyano-2,2':6',2"-terpyridine)
had to be modified for these bipyridine based reactions.

The amount of

palladium catalyst and ligand were kept the same as reported (5 and 10 mol%
per CN group respectively), however I found that an excess of zinc cyanide (as
opposed to the stoichiometric amount used for the tpy-based systems) was
necessary to form the bipyridine complexes. In addition, the 6-12 hour reaction
time necessary for the terpyridine complexes had to be truncated to 1-2 hours for
the bipyridine analogs to avoid the formation of side products which were not
easily removed by recrystallization or column chromatography.

The ideal

reaction time differed depending on the exact rate of heating (~1°C/minute was
used), so monitoring the reaction progress by UV-vis spectroscopy, as shown in
Figures 3-3, A-2, and A-4, was essential for determining when the reaction had
reached completion. Even under the best of conditions, the separation achieved
by column chromatography was less than ideal; multiple fractions were collected
and evaluated by UV-vis and NMR.

Attempts to grow single crystals were

unsuccessful, so the identity and purity of the final complexes were ultimately
determined by NMR, high resolution ESI/MS, and elemental analysis.

3.3.2 Ground State Spectroscopic Properties
The ground state absorption spectra of the three complexes and [Ru(bpy)3]2+
are shown in Figure 3-4.

Metal bipyridyl complexes typically exhibit both

intraligand as well as charge trasfer transitions. Based on extinction coefficients
as well as other ruthenium polypyridyl complexes, the UV absorption features in
complexes 1-3 can be assigned as π

π* absorptions of the bipyridine ligands

whereas the somewhat weaker visible features are as metal-to-ligand charge
transfer (MLCT) in nature. The variation in composition across the series is
reflected in the spectra.

For example, in complex 3 there is a single UV

77

absorption at 315 nm which can be readily assigned as a π
the CN-Me-bpy ligand.

π* absorption(s) of

Complexes 1 and 2 also have UV absorptions at 318

and 317 nm, respectively, which are likewise also assigned to

π

π*

absorptions of the CN-Me-bpy ligand. In the heteroleptic complexes there is also

Wavelength (nm)
350

400

500

600

80

3

-1

-1

Molar Absorptivity (x10 M cm )

300

60

40

20

0

35

30
25
3
-1
Energy (x10 cm )

20

15

Figure 3-4: Electronic absorption spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1,
blue line), [Ru(bpy)(CN-Me-bpy)2](PF6)2 (2, green line), [Ru(CN-Me-bpy)3](PF6)2
(3, red line), and [Ru(bpy)3](PF6)2 (black line) in CH3CN solutions.

a higher energy UV absorption that is not present in complex 3, but is in
[Ru(bpy)3]2+: this is most likely associated with the unsubstituted bipyridine
ligand which is not present in complex 3. The change in relative intensity of the
two bands going from complexes 1 to 3 is consistent with this assignment, as are
the relative energies of the two absorption features given the additional
78

conjugation of the π system of the CN-Me-bpy ligand compared to the
unsubstituted bipyridine. Assignments within the charge transfer band are not a
straight forward, but can be clarified using electrochemistry. To a reasonable
approximation the energy of a metal-to-ligand charge transfer band can be
thought of in terms of the energy.required to oxidize the metal and reduce the
ligand i.e. E(MLCT) ≈ E(M

M+) + E(L

L-) for M-L

M+-L-.62 For two ligands

bound to the same metal, the energy of the MLCT states will therefore correlate
with the reduction potential of the ligands. The electrochemical properties of
complexes 1-3 have been investigated by cyclic voltammetry (CV) and
differential pulse voltammetry (DPV), the results of which are shown in Figure
3-5 and listed in Table 3-1.
The electron-withdrawing nature of the cyano group is immediately apparent
in the trends in the ruthenium oxidation potential across the series. Starting with
[Ru(bpy)3]2+, each successive replacement of 2,2'-bipyridine by CN-Me-bpy in
the coordination sphere systematically shifts the oxidation potential of the metal
center positive by ca. 130 mV. This electropositive shift in charge density at the
ruthenium indicates the generally poorer electron donating ability of the CN-Mebpy relative to the unsubstituted ligand. For the ligand reductions, first compare
the reduction potentials of complex 3 and [Ru(bpy)3](PF6)2 where there is no
question of which ligand is being reduced. The CN-Me-bpy ligand is more easily
reduced than the bpy ligand by 480 mV and in both complexes the three
reduction potentials are evenly spaced, with 150 to 250 mV between them.
Nextwe compare the two heteroleptic complexes (1 and 2) to the two homoleptic
complexes. In complex 1 the first reduction potential is much closer in energy to
that of complex 3 whereas the second and third reduction potentials are much
closer to those of [Ru(bpy)3](PF6)2, this combined with the large energy splitting
between the first and second reductions show that the first reduction potential is

79

1 !A
-1.2

-1.6

-2.0

-2.4

1.2 1.0 0.8

1.0

0.0

-1.0

-2.0

Potential (V)

Figure 3-5: Cyclic voltammogram of complex 1 in CH3CN with 0.1 M TBAPF6 as
the electrolyte. Spectrum was collected using the standard three electrode setup
with a platinum working electrode, graphite counter electrode, and a Ag/AgCl
reference electrode and is referenced to the ferrocene/ferrocenium redox couple,
which was added as an internal standard. The differential pulse voltammogram
is shown in the insets.

Table 3-1: Electrochemical Data for Complexes 1-3 and [Ru(bpy)3](PF6)2a
Complex

E1/2ox

E1/2red1

E1/2red2

[Ru(bpy)3](PF6)2

+0.91

-1.70

-1.89

-2.14

2.61

2.03

(1)

+1.01

-1.38

-1.86

-2.06

2.45

1.82

(2)

+1.13

-1.33

-1.52

-1.99

2.46

1.91

(3)

+1.29

-1.25

-1.40

-1.61

2.54

1.99

aPotentials

E1/2red3 ΔE (eV)b E0 (eV)c

are reported in V versus the ferrocene/ferrocenium couple in
acetonitrile solution as described in the experimental section.
bΔE = (E ox -E red1). cE from Table 3-3 (page 82).
1/2
1/2
0

80

the reduction of the CN-Me-bpy ligand and the second and third reductions are
those of the unsubstituted bpy ligands. A similar comparison can be made for
complex 2 which shows that the first and second reductions are the reductions of
the two CN-Me-bpy ligands and the third reduction is the reduction of the
unsubstituted bpy ligand.
With the electrochemical results in hand, it is now straightforward to assign
the features in the electronic absorption spectrum.

Specifically, for complex 1,

the lower energy absorption at 479 nm can be assigned to an MLCT transition
associated with the CN-Me-bpy ligand whereas the higher energy portion of the
absorption band near 420 nm are charge-transfer features coupled to the
unsubstituted bpy ligands. The same is true for complex 2, the lowest energy
visible absorption feature is an MLCT transition to the CN-Me-bpy ligands and
the higher energy visible absorption is an MLCT transition to the unsubstituted
bpy ligand. In complex 3 there is a single visible absorption at 458 nm which
obviously corresponds to MLCT transitions to the CN-Me-bpy ligand.

It is

interesting to note that the MLCT absorption in the symmetric tris complex is
much narrower and appears at higher energy than its counterpart in the
heteroleptic complexes 1 and 2.

This phenomenon has been seen in other

series of heteroleptic ruthenium complexes and the change in breadth of the
absorption band is most likely due to the reduction of symmetry (i.e. from
nominally D3 to C2) while the difference in energy is primarily due to ancillary
ligand effects.19,50,63

3.3.3 Steady-State and Time-Resolved Emission Spectroscopies
The room temperature emission spectra for all three complexes are plotted in
Figure

3-6a.

The

emission

maxima

for

[Ru(bpy)2(CN-Me-bpy)](PF6)2,

[Ru(bpy)(CN-Me-bpy)2](PF6)2, and [Ru(CN-Me-bpy)3](PF6)2 are at 686 nm, 658

81

Wavelength (nm)
550

600

700

800

(a)

1.0

Normalized Emission Intensity

900

0.5

0.0
(b)

1.0

0.5

0.0

18

16

14
3

12

10

-1

Enengy (x10 cm )

Figure 3-6: Steady-state emission spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1,
blue squares), [Ru(bpy)(CN-Me-bpy)2](PF6)2 (2, green triangles), and [Ru(CNMe-bpy)3](PF6)2 (3, red circles). (a) Room-temperature spectra acquired in
deoxygenated CH3CN solution. The solid lines correspond to fits to an
asymmetric double sigmoidal function – see text for further details. (b) Emission
spectra acquired in a 9:2 butyronitrile/propionitrile glass at 80 K.

nm and 626 nm respectively following the same trend as the ruthenium oxidation
potentials across the series. Radiative quantum yields for complexes 1-3 were
determined relative to the [Ru(bpy)3](PF6)2 standard, and are listed in Table 3-2.
As the emission energy increases the quantum yield of the complexes also

82

Table 3-2: Photophysical Data for Complexes 1-3 and [Ru(bpy)3](PF6)2

Complex
[Ru(bpy)3](PF6)2

λem
(nm)

kobs b
(x105 s-1)

Φa

kr a,c
(x104 s-1)

knr a,d
(x105 s-1)

618 0.095 ± 0.003 10.5 ± 0.10

10.0 ± 0.30 9.53 ± .08

(1)

686

0.13 ± 0.004

6.41 ± 0.08

8.33 ± 0.25

5.58 ± 0.17

(2)

658

0.27 ± 0.008 3.82 ± 0.06

10.3 ± 0.31

2.79 ± 0.08

(3)

626

0.40 ± 0.012

11.4 ± 0.35 1.71 ± 0.05

2.86 ± 0.04

a Error

bars based on the 3% uncertainty in the [Ru(bpy)3](PF6)2 quantum yield
standard. b Uncertainty determined by the standard deviation of multiple
measurements. c kr = Φr• kobs. d knr = kobs- kr.

increases, going from 13% in 1 to 40% in 3. While these quantum yields may
seem unusually large for ruthenium polypyridyl complexes (especially that of
complex 3), many ruthenium tris bipyridine41 and tris phenanthroline47
complexes containing aromatic substituents exhibit radiative quantum yields on
this order. In addition the recent correction in the absolute value for Φr from
6.2% to 9.5% for [Ru(bpy)3](PF6)2 must be borne in mind when comparing
values in the literature.37
To obtain more quantitative insights into the 3MLCT excited states of this
system we carried out a single mode spectral fitting analysis using Equation 2 as
described by Meyer and coworkers.39,64

The room temperature emission

spectra are typical of molecules in this class, showing virtually no vibrational fine
structure; the spectra

can be fit equally well with a wide range of the four

parameters (E0, SM, ωM, and Δν̅0,1/2). We therefore acquired low temperature
emission spectra (Figure 3-6b) which exhibit sufficient fine structure to yield well-

83

defined values for all four parameters. Following the example of Claude and
Meyer, the value of

ωM was assumed to be temperature independent, so the

room temperature spectra were refit using the low temperature value of
a fixed value.
3MLCT

ωM as

Using this approach, well defined (<10% variation) values for

zero-point energy (E0), Huang-Rhys factor (SM), and the spectral

bandwidth (Δν̅0,1/2) could be determined for the room-temperature spectra.
Spectral fitting parameters for complexes 1-3 and [Ru(bpy)3](PF6)2 are listed in
Table 3-3.

Table 3-3: Spectral Fitting Results for Complexes 1-3 and [Ru(bpy)3](PF6)2

Complexes

Low Temperaturea,b
E0 (cm-1)

SM

ωM (cm-1)

Δν̅0,1/2(cm-1)

(1)

15409 ± 40

0.67 ± 0.06

1288 ± 40

985 ± 60

(2)

15969 ± 50

0.69 ± 0.05

1286 ± 40

950 ± 50

(3)

16335 ± 70

0.84 ± 0.07

1365 ± 90

1316 ± 90

[Ru(bpy)3](PF6)2

17220 ± 50

1.05 ± 0.06

1345 ± 40

948 ± 60

Room Temperatureb,c
E0 (cm-1)

SM

ωM (cm-1)d

Δν̅0,1/2(cm-1)

(1)

14653 ± 50

0.65 ± 0.1

1288

1676 ± 110

(2)

15409 ± 50

0.73 ± 0.09

1286

1674 ± 120

(3)

15972 ± 70

0.79 ± 0.11

1365

1813 ± 150

[Ru(bpy)3](PF6)2

16366 ± 50

1.01 ±0.08

1345

1742 ± 120

a Fitting results for 80 K emission spectra in 9:2 butyronitrile:propionitrile.
b Error bars represent an approximate range of visually equivalent fits. c Fitting

results for room temperature emission spectra in CH3CN, using a fixed value of
ωM based on the low temperature data. See text for details.

84

The fitted values for E0 for all three complexes are very close to the observed
emission maxima, indicating that the ν* = 0

ν = 0 transition is the dominant

contribution to the observed emission spectrum. The Huang-Rhys factor is a
measure of the vibrational overlap between the excited state and the ground
state and is proportional to (ΔQ)2, where ΔQ is the nuclear displacement
between the excited state and the ground state. This makes SM a useful
parameter for gauging the amount of structural distortion in the excited state
relative to the ground state.41 Comparing the room temperature Huang-Rhys
factor for complexes 1 - 3 it is difficult to draw any conclusions concerning the
differences across the series since the variations in SM are within the
uncertainties of the fits.
What is important to note, however, is that all three complexes have SM
values that are significantly lower than for [Ru(bpy)3](PF6)2. This is not simply a
consequence of the value of ωM since the fitted value of ωM for complex 3 is
actually slightly larger than for [Ru(bpy)3](PF6)2 yet still exhibits a smaller value
of SM. This result indicates that the three cyano substituted complexes have
excited states which are less displaced relative to the ground state compared to
[Ru(bpy)3](PF6)2, which is evidence that the emissive 3MLCT excited state is
strongly coupled on to the peripheral cyano group.
The lifetimes of complexes 1-3 were also determined using nanosecond timeresolved emission. From the lifetimes and quantum yields, radiative and nonradiative decay rates for all three complexes were calculated (Table 3-2). The
most significant difference between the complexes is the value of knr.

As

predicted by the energy gap law64, we see an inverse relationship between the
zero-point energy differences and the rate of non-radiative decay back down to
the ground state. The smaller non-radiative decay rate in the cyano substituted
complexes as compared to [Ru(bpy)3](PF6)2 suggests that the excited state in

85

these complexes is delocalized onto the cyano group leading to a smaller excited
state geometry distortions and less vibrational overlap between the long lived
excited state and the ground state; this is wholly in agreement with the smaller
values of SM for complexes 1-3 compared to [Ru(bpy)3](PF6)2 discussed above.
The differences in radiative decay rate (kr) are less dramatic than what is
observed for knr. Radiative decay theory specifies a proportionality between kr
and the product μ2 • E03 where μ is the transition dipole moment and E0 is the
energy gap.65 In this context we note that, for complexes 1 through 3, as the
energy gap increases the radiative rate constant also increases, which is
consistent with radiative decay theory. We also note that the energy gap for
[Ru(bpy)3](PF6)2 is slightly larger than complex 3, yet the radiative decay rate is
smaller than that in complex 3. This apparent discrepancy can be explained by
considering the transition moment dipole in each complex. As discussed above,
the charge transfer excited state in [Ru(CN-Me-bpy)3](PF6)2 is most likely
delocalized onto the cyano group: this would also give rise to a larger transition
dipole moment for emission from this state.

This conclusion is consistent with

the larger extinction coefficients for the 1A1

1MLCT

absorption for this series

as compared to [Ru(bpy)3](PF6)2 (Figure 3-4). Previous work from our lab on a
series of aryl-substituted ruthenium (II) polypyridyl complexes revealed similar
trends in kr and knr that were attributed to an extended π system in the MLCT
manifold.41

3.3.4 Time-Resolved Electronic Absorption Spectroscopy
Nanosecond transient absorption spectroscopy is another important tool for
characterizing the lowest energy excited state in transition metal complexes,
particularly as a precursor to ultrafast spectroscopic measurements.66 Transient
absorption spectra of complexes 1-3 acquired following 1A1

86

1MLCT

excitation

Wavelength (nm)

Change in
Absorbance

0.1

350

400

450

500

600
(a)

0.0

Change in
Absorbance

Change in
Absorbance

Change in
Absorbance

-0.1
(b)

0.1
0.0
-0.1

(c)

0.05
0.00
-0.05
-0.10
0.4

(d)

0.2
0.0
-0.2
-0.4
28

26

24

22
3

20

18

16

-1

Energy (x10 cm )

Figure 3-7: Nanosecond time-resolved differential absorption spectra acquired in
room-temperature CH3CN solution for (a) [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1), (b)
[Ru(bpy)(CN-Me-bpy)2](PF6)2 (2), (c) [Ru (CN-Me-bpy)3](PF6)2 (3), and (d)
[Ru(bpy)3](PF6)2. The individual points correspond to the amplitudes of fits of the
kinetics data to single-exponential decay models; a smoothed solid line has been
included in each plot to guide the eye.

87

are shown in Figure 3-7. The most striking aspect of Figure 3 is the similarity
among the three CN-Me-bpy complexes as compared to [Ru(bpy)3](PF6)2,
especially in the near UV region of the spectrum. While changes in the transient
absorption spectrum are affected by both differences in the ground state and
excited state absorption, regions of positive absorption are dominated by the
excited state absorption. The red shift of the near UV absorption in complexes
1-3 relative to [Ru(bpy)3](PF6)2 suggests that the nature of the excited state in
complexes 1-3 is different from that found in [Ru(bpy)3](PF6)2.

This is in

agreement with the electrochemical data, which showed that CN-Me-bpy was the
lowest energy ligand in complexes 1-3.

There are three principle features

observed in the spectra: a strong net absorption in the near-UV, a strong bleach
in the mid-visible, and a weak absorption extending into the red. This overall
pattern is typical of what is observed for the excited state spectra of RuII
polypyridyl complexes. Since an MLCT excited state can be thought of in terms
of an oxidized metal and a reduced ligand (vide supra), spectroelectrochemistry
can be quantatively used as a guide to determine the specific origins of each of
these signals.
The change in absorbance upon oxidation of complex 1 is shown in Figure
3-8b. When the Ru2+ is oxidized to Ru3+, the dominant feature in the differential
absorption spectrum is the loss of the MLCT absorption band in the visible. The
π

π* absorption of the CN-Me-bpy ligand, which occurs at ~315 nm in the

ground state spectrum (Figure 3-4) shifts to ~340 nm when coordinated to Ru3+.
The weak absorption at 725 nm shown in the inset can be assigned as an LMCT
absorption based on comparison to other reported Ru3+ polypyridyl complexes.67
When the CN-Me-bpy ligand is reduced (Figure 3-8c) strong absorptions
associated with the radical anion of the ligand are seen across the UV and visible
portions of the spectrum. A small bleach is also seen in the mid-visible due to

88

Wavelength (nm)

Change in
Absorbance

Change in
Absorbance

Molar Absortivity
3 -1
-1
(x10 M cm )

350

400

500

600 700
(a)

15
10
5
0
0.2
0.1
0.0
-0.1
-0.2

(b)

0.005
0.000
16

14

12

(c)

0.4
0.2

Change in
Absorbance

0.0
(d)

0.05
0.00
-0.05
-0.10
-0.15

28

24

20
3

16
-1

Energy (x10 cm )

Figure 3-8: (a) Ground state absorption spectrum for [Ru(bpy)2(CN-Mebpy)](PF6)2 (1) in room-temperature CH3CN solution. (b) Oxidative difference
spectra acquired at an applied potential of +1.55 V versus Ag/AgCl. The inset
corresponds to an expanded view of the low-energy portion of the spectrum. (c)
Reductive difference spectra acquired at an applied potential of -1.05 V versus
Ag/AgCl. (d) Time-resolved differential absorption spectrum of compound 1
following ca. 10 ns excitation at 500 nm.

89

the loss of the charge transfer transition to the CN-Me-bpy that has undergone
reduction.
The transient absorption spectrum of complex 1 in Figure 4d can now be
qualitatively compared to the oxidative and reductive spectra to identify the
origins of the different transitions. The spectroelectrochemical spectra cannot be
quantitatively combined to give the transient absorption spectrum but a
comparison between the features of the spectroelectrochemical spectra and the
transient absorption spectra can offer insight to the origin of the features of the
transient absorption spectra. In the mid visible region of both the oxidative and
reductive spectra there is a negative feature that correspond to the loss of metalto-ligand charge transfer absorption band(s).

The negative feature in the

transient absorption spectrum is, therefore due to the loss of metal-to-ligand
charge transfer absorption features in the excited state. The negative bleach
feature in the time resolved differential absorption spectrum in the 430 to 520 nm
region indicates that the ground state extinction coefficient is larger than the CNMe-bpy radical absorption in that wavelength region. The intense absorption in
the UV region of the transient absorption spectrum is due to the CN-Me-bpy
radical anion. The fact that the time resolved differential absorption spectrum is
a net absorption in this region indicates that the extinction coefficient of the CNMe-bpy excited state absorption is significantly larger than the ground state
MLCT absorption features in the region of 360 to 410 nm.

For probe

wavelengths blue of 360 nm both the oxidative and reductive contributions give
rise to a positive absorption feature.

Finally, wavelengths red of the ground

state/excited state isosbestic point at 520 nm has contributions from both the CNMe-bpy radical anion as well as LMCT transition(s) from the ancillary bpy ligands.
Corresponding data for complexes 2 and 3 can be found in Figures S4 and S5 in
the supporting information. While there are qualitative differences in the spectra,

90

due to differences in the ground state absorption spectra, the same conclusions
can be drawn for the transient absorption spectra of complexes 2 and 3.

3.3.5 Nanosecond Step-Scan Time-Resolved Infrared Spectroscopy
As mentioned in the introduction, the primary goal in developing this series of
compounds was to use them as probes for vibrational relaxation dynamics in
charge-transfer excited states. Similar to our protocol for studying the ultrafast
absorption spectroscopy of transition metal chromophores,66 it is important to
fully characterize the spectroscopic properties of the long lived excited state as a
foundation for interpreting data acquired on shorter timescales. For vibrational
spectroscopy, both time-resolved resonance Raman (TR3) and time-resolved
infrared (TRIR) methods have been developed; each technique has advantages
and disadvantages. In TR3 visible light is used to create as well as resonantly
scatter off the excited state. The resonance condition means that only those
vibrations which are coupled to the excited electronic transition are enhanced.
This can vastly simplify the spectrum and assigning the nature of the excited
state, but at the same time gives only a partial picture of the excited state
vibrational structure. From a technical perspective, interference from emission
can make using this technique on highly emissive compounds challenging. With
TRIR the entire vibrational spectrum of the excited state is sampled using an
infrared probe beam: this provides a more complete picture of the excited-state
vibrational structure but can make interpretation more difficult. TRIR has been
used to characterize the vibrational structure of long lived excited states under a
variety of circumstances.68-70 Previous work from our group employed TRIR on
a series of heteroleptic ruthenium polypyridyl complexes to identify on which
ligand the emissive excited state was localized.19 Relevant to the present study,
Meyer, Palmer, and coworkers examined a series of ruthenium polypyridyl

91

Table 3-4: Step-Scan IR and DFT Results for Complexes 1-3
Step-Scan IR

ν̅(CN)GS (cm-1)a

ν̅(CN)3MLCT (cm-1)b

Δν̅ (CN) (cm-1)c

(1)

2238

2200

-38

(2)

2238

2200

-38

(3)

2239

2202

-37

DFT

ν̅(CN)GS (cm-1)d,e

ν̅(CN) 3MLCT (cm-1)e,f

Δν̅ (CN) (cm-1)c

(1)

2272.01 (as)
2272.73 (s)

2216.06 (as)
2237.52 (s)

-56 (as)
-35 (s)

a Ground state absorption maximum in CH NO solution (4 cm-1 resolution).
3
2
b Excited state differential absorption maximum in CH NO solution (4 cm-1
3
2
resolution). c Δν̅ (CN) = ν̅(CN)3MLCT - ν̅(CN)GS. d Frequency calculation results
on the ground state optimized geometry. e (s) is the totally symmetric CN

stretching frequency, (as) is the asymmetric CN stretching frequency.
f Frequency calculation results on the lowest energy triplet state optimized
geometry.

complexes substituted with diethyl ester and diethyl amide groups as infrared
tags.71,72 These works were able to demonstrate localization of the long-lived
excited state on a single bipyridine ligand for both the heteroleptic and
homoleptic complexes. Interestingly their data and analysis clearly showed that
in the case of 4-monosubstituted bipyridine ligands, the long-lived excited state
was primarily localized on the substituted pyridine ring rather than being
symmetrically distributed across the entire ligand.
The nanosecond step-scan IR spectra of complex 1 are shown in Figure 3-9.
The ground state exhibits a single CN band at 2238 cm-1. A group theoretical
analysis indicates that both the A1 (symmetric) and B2 (asymmetric) CN
vibrations of the CN-Me-bpy ligand should be IR active.

Ground state DFT

frequency calculations on complex 1 (Table 3-4) does predict two bands, but they
are calculated to be separated by < 1cm-1. The single, fairly sharp feature in

92

-1

Energy (cm )
2250

Absorbance

(a)

2100

0.03
0.02

0.10

0.01

0.05

0.03

0.04

v(CN)*

0.15

0.00

Change in Absorbance

v(CN)

2150

0.00

v(CN)

(b)

Change in Absorbance

0.20

2200

150 ns
200 ns
250 ns
350 ns
500 ns
750 ns
1000 ns
1500 ns
2000 ns

0.02
0.01
0.00
2250

2200

2150

2100

-1

Energy (cm )

Figure 3-9: Steady-state and time-resolved infrared absorption data for
[Ru(bpy)2(CN-Me-bpy)](PF6)2 (1) in room-temperature CH3NO2 solution. (a)
Comparison of the ground-state (blue line) and step-scan infrared differential
absorption data acquired at a time delay of 150 ns following ca. 10 ns excitation
at 540 nm (red line). The ~40 cm-1 red-shift in the CN stretching frequency
reflects the presence of an electron in the π* orbital of the CN-Me-bpy ligand in
the 3MLCT excited state. (b) Nanosecond step-scan infra-red spectra as a
function of time following ca. 10 ns excitation at 540 nm. The kinetics describing
the decrease in amplitude of the excited-state infrared absorption signal are
within experimental error of the time-resolved emission and absorption data for
compound 1 and are therefore assigned to relaxation of the 3MLCT excited state.

93

Figure 3-9a therefore presumably contains both of these vibrations; their
accidental degeneracy is an indication of (relatively) weak coupling of these two
modes in the ground state. This is in stark contrast to the vibrational spectra in
the 3MLCT excited state. The CN stretching frequency is observed to shift 38
cm-1 lower in energy relative to the ground state.

Moreover the band is

significantly broadened. The bathochromic shift is consistent with the population
of a π* antibonding orbital, which if coupled to the cyano group should reduce the
CN bond order; the magnitude of the shift is comparable to those seen by Meyer
and coworkers for amide and ester substituted systems.71,72
The increased breadth of the absorption suggests that the coupling between
the two cyano groups is greater in the excited state, leading to a larger difference
in energy between the symmetric and asymmetric stretches. This interpretation
is supported by DFT frequency calculations on the lowest energy triplet state of
complex 1 which show that Δν̅ has increased to ~20 cm-1 in the emissive
3MLCT state. This can be explained by considering the interaction of the π-type
orbitals of the cyanide groups with the π-type orbitals of the bipyridine ring. For
complex 1, Figure 3-10 shows the highest energy π bonding orbital and lowest
energy π* antibonding orbital localized on the CN-Me-bpy ligand. Figure 3-10a
shoes that the π bonding orbitals of the cyanide groups have little or no
interaction with the π bonding orbital of the bipyridine ring and therefore there is
little coupling between the cyanide groups and the bipyridine ring and between
the two cyanide groups in the ground state. Figure 3-10b shows that the π*
antibonding orbitals of the cyanide groups interact much more strongly with the
π* antibonding orbital on the bipyridine ring, leading to an increased coupling
between the cyanide groups and the bipyridine ring and between the two cyanide
groups in the excited state.

The difference in mixing is consistent with the

energies of the π bonding and π* antibonding orbitals on the cyanide group. The

94

C N bond is significantly stronger than the C=C (and C=N) bonds within the
bipyridine ring. While the cyanide π* antibonding orbital is of an appropriate
energy and symmetry to mix well with the π* orbital on the bipyridine ring, the π
bonding orbital of the cyanide is lower in energy than the π orbital on the
bipyridine ligand, which leads to weaker (or no) mixing between the two.

(a)

(b)

Figure 3-10: (a) Highest energy π bonding orbital on CN-Me-bpy in the geometry
optimized ground state of complex 1. (b) Lowest energy π* antibonding orbital on
CN-Me-bpy in the geometry optimized lowest energy triplet state of complex 1.

The SSIR results for complexes 2 and 3 are summarized in Table 3-4; the
spectra can be found in Figures A-8 and A-9. Within experimental error the
ground state and excited state absorption maxima are at the same energy in all
three complexes. This is compelling evidence that the long-lived excited state is
localized on a single CN-Me-bpy ligand in all three complexes rather than being
delocalized over multiple ligands in complexes 2 and 3. This is in agreement
with previous step-scan IR studies71,72 as well other experimental evidence
showing a localized long lived excited state in ruthenium polypyridyl systems.

95

3.4 Concluding Comments.
This work describes the synthesis and spectroscopic characterization of a
series of cyano substituted ruthenium polypyridyl complexes.

While these

complexes could not be synthesized in reasonable amounts by combining the
ligand and ruthenium starting material under standard conditions, the chemistryon-the-complex approach proved useful.

By first synthesizing the chloro

substituted ruthenium complexes and using a palladium catalyzed cyanation
reaction it was possible to prepare and isolate all three CN-containing members
of the [Ru(bpy)3-n(CN-Me-bpy)n](PF6)2 series with reasonable yield and of a high
degree of purity.
The ground- and 3MLCT-state properties of the [Ru(bpy)3-n(CN-Me-bpy)n]2+
series have been characterized using electrochemical, optical, and infrared
spectroscopies. As expected, the CN stretch of the CN-Me-bpy ligand sits in a
well-isolated region of the infra-red, allowing for an unambiguous analysis of its
behavior in the ground and excited state(s) of these compounds. The data we
have presented definitively show that the lowest energy charge-transfer states in
all three compounds are localized on the CN-containing ligand. In particular,
time-resolved electronic and infra-red absorption spectroscopies reveal an
identical optical signature for the 3MLCT state for each compound in the series;
in the case of the infra-red measurements, the common shift in the CN stretching
frequency indicates that (a) the CN group is strongly coupled to the π* system of
the bipyridyl ligand, and (b) that the 3MLCT state is localized on a single ligand
on the vibrational time scale, even in the case of the tris-homoleptic complex
[Ru(CN-Me-bpy)3]2+. The unexpected increase in spectral bandwidth of the CN
vibration(s) further revealed that the two CN groups of the ligand are more
strongly coupled to each other in the 3MLCT state relative to the ground state.
All of the features just enumerated underscore the potential utility of these

96

compounds as probes of vibrational dynamics in charge-transfer excited states.
With the ground- and excited-state spectroscopic features of this series fully
characterized, we believe that an examination of vibrational relaxation as well as
the more elusive process of intramolecular vibrational redistribution will now be
accessible through the combined application of femtosecond time-resolved
electronic and infra-red absorption spectroscopies. Studies along these lines are
currently underway.

97

REFERENCES

98

3.5 References
(1)

Link, G.; Heinen, U.; Berthold, T.; Ohmes, E.; Weidner, J. U.; Kothe, G. Z.
Phys. Chem. 2004, 218, 171-191.

(2)

Santabarbara, S.; Galuppini, L.; Casazza, A. P. J. Integr. Plant Biol. 2010,
52, 735-749.

(3)

Hagfeldt, A.; Grätzel, M. Acc. Chem. Res. 2000, 33, 269-277.

(4)

Gratzel, M. Inorg. Chem. 2005, 44, 6841-6851.

(5)

Hamann, T. W.; Jensen, R. A.; Martinson, A. B. F.; Ryswyk, H. V.; Hupp, J.
T. Energy Environ. Sci. 2008, 1, 66-78.

(6)

Grätzel, M. Acc. Chem. Res. 2009, 42, 1788-1798.

(7)

Cao, Y.; Bai, Y.; Yu, Q.; Cheng, Y.; Liu, S.; Shi, D.; Gao, F.; Wang, P. J.
Phys. Chem. C. 2009, 113, 6290-6297.

(8)

Bessho, T.; Zakeeruddin, S. M.; Yeh, C. Y.; Diau, E. W.; Grätzel, M. Angew.
Chem. Int. Ed. Engl. 2010, 49, 6646-6649.

(9)

Magnuson, A.; Anderlund, M.; Johansson, O.; Lindblad, P.; Lomoth, R.;
Polivka, T.; Ott, S.; Stensjo, K.; Styring, S.; Sundstrom, V.; Hammarstrom,
L. Acc. Chem. Res. 2009, 42, 1899-1909.

(10) Concepcion, J. J.; Jurss, J. W.; Brennaman, M. K.; Hoertz, P. G.;
Patrocinio, A. O.; Murakami Iha, N. Y.; Templeton, J. L.; Meyer, T. J. Acc.
Chem. Res. 2009, 42, 1954-1965.
(11) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.; Lee, H. E.;
Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J. Am. Chem.
Soc. 2001, 123, 4304-4312.
(12) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B. Top. Curr. Chem. 2007,
281, 143-203.
(13) Hofbeck, T.; Yersin, H. Inorg. Chem. 2010, 49, 9290-9299.
(14) Juban, E. A.; McCusker, J. K. J. Am. Chem. Soc. 2005, 127, 6857–6865.
(15) Záliš, S.; Busby, M.; Kotrba, T.; Matousek, P.; Towrie, M.; Vlček, A. A.
Inorg. Chem. 2004, 43, 1723–1734.

99

(16) Maçôas, E. M.; Kananavicius, R.; Myllyperkiö, P.; Pettersson, M.; Kunttu, H.
J. Am. Chem. Soc. 2007, 129, 8934–8935.
(17) Browne, W. R.; Coates, C. G.; Brady, C.; Matousek, P.; Towrie, M.;
Botchway, S. W.; Parker, A. W.; Vos, J. G.; McGarvey, J. J. J. Am. Chem.
Soc. 2003, 125, 1706-1707.
(18) Henry, W.; Coates, C. G.; Brady, C.; Ronayne, K. L.; Matousek, P.; Towrie,
M.; Botchway, S. W.; Parker, A. W.; Vos, J. G.; Browne, W. R.; McGarvey,
J. J. J. Phys. Chem. A. 2008, 112, 4537-4544.
(19) Curtright, A. E.; McCusker, J. K. J. Phys. Chem. A. 1999, 103, 7032–7041.
(20) Lian, T.; Bromberg, S. E.; Asplund, M. C.; Yang, H.; Harris, C. B. J. Phys.
Chem. 1996, 100, 11994–12001.
(21) Farrell, I. R.; Matousek, P.; Vlček, A. A. J. Am. Chem. Soc. 1999, 121,
5296–5301.
(22) Portius, P.; Yang, J.; Sun, X.; Grills, D. C.; Matousek, P.; Parker, A. W.;
Towrie, M.; George, M. W. J. Am. Chem. Soc. 2004, 126, 10713–10720.
(23) Liard, D. J.; Busby, M.; Matousek, P.; Towrie, M.; Vlček, A. A. J. Phys.
Chem. A. 2004, 108, 2363-2369.
(24) Lindquist, B. A.; Furse, K. E.; Corcelli, S. A. Phys. Chem. Chem. Phys.
2009, 11, 8119-8132.
(25) Doorn, S. K.; Dyer, R. B.; Stoutland, P. O.; Woodruff, W. H. J. Am. Chem.
Soc. 1993, 115, 6398-6405.
(26) Wang, C.; Mohney, B. K.; Akhremitchev, B. B.; Walker, G. C. J. Phys.
Chem. A. 2000, 104, 4314–4320.
(27) Easun, T. L.; Alsindi, W. Z.; Deppermann, N.; Towrie, M.; Ronayne, K. L.;
Sun, X. Z.; Ward, M. D.; George, M. W. Inorg. Chem. 2009, 48, 8759-8770.
(28) Simpson, P. G.; Vinciguerra, A.; Quagliano, J. V. Inorg. Chem. 1963, 2,
282-286.
(29) Kavanagh, P.; Leech, D. Tetrahedron Lett. 2004, 45, 121-123.
(30) Mukkala, V. M.; Kankare, J. J. Helv. Chim. Acta. 1992, 75, 1578-1592.

100

(31) Anderson, S.; Constable, E. C.; Seddon, K.; Turp, J. E.; Baggott, J. E.;
Pilling, M. J. J. Chem. Soc. Dalton Trans. 1985, 2247–2261.
(32) Wenkert, D.; B., W. R. J. Org. Chem. 1983, 48, 283-289.
(33) Evans, I. P.; Spencer, A.; Wilkinson, G. J. Chem. Soc. Dalton Trans. 1973,
204-209.
(34) Sullivan, B. P.; Salmon, D. J.; Meyer, T. J. Inorg. Chem. 1978, 17,
3334-3341.
(35) Wang, J.; Fang, Y.; Hanan, G. S.; Loiseau, F.; Campagna, S. Inorg. Chem.
2005, 44, 5-7.
(36) Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278-1285.
(37) Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara, T.; Ishida,
H.; Shiina, Y.; Oishi, S.; Tobita, S. Phys. Chem. Chem. Phys. 2009, 11,
9850-9860.
(38) Damrauer, N. H.; McCusker, J. K. Inorg. Chem. 1999, 38, 4268–4277.
(39) Claude, J. P.; Meyer, T. J. J. Phys. Chem. 1995, 99, 51-54.
(40) Parker, C. A.; Rees, W. T. Analyst (London). 1960, 85, 587-600.
(41) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J. Chem.
Soc. 1997, 119, 8253–8268.
(42) Picraux, L. B.; Smeigh, A. L.; Guo, D.; McCusker, J. K. Inorg. Chem. 2005,
44, 7846–7865.
(43) Sun, H.; Frei, H. J. Phys. Chem. B. 1997, 101, 205–209.
(44) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J., J. A.; Vreven, T.; Kudin, K. N.; Burant,
J. C.; Millam, J. M. ; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.;
Cossi, M. ; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;

101

Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,
J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin,
R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, revision D.01; Gaussian, Inc.,
Wallingford CT, 2004
(45) Monat, J. E.; Rodriguez, J. H.; McCusker, J. K. J. Phys. Chem. A. 2002,
106, 7399–7406.
(46) Fodor, L.; Lendvay, G.; Horváth, A. J. Phys. Chem. A. 2007, 111, 12891–
12900.
(47) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; Von
Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85-277.
(48) Anderson, P. A.; Deacon, G. B.; Haarmann, K. H.; Keene, F. R.; Meyer, T.
J.; Reitsma, D. A.; Skelton, B. W.; Strouse, G. F.; Thomas, N. C. Inorg.
Chem. 1995, 34, 6145-6157.
(49) Cook, M. J.; Lewis, A. P.; McAuliffe, G. S. G.; Skarda, V.; Thomson, A. J.;
Glasper, J. L.; Robbins, D. J. J. Chem. Soc. Perkin Trans. 1984, 417,
1293-1301.
(50) Furue, M.; Maruyama, K.; Oguni, T.; Naiki, M.; Kamachi, M. Inorg. Chem.
1992, 31, 3792-3795.
(51) Masui, H.; Murray, R. W. Inorg. Chem. 1997, 36, 5118-5126.
(52) Losse, S.; Gorls, H.; Groarke, R.; Vos, J. G.; Rau, S. Eur. J. Inorg. Chem.
2008,
(53) Pichot, F.; Beck, J. H.; Elliott, C. M. J. Phys. Chem. A. 1999, 103,
6263-6267.
(54) Constable, E. C.; Thompson, A. M. W. C.; Harveson, P.; Macko, L.;
Zehnder, M. Chem. Eur. J. 1995, 1, 360-367.
(55) Chodorowski-Kimmes, S.; Beley, M.; Collin, J. P.; Sauvage, J. P.
Tetrahedron Lett. 1996, 37, 2963-2966.
(56) Aspley, C. J.; Williams, J. A. G. New J. Chem. 2001, 25, 1136-1147.
(57) Fang, Y. Q.; Polson, M. I.; Hanan, G. S. Inorg. Chem. 2003, 42, 5-7.

102

(58) Kozlov, D. V.; Tyson, D. S.; Goze, C.; Ziessel, R.; Castellano, F. N. Inorg.
Chem. 2004, 43, 6083-6092.
(59) Pomestchenko, I. E.; Polyansky, D. E.; Castellano, F. N. Inorg. Chem.
2005, 44, 3412-3421.
(60) Jin, F.; Confalone, P. N. Tetrahedron Lett. 2000, 41, 3271-3273.
(61) Weissman, S. A.; Zewge, D.; Chen, C. J. Org. Chem. 2005, 70, 1508-1510.
(62) Vlček, A. A.; Dodsworth, E. S.; Pietro, W. J.; Lever, A. B. P. Inorg. Chem.
1995, 34, 1906-1913.
(63) Mabrouk, P. A.; Wrighton, M. S. Inorg. Chem. 1986, 25, 526–531.
(64) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys. Chem.
1986, 90, 3722-3734.
(65) Kestner, N. R.; Logan, J.; Jortner, J. J. Phys. Chem. 1974, 78, 2148-2166.
(66) McCusker, J. K. Acc. Chem. Res. 2003, 36, 876–887.
(67) Nazeeruddin, M. K.; Zakeeruddin, S. M.; Kalyanasundaram, K. J. Phys.
Chem. 1993, 97, 9607-9612.
(68) Omberg, K. M.; Schoonover, J. R.; Treadway, J. A.; Leasure, R. M.; Dyer,
R. B.; Meyer, T. J. J. Am. Chem. Soc. 1997, 119, 7013–7018.
(69) Omberg, K. M.; Schoonover, J. R.; Bernhard, S.; Moss, J. A.; Treadway, J.
A.; Kober, E. M.; Dyer, R. B.; Meyer, T. J. Inorg. Chem. 1998, 37, 3505–
3508.
(70) Yeom, Y.; Frei, H. In-situ Spectroscopy of Catalysts; American Scientific
Publishers: 2004; 32–46.
(71) Chen, P.; Omberg, K. M.; Kavaliunas, D. A.; Treadway, J. A.; Palmer, R. A.;
Meyer, T. J. Inorg. Chem. 1997, 36, 954–955.
(72) Omberg, K. M.; Smith, G. D.; Kavaliunas, D. A.; Chen, P.; Treadway, J. A.;
Schoonover, J. R.; Palmer, R. A.; Meyer, T. J. Inorg. Chem. 1999, 38, 951–
956.

103

Chapter 4: Probing Vibrational Relaxation Dynamics With Ultrafast
Electronic and Infrared Absorption Spectroscopy

4.1 Introduction
A number of important photophysical and photochemical processes occur on
ultrafast timescales.1-5 The very rapid rates of these reactions suggest that they
are occurring from vibrationally excited "hot" states and are possibly in
competition with vibrational relaxation of that hot state. In most charge transfer
complexes, a majority of the energy from an absorbed photon is dissipated nonradiatively and the relaxation from the initially excited state to the long lived
excited state is entirely a non-radiative process.
Despite its important role in excited state dynamics, there has not been much
investigation into the mechanism of non-radiative decay in charge transfer
excited states of transition metal complexes. As discussed in Chapter 1, there is
even question as to the timescale for ruthenium(II) polypyridyl systems. Chergui
and coworkers conclude from their fluorescence upconversion study on
[Ru(bpy)3]2+ that the thermalized 3MLCT state is formed within 15 ± 10 fs and
that vibrational relaxation occurs by very rapid IVR to low frequency vibrational
modes, such as metal ligand bond stretches, followed by slower transfer to
solvent.6

Contradicting this, Browne and McGarvey monitored the growth of

vibrational features of the thermalized 3MLCT state in [Ru(bpy)3]2+ with time
resolved resonance Raman. Their experiments showed that the formation of the
thermalized 3MLCT state was complete after 20 ps.7

A picosecond rate of

vibrational relaxation has also been observed in many other transition metal
complexes,8-13 which suggests that the interpretation of Browne, McGarvey and
coworkers is the correct one

104

(a)

(b)
ES2

ES2

ES1

ES1

Figure 4-1: Potential energy surfaces illustrating the available Franck-Condon
transitions between the optically prepared excited state (ES1) and a higher lying
excited state (ES2) when (a) ES1 is in a excited vibrational (hot) state and (b)
ES1 is in a thermalized vibrational (cooled) state

In ultrafast spectroscopy, information about vibrational relaxation from the
initially excited Franck-Condon state to the long lived excited state is generally
inferred from small changes in electronic absorption spectroscopy.12 After all
possible electronic transitions are assigned, small changes in the spectrum over
time can be assigned to vibrational relaxation. These changes are usually seen
as a narrowing of an absorption feature over time, but the nature of the change
depends on the relative displacement and slope of the two excited state potential
energy surfaces (Figure 4-1).

A more direct way to monitor the vibrational

relaxation is to use infrared spectroscopy, where the vibrational state of the
molecule is directly monitored. With infrared spectroscopy, vibrational relaxation
is generally observed as a narrowing and blue shifting of the infrared spectrum.
105

This is because the Franck-Condon excitation creates a distribution of
vibrationally excited states which all relax to a single νʼ = 0 vibrational state which
causes the spectrum to narrow and the anharmonic nature of the potential
surface means the vibrational spacing is smaller at the top of the potential well
and larger at the bottom, leading to a blue shift.

Because transitions are

occurring within a single electronic state, rather than between two electronic
states, the nature of the signal is only a function of a single electronic state.
Electronic absorption and infrared spectroscopy can offer complementary
information about excited state dynamics. For example, work in our group on the
ultrafast absorption spectroscopy of Cr(acac)3 (acac = acetylacetonate) showed
that upon excitation into the 4T2 ligand field state, the 2E state was formed very
rapidly (>100 fs).12

Monitoring excited state absorptions of the vibrational

relaxation time on 2E surface was 1.1 ± 0.01 ps and ground state recovery was
~700 ps.

Maçôas and coworkers studied Cr(acac)3 by ultrafast infrared

spectroscopy, monitoring the excited state ligand C=C and C=O stretches, and
found biexponential ground state recovery kinetics.14 A majority of the excited
state population (70-85%) had a ground state recovery time of 15 ps and a small
fraction has a ground state recovery rate of 700-900 ps. Using the results from
the electronic absorption spectroscopy above, the authors theorized that the 15
ps component was ground state recovery from the 4T2 state, which was formed
through back intersystem crossing, and the 700-900 ps component was the
same ground state recovery from the 2E state seen in the electronic absorption
spectroscopy.
Maçôas and coworkers have also collected time-resolved infrared data on
Fe(acac)3.13 Unlike the chromium system, there was not any reported timeresolved electronic absorption available for the iron system.

Without any

information about the electronic excited state(s) involved, the authors were

106

unable to definitively assign any of the kinetic features beyond ground state
recovery. These examples show that time-resolved electronic absorption and
time-resolved infrared absorption are complementary experimental techniques.
When used together, the two techniques can give more information about the
excited state dynamics than either technique alone can provide.

CN

CN

N
N

N
Ru
N

CN

NC

N

N

N

N
CN

N
Ru
N

CN

NC
N

N

N

N
CN

NC

N
Ru
N

CN
N
N
CN

CN
(1)

(2)

(3)

Figure 4-2: The series of cyano-substituted ruthenium(II) complexes investigated
in this chapter.

This chapter discusses vibrational relaxation in the series of cyanosubstituted ruthenium(II) bipyridine complexes introduced in Chapter 3, shown in
Figure 4-2. Computational methods are used to identify the nature of the initially
excited Franck-Condon state and the long lived, thermalized 3MLCT states of the
three complexes have been characterized in Chapter 3. In this chapter, both
ultrafast electronic absorption and infrared absorption are used monitor the
relaxation process from the initially excited state to the long lived excited state.

107

4.2 Experimental
4.2.1 Calculations
Calculations on complexes 1-3 were performed using the Gaussian 03
software package.15 Geometry optimizations were done on the ground state
using a spin unrestricted formalism at the B3LYP/LANL2DZ level of theory.16,17
No symmetry restrictions were placed on the geometry optimizations.

The

effects of the acetonitrile solvent environment were included by using the
polarizable continuum model (PCM). Frequency calculations were performed on
the optimized structures to ensure that these geometries corresponded to global
minima.

No imaginary frequencies were obtained for any of the optimized

geometries.

The first 80 electronic transitions (singlets and triplets) of the

optimized structures were calculated by time dependent DFT using spin
unrestricted B3LYP/LANL2DZ. The orbitals contributing to each transition were
visualized using GaussView.

The cartesian coordinates for the optimized

geometries can be found in Appendix D.

4.2.2 Time-Resolved Spectroscopic Measurements
Time-resolved infrared absorption data was collected by Professor Antonín
Vlček at the Rutherford Appleton Laboratory. A description of the experimental
setup and data collection can be found in the literature18,19.

Samples were

prepared by dissolving the complexes in nitromethane so that, in a 0.1 mm
pathlength cell, the absorbance was approximately 0.7 at the excitation
wavelength. This resulted in a ground state infrared absorption of 0.005 to 0.01
for the CN vibration. The samples were excited at two wavelengths: 400 nm and
490 (1 and 2) or 475 (3). The raw data was background corrected with a spline
function in IGOR pro. The energy axis was calibrated using Equation 4-1, where
d is dispersion, a and b are pixels, and ν̅a and ν̅b are the energies (in cm-1) of

108

pixels a and b and Equation 4-2, where ν̅i is the energy (in cm-1) of pixel i, λa is
the wavelength of pixel a (in cm). Kinetic traces were created by plotting the
peak intensity or peak area vs time.
1
1
!
!a !b
¯
¯
d=
b! a
!i =
¯

1
[! a + d( a " i)]

4-1

4-2

Time resolved electronic absorption data was collected by Allison Brown here
at Michigan State University. A general description of the experimental setup
and data collection and processing has previously been published.12,20.
Spectrophotometric grade acetonitrile, which had been freeze-pump-thaw
degassed prior to use, was used as the solvent. Samples were prepared with an
absorbance of 0.4 - 0.7 at the excitation wavelength in air-free 1 mm pathlength
cells.

Stability of the samples was monitored by electronic absorption

spectroscopy before and after each measurement. Compounds 1 and 2 showed
no changes in their UV-visible absorption spectra and no other signs of
decomposition, so the samples were used for several days before replacing.
Compound 3 showed a small change in the UV-vis spectrum over a two day
period, so a fresh sample of 3 was prepared each day.

4.3 Results and Discussion
4.3.1 Time Dependent DFT
In the heteroleptic complexes (1 and 2) it is possible for the Franck-Condon
state to be localized on either of the two ligands.

Interpretation of the

spectroscopic data requires knowledge of where the initial excited state is
localized for each excitation wavelength.

109

Initial attempts to use resonance

Raman spectroscopy to answer this question were unsuccessful because of
interference from the sample emission.

Instead, time dependent DFT

calculations were used to visualize the orbital contributions to each transition.
The ground state geometries for each complex were first optimized in an
acetonitrile dielectric continuum as described in Chapter 3. These optimized
geometries were used as the starting point for the time dependent DFT
calculations. For each structure the first 80 transitions (singlets and triplets) were
calculated. Figure 4-3 shows the results for complex 1. The blue squares are
the calculated singlet transitions, the red triangles are the calculated triplet
transitions, which have a calculated oscillator strength of 0 due to their formally
spin forbidden nature, and the black line is the experimental electronic absorption
spectrum measured in acetonitrile.
The first thing to notice is how well the calculated transitions align with the
experimental absorption spectrum without any of the corrections that are
sometimes necessary for calculations of this sort.16 This excellent agreement
with experimental results is evidence for the accuracy of the TD-DFT calculation.
The calculations show that the predictions from electrochemistry in Chapter 3 are
correct and the lowest energy absorption band is a metal-to-ligand charge
transfer to the CN-Me-bpy. The orbital pictures for that transition shows that the
transition is from a (mostly) metal based orbital to a (mostly) ligand based orbital.
The selected higher energy MLCT transition has two approximately equal
contributions, one from a metal based orbital to an orbital localized on the two
bpy ligands and one from a metal based orbital to an orbital that is somewhat
delocalized over all three ligands, but mostly localized onto the two bipyridine
ligands.

The third major absorption band is an intraligand transition.

The

dominant orbital contribution to that transition is from a CN-Me-bpy π bonding
orbital to a CN-Me-bpy π* antibonding orbital. This means that with the 490 nm

110

600

50

0.5
40

0.4

30

0.3
0.2

20

0.1

10

3

4.0

3.5

3.0
Energy (eV)

2.5

2.0

0

-1

0.0

-1

Oscillator Strength (f)

350

Molar Absortivity (x10 M cm )

300

Wavelength (nm)
400
450 500

Figure 4-3: TD-DFT results for complex 1 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory.
Acetonitrile solvent
environment was modeled with the polarizable continuum model (PCM). The red
triangles are calculated triplet transitions the blue squares are calculated singlet
transitions, and the black line is the experimental electronic absorption spectrum
in CH3CN. Pictured orbitals are the major contributions to the selected
transitions.

111

500

600

0.6

50

0.5

40

0.4

30

0.3

20

0.2

3

10

0.1
3.5

3.0
Energy (eV)

2.5

0
2.0

-1

4.0

cm )

0.0

-1

Oscillator Strength (f)

300

Molar Absortivity (x10 M

0.7

Wavelength (nm)
350
400
450

Figure 4-4: TD-DFT results for complex 2 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory.
Acetonitrile solvent
environment was modeled with polarizable continuum model (PCM). The red
triangles are calculated triplet transitions the blue squares are calculated singlet
transitions, and the black line is the experimental electronic absorption spectrum
in CH3CN. Pictured orbitals are the major contributions to the selected
transitions.

112

500

0.5

80
60

0.4
0.3

40

0.2

20

-1

0.1
3.0
Energy (eV)

2.5

0
2.0

-1

3.5

cm )

0.0
4.0

600

3

Oscillator Strength (f)

350

Molar Absortivity (x10 M

0.6

Wavelength (nm)
400
450

Figure 4-5: TD-DFT results for complex 3 calculated using a spin unrestricted
formalism at the B3LYP/LANL2DZ level of theory.
Acetonitrile solvent
environment was modeled with polarizable continuum model (PCM). The red
triangles are calculated triplet transitions the blue squares are calculated singlet
transitions, and the black line is the experimental electronic absorption spectrum
in CH3CN. Pictured orbitals are the major contributions to the selected
transitions.

113

excitation wavelength the Franck-Condon state will be localized on the CN-Mebpy.

The higher energy excitation wavelengths, 400 nm for the infrared

absorption and 425 for the visible absorption, will generate a Franck-Condon
state primarily localized on the unsubstituted bipyridine ligand(s). Calculations
on complex 2 (Figure 4-4) show similar results. In complex 3 there is only one
possible ligand for the initial excited state to be localized on. The computational
results (Figure 4-5) show that both the high and low energy excitation will
generate metal-to-ligand charge transfer states localized on either one or two of
the CN-Me-bpy ligands.
4.3.2 Time-Resolved Infrared Spectroscopy
Vibrational relaxation in transition metal complexes can appear in different
ways in vibrational spectroscopy. In complexes with a low density of vibrational
states, like the W(CO)5L complexes studied by Vlček and coworkers, vibrational
relaxation is seen as a narrowing and blueshifting of the excited state
absorption.10

The spectrum narrows because multiple Franck-Condon

transitions are allowed, creating a distribution of vibrational levels in the initial
excited state which all relax to a single ν' = 0 vibrational level. The anharmonic
nature of the potential surface means that the vibrational spacing is smaller at the
top of the potential well and larger at the bottom, which leads to the vibrational
transition shifting to higher energy as the system cools. In these types of low
density vibrational density metal complexes, each excited state absorption
generally corresponds to a single vibrational mode. In complexes with higher
densities of vibrational states, the observed absorption features are often a
superposition of many closely spaced vibrational modes. The work of Browne,
McGarvey, and coworkers on the time resolved resonance Raman spectroscopy
of [Ru(bpy)3]2+ shows vibrational relaxation in a system with high density of

114

0.0025
0.0020
0.0015
0.0010
0.0005

0.005

(a)

0 ps
1 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

Change in Absorbance

Change in Absorbance

0.0030

0.0000

0.003
0.002
0.001
0.000
-0.001

-0.0005
2400

0.004

(b)

0 ps
1 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

2300

2200
-1

2100

2400

2300

2200
-1

2100

Energy (cm )

Energy (cm )

Figure 4-6: Time resolved infrared spectra collected for complex 1 in a
nitromethane solution. (a) 490 nm excitation wavelength where the initial excited
state is localized on the CN-Me-bpy ligand (b) 400 nm excitation wavelength
where the initial excited state is localized on the unsubstituted bpy ligand.

vibrational states. The authors observe a increase in peak intensity over time
without any significant changes in the peak shape or energy.
The time resolved infrared absorption spectra for all three complexes were
collected at two different excitation wavelengths, 490 (1 and 2) or 475 (3) and
400 nm. The Franck-Condon state with the 400 nm excitation (for complexes 1
and 2) will be localized on the unsubstituted bipyridine ligand and the lower
energy excitation will create a Franck-Condon state that is localized on the CNMe-bpy ligand. The data for complex 1 can be seen in Figure 4-6. The most
notable observation is that there is little or no difference between the 490 nm and
the 400 nm spectra. In both spectra, there is an increase in the intensity of the
excited state absorption feature without any measurable change in the energy or
shape of the peak, similar to what Browne, McGarvey and coworkers observed
115

for [Ru(bpy)3]2+. A kinetic analysis shows that the rise time of the peak in both
spectra is approximately 2 ps. Figure 4-9 shows the change in excites state
absorption peak intensity over time for the 400 nm spectra and similar results are
seen for the 490 nm data.

0.0015
0.0010
0.0005

0 ps
1 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

(a)

0.003

Change in Absorbance

Change in Absorbance

0.0020

0.0000
-0.0005

0.002

0.001

(b)

0 ps
1.0 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

0.000

-0.001

-0.0010
2400

2300

2200
-1

2100

Energy (cm )

2400

2300

2200
-1

2100

Energy (cm )

Figure 4-7: Time resolved infrared measurements on complex 2. in a
nitromethane solution. (a) 490 nm excitation wavelength where the initial excited
state is localized on the CN-Me-bpy ligand (b) 400 nm excitation wavelength
where the initial excited state is localized on the unsubstituted bpy ligand.

As discussed in Chapter 3, when the excited state is localized on the CN-Mebpy ligand the excited electron is in a π* antibonding orbital of the CN-Me-bpy
ligand. The cyanide group is conjugated into the π system of the bipyridine rings,
so populating a π* antibonding orbital leads to a weaker CN bond and lower
frequency vibration. When the excited state is localized on the unsubstituted
bipyridine ligand, the excited state cyanide stretching frequency would be

116

0 ps
1 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

0.0015
0.0010
0.0005

(a)

0 ps
1.0 ps
1.2 ps
1.6 ps
2 ps
3 ps
5 ps
8 ps
10 ps
20 ps
50 ps
100 ps

0.003

Change in Absorbance

Change in Absorbance

0.0020

0.0000
-0.0005

0.002

0.001

(b)

0.000

-0.001

-0.0010
2400

2300

2200
-1

2400

2100

2300

2200
-1

2100

Energy (cm )

Energy (cm )

Figure 4-8: Time resolved infrared measurements on complex 3. in a
nitromethane solution. (a) 475 nm excitation wavelength where the initial excited
state is localized on the CN-Me-bpy ligand (b) 400 nm excitation wavelength
where the initial excited state is localized on the CN-Me-bpy bpy ligand.

Change in Absorbance

0.005
0.004
0.003
0.002
0.001
0.000
0

20

40

60

80

100

Time (ps)

Figure 4-9: Kinetic trace of the absorption maximum (2200 cm-1) for complex 1
excited at 400 nm. The red line is the result of a single exponential fit of the data
yielding a time constant of 2.29 ± 0.7 ps

117

somewhat higher in energy relative to the ground state because the backbonding
interaction between the ruthenium and CN-Me-bpy will be weaker in the excited
state (where the ruthenium is formally Ru3+), which leads to a stronger CN bond
and higher frequency vibration in the excited state. This suggests that on the
picosecond timescale of this measurement, the excited state in both cases is
already localized on the CN-Me-bpy ligand and the interligand electron transfer
rate from the unsubstituted bpy to the CN-Me-bpy ligand is much less than a
picosecond.

0.2

Anisotropy

0.1
0.0
-0.1
-0.2
-0.3
-0.4

0

20

40
60
Time (ps)

80

100

Figure 4-10: Results of polarized absorption spectroscopy, showing no
anisotropic decay. This shows that the ILET from bpy to CN-Me-bpy is much
faster than 1 ps.

This idea of rapid ILET is also confirmed by looking at the polarized time
resolved absorption spectrum. Along with the magic angle spectra shown in
Figures 4-6 -4-8, spectra were also collected with the pump and probe beams
parallel and perpendicular to each other. The anisotropy can be calculated from

118

the parallel and perpendicular spectra using Equation 4-3, where r is the
r=

I ! "I!

4-3

I ! + 2 I!

anisotropy, I‖ is the parallel transient absorption signal, and I
perpendicular transient absorption signal.

is the

The time dependance of the

anisotropic signal for complex 1 with 400 nm excitation is shown in Figure 4-10.
The lack of any anisotropic signal shows that the excited state is completely
randomized on this time scale, again suggesting that the ILET rate is much less
than 1 ps.
A comparison of Figures 4-6 - 4-8 shows that the the spectra for the three
complexes are more or less identical. The fact that the excited state absorption
maximum is the same for all three complexes shows that the amount of excited
state delocalization is also the same. The absorption maximum is also the same
as what is observed in the nanosecond step-scan IR experiment. Therefore, in
all three complexes, the excited state must be localized on a single CN-Me-bpy
ligand and the increase in the peak intensity is due to vibrational relaxation on
the long lived excited state potential energy surface. Like the [Ru(bpy)2(CN-Mebpy)](PF6)2, the only kinetic feature seen in the other two complexes is the
growth of the excited state absorption maximum. For all three complexes, at
both excitation wavelengths, the growth could be fit with a single exponential to
yield time constants from ~1 to 8 ps, which is similar to the the vibrational
relaxation time measured by Browne, McGarvey and coworkers.7
4.3.3 Time-Resolved Electronic Absorption Spectroscopy
As a complement to the infrared data, time-resolved visible absorption
spectra were also collected on the three complexes, exciting at two different
wavelengths. Like the infrared data, the two excitation wavelengths were chosen

119

so that at the lower energy wavelength (490 nm) the initial excited state would be
localized on the CN-Me-bpy ligand and at the higher energy wavelength (425
nm) the initial excited state in complexes 1 and 2 would be localized on the
unsubstituted bipyridine ligand.

Wavelength (nm)
400

450

500

600

0.02
-0.5 ps
0.7 ps
1.7 ps
5.1 ps
9.5 ps
15.1 ps
19.5 ps
22.8 ps

0.01
0.00
-0.01
-0.02
-0.03
(a)

-0.04
0.02

0.05

0.01
0.00

0.00

-0.01

-0.05

-0.02

-0.10

-0.03
-0.04
30

(b)
28

26

24

22
3

20

18

-0.15

Change in Absorbance

Change in Absorbance

Change in Absorbance

350

16

-1

Energy (x10 cm )
Figure 4-11: Full spectra for compound 1 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the nanosecond
transient absorption spectrum (blue squares)

4.3.3.1 Low Energy Excitation
As a parallel to the time resolved infrared spectra, Figure 4-11a shows the full
spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 after 490 nm excitation, collected in

120

Wavelength (nm)
400

450

500

600

0.02

-0.5 ps
0.7 ps
1.7 ps
5.1 ps
9.5 ps
15.1 ps
19.5 ps
22.8 ps

0.00
-0.02
-0.04
-0.06
-0.08
(a)

-0.10
0.02

0.05

0.00

0.00

-0.02

-0.05

-0.04

-0.10

-0.06
(b)

-0.08
30

28

26

24

22
3

20

18

-0.15

Change in Absorbance

Change in Absorbance

Change in Absorbance

350

16

-1

Energy (x10 cm )
Figure 4-12: Full spectra for compound 2 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the nanosecond
transient absorption spectrum (green triangle). The scatter at 490 pump
wavelength was removed for a better comparison.

1 ps steps.

This spectrum shows a slight rise in the CN-Me-bpy radical

absorption over time, without any significant change in the peak position or
shape. This mirrors the behavior seen in the infrared spectra. By 15 ps the
spectrum matches the spectrum seen on the nanosecond timescale (Figure
4-11b). This shows that the thermalized long lived triplet state, which is present
on a nanosecond timescale, is formed within 15 ps.

121

Complexes 2 and 3

600

0.04
-0.5 ps
0.7 ps
1.7 ps
5.1ps
9.5 ps
15.1 ps
19.5 ps
22.8 ps

0.02
0.00
-0.02
-0.04
-0.06

(a)

0.04

0.04

0.02

0.02

0.00

0.00

-0.02

-0.02
-0.04

-0.04

-0.06

-0.06
30

(b)
28

26

24

22
3

20

18

-0.08

Change in Absorbance

Change in Absorbance

Change in Absorbance

350

Wavelength (nm)
400
450
500

16

-1

Energy (x10 cm )

Figure 4-13: Full spectra for compound 3 excited at 490 nm (a) probed in 1 ps
steps. (b) Overlay of the 15 ps spectrum (black line) with the nanosecond
transient absorption spectrum (red circle)

(Figures 4-12 and 4-13) show a similar rise in the CN-Me-bpy radical absorption
over time.
To get a more accurate estimate of the timescale of the peak increase, single
wavelength traces were measured for all three complexes (Figures 4-14 - 4-16).
For each complex, a single wavelength trace was collected at the CN-Me-bpy
radical absorption (370 nm), the isosbestic point between the radical absorption

122

Change in
Absorbance

0.004

0.002

0.000

(a)

Change in
Absorbance

0.000
-0.005
-0.010

(b)

Change in
Absorbance

Change in
Absorbance

-0.015
0.002
0.000
-0.002
-0.004

(c)

-0.006
0.010
0.005
0.000
-0.005

(d)
0

5

10

15

20

25

30

Time (ps)

Figure 4-14: Single wavelength kinetics for complex 1 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 400 nm (red), (c) 460 nm (green),
and (d) 540 nm (purple). The black line is the result of a double exponential fit,
yielding lifetimes of 0.42 ± 0.05 ps and 9.20 ± 2.5 ps.

123

Change in
Absorbance

0.004
0.002
0.000

Change in
Absorbance

(a)
0.010
0.005
0.000

Change in
Absorbance

-0.005

(b)

0.00
-0.01
-0.02

(c)

Change in
Absorbance

-0.03
0.010
0.005
0.000
-0.005

(d)
0

5

10

15

20

25

30

Time (ps)

Figure 4-15: Single wavelength kinetics for complex 2 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 410 nm (red), (c) 460 nm (green),
and (d) 540 nm (purple). Black line is the results of a double exponential fit,
yielding lifetimes of 0.39 ± 0.13 ps and 7.16 ± 2.1 ps.

124

Change in
Absorbance

0.010

0.005

Change in
Absorbance

0.000

(a)

0.006
0.004
0.002
0.000

(b)

Change in
Absorbance

-0.002
0.00
-0.01
-0.02

(c)

-0.03

Change in
Absorbance

0.006
0.004
0.002
0.000
-0.002

(d)
0

5

10

15

20

25

30

Time (ps)

Figure 4-16: Single wavelength kinetics for complex 3 after 490 nm excitation.
Probe wavelengths are (a) 370 nm (blue), (b) 410 nm (red), (c) 460 nm (green),
and (d) 540 nm (purple). Black lines are the results of double exponential fits,
yielding lifetimes of 0.15 ± 0.07 ps and 5.25 ± 1.5 ps for 410 nm and 0.10 ± .01
ps and 5.45 ± 0.87 ps for 540 nm.

125

Wavelength (nm)

Change in Absorbance

0.02

350

400

450

500

22

20

600

0.00
-0.02
-0.04
30

28

26

24

3

18

-300 fs
0 fs
33 fs
66 fs
100 fs
200 fs
300 fs
400 fs
600 fs
800 fs
1200 fs
1500 fs
1900 fs

16

-1

Energy (x10 cm )

Figure 4-17: Full spectra for compound 1 excited at 490 nm, probed in 33 fs
steps.

Change in Absorbance

350

Wavelength (nm)
400
450 500

600
-300 fs
0 fs
33 fs
66 fs
100 fs
200 fs
300 fs
400 fs
800 fs
1200 fs
1500 fs
1900 fs

0.02
0.00
-0.02
-0.04
-0.06
30

28

26

24

22
3

20

18

16

-1

Energy (x10 cm )

Figure 4-18: Full spectra for compound 2 excited at 490 nm, probed in 33 fs
steps.

126

and bleach features (390 or 410 nm), the MLCT bleach (460 nm), and the red
absorption (540 nm). The single wavelength trace collected at the isosbestic
point on all three complexes shows a fast increase (100-500 fs) followed by a
slower (5-10 ps) decay. This is in agreement with previous work in our group on
[Ru(dmb)3]2+ and [Ru(dpb)3]2+ (dmb = 4,4'-dimethyl-2,2'-bipyridine and dpb =
4,4'-diphenyl-2,2'-bipyridine).21 By comparison to those previous results and the
infrared results above, the slower component can be assigned as vibrational
relaxation within the lowest energy 3MLCT state.

This result confirms the

assumption that the subtle, small amplitude changes in the time resolved
electronic

absorption

spectrum

do

correspond

to

vibrational

relaxation

dynamics.12,21,22

Change in Absorbance

Wavelength (nm)
340 360 380 400
450 500 550 600
-300 fs
0 fs
33 fs
66 fs
100 fs
200 fs
300 fs
400 fs
600 fs
800 fs
1200 fs
1500 fs
1900 fs

0.02
0.00
-0.02
-0.04
-0.06
-0.08
30

28

26

24

22

Enregy x10

3

20

18

16

-1

cm )

Figure 4-19: Full spectra for compound 2 excited at 490 nm, probed in 33 fs
steps.

127

The origin of the faster component is more uncertain. In order to get a better
picture of the origin of the faster component full spectra were collected in 33 fs
steps. The shorter time spectra of complex 1, (Figure 4-17) shows some shifting
of the CN-Me-bpy radical absorption and a growing in of the MLCT bleach over
~200 fs. With the 490 nm excitation both the initial and long lived excited state
are localized on the CN-Me-bpy ligand. This means that interligand electron
transfer can be eliminated as the origin of the fast kinetic component. In other
ruthenium tris bipyridyl complexes, such as [Ru(bpy)3]2+, [Ru(dmb)3]2+ and
[Ru(dpb)3]2+ (dpb = 4,4'-diphenyl-2,2'-bipyridine a similar fast kinetic component

Wavelength (nm)

Change in
Absorbance

0.010

360

380

400

450

500

550

600
-1.1 ps
1.1 ps
2.2 ps
5.5 ps
10 ps
15.5 ps
20 ps
22.2 ps

0.000

-0.010

(a)
-0.020

Change in
Absorbance

0.01
0.00
-0.01
-0.02

(b)

-0.03
28

26

24

22
3

20
-1

18

16

-100 fs
0 fs
33 fs
66 fs
100 fs
200 fs
400 fs
600 fs
800 fs
1200 fs
1500 fs
1900 fs

Energy (x10 cm )

Figure 4-20: Full spectra for [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1) after 425 nm
excitation, probed in (a) 1 ps steps and (b) 33 fs steps

128

on the order of 100 fs has been observed.21,23,24

In the literature this fast

component has been assigned as the formation of the 3MLCT state. The slow
grow in of the bleach feature, however, may suggest that the observed kinetic
component is due to intramolecular vibrational redistribution (IVR).

Similar

behavior is seen for complexes 2 and 3 (Figures 4-18 and 4-19).

Wavelength (nm)

Change in
Absorbance

0.01

360

380

400

450

500

550

600
-1.1 ps
1.1 ps
2.2 ps
5.5 ps
10.0 ps
15.5 ps
20.0 ps
22.2 ps

0.00
-0.01
-0.02

(a)

Change in
Absorbance

0.02

0.00

-0.02
-0.04
28

(b)
26

24

22

20

18

16

-100 fs
0 fs
33 fs
66 fs
100 fs
200 fs
400 fs
600 fs
800 fs
1200 fs
1500 fs
1900 fs

Energy (x10 3 cm-1)
Figure 4-21: Full spectra for compound 2 with 425 nm excitation probed in (a) 1
ps intervals and (b) 33 fs intervals.

4.3.2.2 High Energy Excitation
To mirror the time-resolved infrared experiments, time-resolved electronic
absorption spectra were collected after excitation at higher energy (425 nm).

129

With the 425 nm excitation, the TD-DFT calculations show that the initial excited
state is localized on the unsubstituted bipyridine ligand(s) in complexes 1 and 2.
This opens up the possibility of interligand electron transfer. Figure 4-20 shows
the full spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 after 425 nm excitation. At this
excitation wavelength, the initial excited state is localized on the unsubstituted
bipyridine ligand(s). In the long time spectra, collected in 1 ps steps, there is an
increase in the ligand radical absorption over time without a significant change in
the energy or shape of the peak. This is the same behavior that was seen in the
spectra collected after 490 nm excitation. This is not surprising, considering the
time resolved infrared spectra showed no differences between 400nm excitation

Wavelength (nm)
360

380

400

450

500

550

600
-1.1 ps
1.1 ps
2.2 ps
5.5 ps
10.0 ps
15.5 ps
20.0 ps
22.2 ps

Change in
Absorbance

0.01
0.00
-0.01
-0.02

(a)

Change in
Absorbance

0.02
0.00
-0.02
-0.04
28

(b)
26

24

22
3

20
-1

18

16

-100 fs
0 fs
33 fs
66 fs
100 fs
200 fs
400 fs
600 fs
800 fs
1200 fs
1500 fs
1900 fs

Energy (x10 cm )

Figure 4-22: Full spectra for compound 3 with 425 nm excitation probed in (a) 1
ps intervals and (b) 33 fs intervals.

130

0.004

Change in
Absorbance

0.003
0.002
0.001
0.000

(a)

-0.001

Change in
Absorbance

0.000
-0.001
-0.002
-0.003

(b)

Change in
Absorbance

0.0010

0.0005

0.0000

(c)
-0.0005

0

5

10

15

20

25

30

Time (ps)

Figure 4-23: Single wavelength kinetics for complex 1 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c) 520 nm
(green). The black lines on the 460 nm and 520 nm traces are the results of
biexponential fits, yielding lifetimes of 0.12 ± .03 ps and 0.33 ± 0.13 ps for 460
nm and 0.14 ± .07 ps and 1.46 ± 2.3 ps for 540 nm.

131

Change in
Absorbance

0.004

0.002

0.000

(a)

Change in
Absorbance

-0.002

0.000
-0.005
-0.010

(b)
-0.015

Change in
Absorbance

0.004

0.002

0.000

(c)
0

5

10

15

20

25

30

Time (ps)

Figure 4-24: Single wavelength kinetics for complex 2 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c) 520 nm
(green). The black line on the 520 nm trace is the result of a single exponential
fit yielding a lifetime of 7.25 ± 1.3 ps.

132

Change in
Absorbance

0.008
0.006
0.004
0.002
0.000

(a)

-0.002

Change in
Absorbance

0.000

-0.005

-0.010

(b)

Change in
Absorbance

0.0010

0.0005

0.0000

(c)
0

5

10

Time (ps)

15

20

Figure 4-25: Single wavelength kinetics for complex 3 after 425 nm excitation.
Probe wavelength are (a) 370 nm (blue), (b) 460 nm (red), and (c) 540 nm
(green). The black line on the 540 nm trace is the result of a biexponential fit
yielding lifetimes of 0.15 ± 0.08 ps and 0.32 ± 0.27 ps.

133

and 490 nm excitation. In the shorter time spectra, however, the poor signal-tonoise ratio in the near UV region of the spectrum make it difficult to determine if
there is any shifting of the ligand radical absorption band, as was seen in the
spectra after 490 nm excitation. Similar results were seen for complexes 2 and 3
(Figures 4-21 and 4-22).
Again, to get a more accurate estimate of any kinetic processes, single
wavelength traces were collected for all three complexes (Figures 4-23 - 4-25).
For each complex, a single wavelength trace was collected at the CN-Me-bpy
radical absorption (370 nm), the MLCT bleach (460 nm), and the isosbestic point
between the MLCT bleach and the red absorption (520 or 540 nm).

At the

isosbestic points in complexes 1 and 2 there is a long component, ~ 1.5 ps and
~7 ps respectively, which most likely corresponds to vibrational relaxation in the
lowest energy 3MLCT state. All three complexes also show shorter dynamics,
similar to those seen after 490 nm excitation.

These possibly represent the

formation of the 3MLCT state. In the long time full spectra, it is clear that the long
lived 3MLCT is already established and the observed changes are due to the
vibrational relaxation within that state. Unfortunately, the poor signal-to-noise of
the shorter time full spectra make it difficult to determine whether the nature of
the excited state is changing (through interligand electron transfer) or otherwise
give clues as to what processes could be giving rise to the shorter kinetic
components.
4.4 Future Works
4.4.1 Solvent Effects
Preliminary transient absorption spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 at
longer times revealed an additional kinetic component (180 ± 30 ps) to the
decay, which can been seen in Figure 4-26. The ground state recovery time in

134

Change in Absorbance

0.000
-0.005
-0.010
-0.015
-0.020
-0.025
0

400
Time (ps)

800

Figure 4-26:
Single wavelength trace for [Ru(bpy)2(CN-Me-bpy)](PF6)2
collected at 450 nm, following excitation at 490 nm. The red line is the result of a
single exponential fit, resulting in a lifetime of 180 ± 30 ps.

complex 1 from Chapter 3 is around 1.5 μs, so this new component is far too
short to be associated with ground state recovery. The ultrafast infrared and
visible absorption spectroscopy show that the thermalized 3MLCT state is formed
within 15-20 ps (vide supra) and therefore the ~200 ps process must be
occurring from the thermalized long lived 3MLCT state. In related ruthenium(II)
complexes of 4,4'-dicarboxy-2,2'-bipyridne (dcbpy) the carboxylate groups are
mostly deprotonated in the ground state (pKa ≈ 2), but in the long lived 3MLCT
state they become more basic (pKa* ≈ 4)25,26. One result of this shift in pKa is
that the excited state on the complex can be protonated by the solvent (or water
in the solvent). Work in our group has shown that this excited state protonation
reaction most likely occurs with a time constant of hundreds of picoseconds.27
Scandola and coworkers performed similar experiments with Ru(bpy)2CN2 and
showed that the cyanide groups have similar acid-base equilibria.28

135

Like the carboxylate groups in the ruthenium(II) dcbpy complexes, the
cyanide groups in complex 1 also have an increased electron density in the
excited state, potentially making them more basic. It may be possible that the
observed 180 ps decay component is due to an interaction between the excited
state and the solvent (or residual water in the solvent). The solvatochromism of
the lowest energy charge transfer feature, shown in Figure 4-27, is another
indication that the excited state has a significant interaction with the solvent.
Solvatochromism and what it implies about solvent-molecule interactions will be
discussed in Chapter 5. A variable solvent time-resolved absorption study using
non-polar, dipolar aprotic, and protic solvents is a way to either support or
disprove this theory.

Wavelength (nm)
400

1.2

450

500

600
water
pyridine
C2H4Cl2
CH3NO2
DMSO
DMF
CH3CN
Acetone
EtOH
THF
CH2Cl2
MeOH
DME

Normalized Absorbance

1.0
0.8
0.6

1.0
0.9

0.4
0.8
0.2
0.0
28

0.7

20.0
3
-1
Energy (x10 cm )

26

21.0

24

22
3

20
-1

18

16

Energy (x10 cm )

Figure 4-27: Electronic absorption spectra of [Ru(bpy)2(CN-Me-bpy)](PF6)2 (1)
in different solvents, showing the solvatochromatic behavior of the lowest energy
MLCT absorption band. The inset shows an expanded view of the lowest energy
absorption maxima.
136

4.4.2 ILET and Localization
As mentioned in Chapter 1, one of the major unresolved questions in
ruthenium (II) photophysics is the localized or delocalized nature of the excited
state. It is well established that the thermalized lowest energy 3MLCT state is
localized on a single ligand, even in homoleptic complexes.29-32. A number of
experiments also show that the initially excited Franck-Condon state is most
likely also localized on a single ligand.33-35 The question remains, does the
excited state stay localized on the same bipyridine ligand as it moves from the
Franck-Condon state to the thermalized 3MLCT state?

Ultrafast polarized

absorption spectra collected by Hammarström and coworkers show that the
excited state in [Ru(bpy)3]2+ is randomized in < 300 fs, meaning it has no
memory of where the initial excited state was localized. It is unclear, however,
whether this randomization is due to the formation of a delocalized state or very
rapid interligand electron transfer (ILET).
The cyano-substituted ruthenium(II) bipyridine complexes offer a unique
opportunity to resolve this question. By changing the ancillary bipyridine ligands,
as shown in Figure 4-18, it is possible to design a series of compounds where
the driving force for ILET varies across the series. In these examples, the energy
difference between the two different bipyridine ligands is estimated by comparing
the first reduction potentials for the respective homoleptic complexes.

For

example, the energy difference between CF3-bpy (CF3-bpy = 4,4'-bis
(trifluoromethyl)-2,2'-bipyridine) and CN-Me-bpy in [Ru(CF3-bpy)2(CN-Me-bpy)]2+
is estimated by comparing [Ru(CF3-bpy)3]2+ (Ered = 0.77 V vs SCE)36 and
[Ru(CN-Me-bpy)3]2+ (Ered = 0.85 V vs SCE). This predicts that the CF3-bpy
ligand is ~ 80 mV lower in energy than the CN-Me-bpy ligand so ILET from CF3bpy to CN-Me-bpy would be a slightly uphill process.

137

CN

CF3
F3C

N

N

Ru

N

N

F3C

NC

CN
N

N

N

N
NC

CN

[Ru(CF3-bpy)2(CN-Me-bpy)]2+

N

Ru
N

N

N
N
CN

[Ru(CN-Me-bpy)3]2+

+80 mV

N

Ru

CN

CN

CF3

N

N

0 mV

CN
N

N

N

N
CN

N
Ru
N

CN
N
N
CN

[Ru(bpy)2(CN-Me-bpy)]2+

[Ru(tmb)2(CN-Me-bpy)]2+

-480 mV

-780 mV

Figure 4-28: A series of [Ru(bpy')2(CN-Me-bpy)](PF6)2 complexes with varying
energy differences between bpy' and CN-Me-bpy. Listed energy differences are
estimated by the difference in the first reduction potential for the corresponding
homoleptic complexes, i.e. Ered{[Ru(CN-Me-bpy)3]2+} - Ered{[Ru(bpy')3]2+}.

Polarized absorption measurements, like those done by Hammarström and
coworkers37, will be able to measure the timescale of excited state randomization
but cannot distinguish between delocalization and rapid ILET.

The cyanide

substituents in this series of complexes provide a way by which the two can be
distinguished. The excited state stretching frequency of the cyanide group will be
very different in a delocalized excited state compared to a localized excited state
undergoing rapid ILET. Time resolved infrared measurements on [Ru(CN-Me-

138

bpy)3]2+ on a nanosecond timescale show a ~40 cm-1 shift between the ground
state and the excited state. As mentioned above, it is well established that the
excited state is localized on a single ligand on a nanosecond timescale, so that
40 cm-1 shift represents a localized excited state. In the case of a delocalized
excited state, the excited electron is delocalized over all three CN-Me-bpy
ligands so the effect on the cyanide stretching frequency should be about a third
of what it is in the localized state (~13 cm-1). The difference between a 13 cm-1
shift and a 40 cm-1 shift should be easily detectable by time-resolved infrared
spectroscopy or femtosecond stimulated Raman spectroscopy.

4.5 Concluding Comments
Both the time-resolved infrared and visible absorption experiments show
evidence that vibrational relaxation occurs on 1-10 ps timescale in these
complexes. Vibrational relaxation on this timescale has been observed in other
ruthenium bipyridine systems, most notably in the work of Browne, McGarvey
and coworkers on the time-resolved resonance Raman spectroscopy of
[Ru(bpy)3]2+. This is further evidence that the conclusions drawn by Chegui and
coworkers, that the thermalized long lived 3MLCT state in [Ru(bpy)3]2+ is fully
formed in 15 fs, is incorrect.
The time resolve infrared experiments also show that interligand electron
transfer and/or intermolecular vibrational energy redistribution both occur much
faster than the picosecond resolution of the experiment. Localizing the initial
excited state on the CN-Me-bpy or unsubstituted bpy lead to identical spectra,
meaning in both cases only the long live excited state, localized on a single CNMe-bpy ligand was being probed. The time-resolved electronic absorption data
also shows a ~1-10 ps vibrational relaxation time. The faster kinetic components
of the electronic absorption spectra are most likely related to the formation of the

139

long lived 3MLCT state, but without further experiments, it is unclear exactly
which process (intersystem crossing, interligand electron transfer, intermolecular
vibrational energy redistribution, etc.) corresponds to the observed rates.

140

REFERENCES

141

4.6 References
(1) Peteanu, L. A.; Schoenlein, R. W.; Wang, Q.; Mathies, R. A.; Shank, C. V.
P. Natl. Acad. Sci. USA. 1993, 90, 11762.
(2)

Kuciauskas, D.; Monat, J. E.; Villahermosa, R.; Gray, H. B.; Lewis, N. S.;
McCusker, J. K. J. Phys. Chem. B. 2002, 106, 9347–9358.

(3)

McCamant, D. W.; Kukura, P.; Mathies, R. A. J. Phys. Chem. B. 2005, 109,
10449-10457.

(4)

Forster, L. S. Coord. Chem. Rev. 2006, 250, 2023-2033.

(5)

Frontiera, R. R.; Dasgupta, J.; Mathies, R. A. J. Am. Chem. Soc. 2009, 131,
15630-15632.

(6)

Cannizzo, A.; van Mourik, F.; Gawelda, W.; Zgrablic, G.; Bressler, C.;
Chergui, M. Angew. Chem. Int. Ed. Engl. 2006, 45, 3174-3176.

(7)

Henry, W.; Coates, C. G.; Brady, C.; Ronayne, K. L.; Matousek, P.; Towrie,
M.; Botchway, S. W.; Parker, A. W.; Vos, J. G.; Browne, W. R.; McGarvey,
J. J. J. Phys. Chem. A. 2008, 112, 4537-4544.

(8)

King, J. C.; Zhang, J. Z.; Schwartz, B. J.; Harris, C. B. J. Chem. Phys. 1993,
99, 7595.

(9)

Liard, D. J.; Busby, M.; Matousek, P.; Towrie, M.; Vlček, A. A. J. Phys.
Chem. A. 2004, 108, 2363-2369.

(10) Záliš, S.; Busby, M.; Kotrba, T.; Matousek, P.; Towrie, M.; Vlček, A. A.
Inorg. Chem. 2004, 43, 1723–1734.
(11) Shaw, G. B.; Styers-Barnett, D. J.; Gannon, E. Z.; Granger, J. C.;
Papanikolas, J. M. J. Phys. Chem. A. 2004, 108, 4998-5006.
(12) Juban, E. A.; McCusker, J. K. J. Am. Chem. Soc. 2005, 127, 6857–6865.
(13) Maçôas, E. M.; Kananavicius, R.; Myllyperkiö, P.; Pettersson, M.; Kunttu, H.
J. Phys. Chem. A. 2007, 111, 2054–2061.
(14) Maçôas, E. M.; Kananavicius, R.; Myllyperkiö, P.; Pettersson, M.; Kunttu, H.
J. Am. Chem. Soc. 2007, 129, 8934–8935.
(15) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Montgomery, J., J. A.; Vreven, T.; Kudin, K. N.; Burant,
J. C.; Millam, J. M. ; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.;

142

Cossi, M. ; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian,
H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain,
M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski,
J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin,
R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, revision D.01; Gaussian, Inc.,
Wallingford CT, 2004
(16) Monat, J. E.; Rodriguez, J. H.; McCusker, J. K. J. Phys. Chem. A. 2002,
106, 7399–7406.
(17) Fodor, L.; Lendvay, G.; Horváth, A. J. Phys. Chem. A. 2007, 111, 12891–
12900.
(18) Gabrielsson, A.; Towrie, M.; Zalis, S.; Vlcek, A. J. Inorg. Chem. 2008, 47,
4236-4242.
(19) Vlček, A. A.; Farrell, I. R.; Liard, D. J.; Matousek, P.; Towrie, M.; Parker, A.
W.; Grills, D. C.; George, M. W. J. Chem. Soc. Dalton Trans. 2002, 701–
712.
(20) Schrauben, J. N. Ph.D. Dissertation, Michigan State University, East
Lansing, MI, 2010
(21) Damrauer, N. H.; McCusker, J. K. J. Phys. Chem. A. 1999, 103, 8440–
8446.
(22) Schrauben, J. N.; Dillman, K. L.; Beck, W. F.; McCusker, J. K. Chem. Sci.
2010, 1, 405.
(23) Damrauer, N. H.; Cerullo, G.; Yeh, A.; Boussie, T. R.; Shank, C. V.;
McCusker, J. K. Science. 1997, 275, 54–57.
(24) Yoon, S.; Kukura, P.; Stuart, C. M.; Mathies, R. A. Molecular Physics. 2006,
104, 1275-1282.
(25) Nazeeruddin, M. K.; Kalyanasundaram, K. Inorg. Chem. 1989, 28, 4251–
4259.

143

(26) Nazeeruddin, M. K.; Zakeeruddin, S. M.; Humphry-Baker, R.; Jirousek, M.;
Liska, P.; Vlachopoulos, N.; Shklover, V.; H., F. C.; Grätzel, M. Inorg.
Chem. 1999, 38, 6298-6305.
(27) Smeigh, A. L. Unpublished Work.
(28) Davili, J.; Bignozzi, C. A.; Scandola, F. J. Phys. Chem. 1989, 93, 1373–
1380.
(29) Bradley, P. G.; Kress, N.; Hornberger, B. A.; Dallinger, R. F.; Woodruff, W.
H. J. Am. Chem. Soc. 1981, 103, 7441-7446.
(30) Morris, D. E.; Hanck, K. W.; DeArmond, M. K. J. Am. Chem. Soc. 1983,
105, 3032-3038.
(31) Mabrouk, P. A.; Wrighton, M. S. Inorg. Chem. 1986, 25, 526–531.
(32) Omberg, K. M.; Schoonover, J. R.; Treadway, J. A.; Leasure, R. M.; Dyer,
R. B.; Meyer, T. J. J. Am. Chem. Soc. 1997, 119, 7013–7018.
(33) Kober, E. M.; Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1984, 23,
2098-2104.
(34) Oh, D. H.; Boxer, S. G. J. Am. Chem. Soc. 1989, 111, 1130-1131.
(35) Webb, M. A.; Knorr, F. J.; McHale, J. L. J. Raman Spectrosc. 2001, 32,
481-485.
(36) Furue, M.; Maruyama, K.; Oguni, T.; Naiki, M.; Kamachi, M. Inorg. Chem.
1992, 31, 3792-3795.
(37) Wallin, S.; Davidsson, J.; Modin, J.; Hammarström, L. J. Phys. Chem. A.
2005, 109, 4697-4704.

144

Chapter 5: Synthesis and Spectroscopic Characterization of Cyano-Substituted Terpyridyl
Complexes of Ruthenium(II)

5.1 Introduction

Ruthenium tris bipyridine complexes have many properties, which have been
discussed in Chapters 1 and 3, that make them ideally suited for a wide array of
applications. Briefly, their lowest energy excited state is a 3MLCT state that lives
100's of nanoseconds to microseconds. The charge transfer nature of the long
lived excited state means that the excited state can reversibly act as either an
oxidant or a reductant and the long lifetime of the excited state allows for
bimolecular excited state reactions to occur. The relatively high quantum yields
of this class of compound also makes them potentially useful as fluorescent dyes
or sensors. The properties of ruthenium(II) bipyridine complexes, particularly
[Ru(bpy)3]2+, have been widely studied on all timescales and the excited state
dynamics in these compounds is fairly well (but not completely) understood.
The geometry is one aspect of tris bipyridine complexes that is not ideal for
many applications is the geometry. Many different applications, such as light
harvesting systems or molecular arrays, are best served by well ordered linear
geometries. With a tris bipyridine complex, there are many different ways to
arrange two substituent around the metal center (a few of which can be seen in
Figure 5-1), and any reaction will yield a mixture of these different isomers. In
order to get a linear arrangement of substituents with only one isomer, it is
necessary to move to a bis terpyridine framework.
While the geometry of bis(terpyridyl) ruthenium(II) complexes makes them
perfectly suited for many applications, their photophysical properties are less
than ideal.

The bis(terpyridyl)ruthenium(II) systems have electrochemical

properties that are similar to related bipyridine systems, so they can still act as

145

A

A

N
N

N
N

N

N

N

Ru
N
N

A
N
N

N

N

N

Ru

Ru

N

B

N
N
N

B

B
A

N
N

N
Ru

N

N
N

B
Figure 5-1: Examples of different isomers formed with two substituents (A and B)
in a ruthenium tris(bipyridine) complex (top) and a ruthenium bis(terpyridine)
complex (bottom)

both oxidants and reductants when photoexcited. The longest lived excited state
is also charge transfer in nature. The problem is the terpyridine with the lifetime
of the longest lived excited state; [Ru(tpy)2]2+ has a lifetime of only 250 ps,
compared to 950 ns for [Ru(bpy)3]2+. The main reason for the difference in
lifetimes is the difference in ligand field strength between bipyridine and
terpyridine. The geometry of the terpyridyl ligand distorts the ruthenium center
from an ideal octahedral coordination environment and weakens the ligand field
strength of the terpyridine ligand. The weaker ligand field strength means that
the ligand field states are lower in energy in terpyridine complexes than they are
146

in bipyridine. In [Ru(tpy)2]2+ and other terpyridine complexes the ligand field
states are low enough in energy that they are thermally accessible form the long
lived excited state. In general, ligand field states couple to the ground state very
efficiently through non-radiative decay, so thermal population of the ligand field
states in ruthenium(II) terpyridine complexes leads to very rapid, non-radiative
deactivation of the excited state.
Different approaches have been taken in attempt to lengthen the lifetime of
bis(terpyridyl) ruthenium(II) complexes by increasing the energy between the
charge transfer and ligand field states.

Increasing the conjugation in the

terpyridyl π system both lowers the energy of the charge transfer state, which
increases the energy barrier between the charge transfer and ligand field states
and makes the charge transfer state more nested, which lowers the rate of nonradiative decay directly from the charge transfer state to the ground state. The
increased conjugation can also increase the transition dipole (μ), which increases
kr according to radiative decay theory, therefore increasing the extinction
coefficient and the quantum yield.

Harriman and coworkers used a

phenylethynyl substituent to extend the lifetime to 44 ± 3 ns (compared to 250 ps
for [Ru(tpy)2]2+) and increased the room temperature quantum yield to 3.6 x 10-4
(compared to < 5 x10-6 for [Ru(tpy)2]2+).1 Increasing the ligand field strength of
the ligands bound to the ruthenium center will increase the energy of the ligand
field states. Scandola and coworkers replaced one terpyridine ligand with strong
field cyanide ligands in order to increase the energy separation between the
charge transfer and ligand field states. The [Ru(tpy)(CN)3]1- complex exhibited
extended lifetimes but the results were highly solvent dependent, ranging from <1
ns to 49 ns in different solvents.2

The most dramatic improvements in excited

state lifetime and quantum yield come from modifying the terpyridine ligand to
increase its flexibility and make the coordination environment around the

147

ruthenium center more octahedral.3 Johansson, Hammarström, and coworkers
replaced terpyridine with 2,6-di(quinolin-8-yl)pyridine (Figure 5-2a).4 This change
increased the bite angle of the ligand to 178° (from 158° for terpyridine) and
resulted in a complex with a 3 μs lifetime and a quantum yield of 0.02, which is
comparable to ruthenium tris bipyridine complexes.

Schramm et. al. used a

ketone bridged tripyridine ligand to form a perfectly octahedral coordination
environment around the ruthenium center (i.e. bond angles of 180° and 90°).5
This bis(tripyridine)ruthenium(II) complex exhibited a 3.5 μs lifetime and a
quantum yield of 30% (Figure 5-2b). This is actually an improvement over many
tris(bipyridine)ruthenium(II) complexes. For example [Ru(bpy)3]2+ has a 950 ns
lifetime and a 9.5% quantum yield.

2+

(a)

2+

(b)
O

O
N

N

N

N
Ru

N
N

N

N

O

N

Ru

N

N

N
O

Figure 5-2: Examples of ruthenium(II) complexes with tridentate ligands having
a near octahedral coordination geometry.
(a) [Ru(dqp)2]2+ (dqp = 2,6di(quinolin-8-yl)pyridine)4 (b) [Ru(tripy)2]2+ (tripy =1,1'-(2,6-pyridinediyl)bis[1-(2pyridinyl)methanone])5

This chapter investigates the ground state and long lived excited states of a
series of cyano-substituted terpyridine complexes (Figure 5-3).

These

complexes served as the synthetic inspiration for the bipyridine complexes
discussed in Chapters 3 and 4. Compared to tris bipyridine complexes, there

148

has been relatively little work done to investigate the excited state dynamics in
bis terpyridine complexes. The work in this chapter lays the groundwork for a
wider study into the excited state dynamics of bis(terpyridyl)ruthenium(II)
complexes.

CN

CN

N

N
N

N

N

N
N

N

N

Ru

Ru

N
N

N

CN
[Ru(tpy)(CN-tpy)]2+
(1)

[Ru(CN-tpy)2]2+
(2)

Figure 5-3:
Structures of the cyano-substituted terpyridine ruthenium(II)
complexes discussed in this chapter.

5.2 Experimental
5.2.1 General
All chemicals and solvents were obtained from Fisher or Aldrich Chemical Co.
and

used

without

further

purification

unless

otherwise

stated.

Tris(benzylidineacetone)dipalladium(0) (Pd2(dba)3), diphenylphosphinoferrocene
(dppf), and 4'-chloro-2,2':6'2"-terpyridine were purchased from Strem Chemicals,
Inc.

NMR spectra were collected on the Varian UnityPlus-500 (500 MHz)

spectrometer. Ground state infrared spectra (4000-400 cm-1) were measured as
KBr pellets using a Mattson Galaxy series 3000 FT-IR spectrophotometer. Mass

149

spectra were obtained through the Michigan State University Mass Spectrometry
Facility. Elemental analyses were obtained through the analytical facilities at
Michigan State University.

Solvent included in calculated elemental analysis

percentages was included to give the best fit based on solvents identified by
NMR and IR.

5.2.2 Synthesis of [Ru(tpy)(CN-tpy)](PF6)2
Trichloro(terpyridine)ruthenium(III)
Ru(tpy)Cl3
Ru(tpy)Cl3 was synthesized as reported by Constable et. al.6 Briefly, 1.07 g
of RuCl3•nH2O and 0.96 g of 2,2':6',2"-terpyridine were refluxed in 200 mL of
ethanol for one hour. The solution was cooled to room temperature and a brown
solid was collected by vacuum filtration; rinsed with ethanol, water, and ether;
and dried further under vacuum.

The insolubility of the product prevented

characterization by ESI/MS, NMR, or electronic absorption spectroscopy. Yield:
1.37 g

(4'-chloro-terpyridine)terpyridineruthenium(II) hexafluorophosphate
[Ru(tpy)(Cl-tpy)](PF6)2
[Ru(tpy)(Cl-tpy)](PF6)2 was synthesized as reported by Constable et. al.6
Briefly, 0.05 g of Ru(tpy)Cl3 and 0.03 g of 4'-chloro-terpyridine (Cl-tpy) were
suspended in 20.0 mL of methanol. Five drops of N-ethylmorpholine were added
and the solution was heated to reflux for one hour. After an hour the soultion
was cooled to room temperature and filtered through celite to remove any
unreacted Ru(tpy)Cl3.

Excess aqueous NH4PF6 was added to the filtrate to

precipitate a red powder. The red solid was collected be vacuum filtration; rinsed
once with cold methanol, three times with water and three times with ether; and

150

dried under vacuum.

Product was recrystalized once by acetonitrile/ether

diffusion. Yield: 0.08 g (75.5%) 1H NMR (500 MHz, (CD3)2CO): δ 9.20 (s, 2H),
9.08 (d, J=8.2 Hz, 2H), 8.90 (d, J=7.6 Hz, 2H), 8.82 (d, J=8.1 Hz, 2H), 8.60 (t,
J=8.2 Hz, 1H), 8.09 (m, 4H), 7.80 (d, J=6.3 Hz, 2H), 7.76 (d, J=6.3 Hz, 2H), 7.37
(t, J=6.6 Hz, 2H), 7.32 (t, J=5.9 Hz, 2H).

(4'-cyano-terpyridine)terpyridineruthenium(II) hexafluorophosphate
[Ru(tpy)(CN-tpy)](PF6)2
[Ru(tpy)(CN-tpy)](PF6)2 was synthesized as reported by Hanan, Campagna,
and coworkers.7 Brielfy, 0.025 g of [Ru(tpy)(Cl-tpy)](PF6)2, 0.002 g of Zn(CN)2,
0.001 g of Pd2(dba)3, 0.001 g of dppf, and a trace amount of zinc dust were
combined with 5.0 mL of dimethylacetamide (DMA) in a inert atmosphere
glovebox. The solution was transfered from the glovebox to a Schlenk line and
heated to 120°C for six hours. After six hours, the warm solution was filtered
through celite and the filtrate was evaporated to dryness. The red-brown residue
was

purified

by

column

chromatography

using

silica

gel

and

7:1

CH3CN:saturated aqueous KNO3. The purified product was recrystalized once
by acetonitrile and ether diffusion. The purity of the product was confirmed by
NMR and elemental analysis. Anal. Calcd (Found) for C31H21F12N7P2Ru•0.25
CH3CN: C, 42.38 (42.51); H, 2.46 (2.10); N, 11.37 (11.70). 1H NMR (500 mHz,
CD3CN): δ 9.07 (s, 2H), 8.79 (d, J=8.2 Hz, 2H), 8.54 (d, J=8.0 Hz, 2H), 8.51 (d,
J=8.7 Hz, 2H), 8.48 (t, J=8.1 Hz, 1H), 7.98 (t, j=7.9 Hz, 2H), 7.92 (t, J=8.0 Hz,
2H), 7.41 (d, J=5.1 Hz, 2H), 7.29 (t, J=5.5 Hz, 2H), 7.24 (t, J=6.6 Hz, 2H), 7.14 (t,
J=6.1 Hz, 2H).

FT-IR (selected frequencies in KBr pellet, cm-1) 3116(m),

2240(m), 1602(s), 1449(s), 835(vs), 558(vs).

151

5.2.3 Synthesis of [Ru(CN-tpy)2](PF6)2
Trichloro(4'-chloro-terpyridine)ruthenium(III)
Ru(Cl-tpy)Cl3
Ru(tpy)Cl3 was synthesized as reported by Constable et. al.6 Briefly, 0.52 g
of RuCl3•nH2O and 0.55 g of 4'-chloro-2,2':6',2"-terpyridine were refluxed in 100
mL of ethanol for one hour. The solution was cooled to room temperature and a
brown solid was collected by vacuum filtration; rinsed with ethanol, water, and
ether; and dried further under vacuum. The insolubility of the product prevented
characterization by ESI/MS, NMR, or electronic absorption spectroscopy. Yield:
0.78 g

Bis(4'-chloro-terpyridine)ruthenium(II) hexafluorophosphate
[Ru(Cl-tpy)2](PF6)2
[Ru(Cl-tpy2)](PF6)2 was synthesized as reported by Constable et. al.6 Briefly,
0.175 g of Ru(Cl-tpy)Cl3 and 0.099 g of 4'-chloro-terpyridine (Cl-tpy) were
suspended in 60.0 mL of methanol. Fifteen drops of N-ethylmorpholine were
added and the solution was heated to reflux for one hour. After an hour the
soultion was cooled to room temperature and filtered through celite to remove
any unreacted Ru(Cl-tpy)Cl3. Excess aqueous NH4PF6 was added to the filtrate
to precipitate a red powder. The red solid was collected by vacuum filtration;
rinsed once with cold methanol; three times with water and three times with
ether; and dried under vacuum. Product was recrystalized once by acetonitrile/
ether diffusion. Yield: 0.13 g (38.2%) 1H NMR (500 MHz, CD3CN): δ 9.10 (s,
2H), 8.80 (d, J=8.0 Hz, 2H), 8.00 (td, J=7.9, 1.5 Hz, 2H), 7.74 (d, J=5.6 Hz, 2H),
7.26 (dt, J=4.8, 1.3 Hz, 1H).

152

Bis(4'-cyano-terpyridine)ruthenium(II) hexafluorophosphate
[Ru(CN-tpy)2](PF6)2
[Ru(CN-tpy)2](PF6)2 was synthesized as reported by Hanan, Campagna, and
coworkers.7 Briefly, 0.108 g of [Ru(Cl-tpy)](PF6)2, 0.0162 g of Zn(CN)2, 0.012 g
of Pd2(dba)3, 0.013 g of dppf, and a trace amount of zinc dust were combined
with 25.0 mL of dimethylacetamide (DMA) in a inert atmosphere glovebox. The
solution was transfered from the glovebox to a Schlenk line and heated to 120°C
for overnight. The warm solution was filtered through celite and the filtrate was
evaporated to dryness.

The red-brown residue was purified by column

chromatography using silica gel and 7:1 CH3CN:saturated aqueous KNO3. The
purified product was recrystalized once by acetonitrile and ether diffusion. After
recrystallization, the solid was isolated as mixed salt.

Elemental analysis

suggests ~1.5 NO3- and 0.5 PF6- as the average counterion composition. The
purity of the product was also confirmed by NMR and mass spectroscopy. Anal.
Calcd (Found) for C32H20N8Ru•1.5 NO3 0.5 PF6: C, 49.08 (49.19); H, 2.57
(2.65); N, 16.99 (16.72). 1H NMR (500 MHz, CD3CN): δ 9.08 (s, 2H), 8.52 (d,
J=8.1 Hz, 2H), 7.99 (td, J=7.9, 1.6 Hz, 2H), 7.35 (d, J=5.6 Hz, 2H), 7.22 (td,
J=5.6, 1.3 Hz, 2H). MS [ESI (CH3CN), m/z (rel. int.)]: 309.1 (100) [M-2PF6]2+.
MS [HR-ESI (CH3CN)] m/z

309.0430 [M-2PF6]2+, calcd. (C32H20N8Ru)

309.0427. FT-IR (selected frequencies in KBr pellet, cm-1) 3121(w), 2240(m),
1625(w), 1603(s), 1426(s), 1353(s), 831(vs), 558(vs).

5.2.4 Physical Measurements
Electrochemistry
Electrochemical measurements were carried out in a Ar-filled drybox
(Vacuum Atmospheres) using a CHI 630B electrochemical analyzer. A standard
three-electrode arrangement was used, consisting of a Pt working electrode, a

153

graphite counter electrode, and a Ag/AgCl reference electrode (Cypress
Systems). Measurements were carried out in spectrophotometric grade CH3CN,
which was freeze-pump-thaw degassed before use, and using 0.1 M
tetrabutylammonium

hexafluorophosphate

(TBAPF6)

as

the

supporting

electrolyte. Data were acquired by cyclic voltammetry (CV) and differential pulse
voltammetry (DPV); the scan rate for the CV measurements was 50 mV/s and
the scan rate and pulse width for the DPV measurements were 20 mV/s and 50
mV, respectively.
comparable.

Values for E1/2 obtained by the two techniques were

All oxidation and reduction waves were reversible over several

consecutive scans.

Potentials are listed versus the ferrocene/ferrocenium

couple, which was used as an internal standard, and quoted as E1/2 values as
calculated from the DPV peak potentials.8

Electronic Steady-State and Time-Resolved Spectroscopies
All spectra were collected in spectrophotometric grade CH3CN unless
otherwise noted. For steady state emission and time-resolved measurements,
the solvent was freeze-pump-thaw degassed before use. Electronic absorption
spectra for all compounds were acquired using a Cary 50 spectrophotometer.
Steady-state emission spectra were acquired using a Spex Fluoromax
fluorimeter and corrected for instrumental response using a NIST standard of
spectral irradiance (Optronic Laboratories, Inc., OL220 M tungsten quartz lamp).
Samples were prepared in an inert atmosphere glovebox in 1 cm quartz cuvettes
and measured under optically dilute conditions (o.d. 0.1-0.2). The resulting
emission spectra were fit with an asymmetric double sigmoidal function using
IGOR pro.

This function has no mathematical significance but it is able to

accurately reproduce the shape of the entire emission curve and thereby capture
the small area (< 10%) that lies outside of the detector range thereby providing

154

for a more accurate estimate of the integrated spectrum.

Relative radiative

quantum yields (Φr) were determined using [Ru(bpy)3](PF6)2 or [Os(bpy)3](PF6)2
as a standard (Ru Φstd = 0.095 in degassed CH3CN9, Os Φstd = 0.0077 in
degassed CH3CN10,11).

Quantum yields were calculated as described in

Chapter 3, using Equation 5-1,where Φunk is the relative radiative quantum yield
!unk = !std

I unk
Aunk

A std
I std

5-1

"unk 2
"std

of the sample, Iunk and Istd are the integrated areas of the corrected emission
spectra of the sample and standard, respectively, Aunk and Astd are the
absorbances of the sample and the standard at the excitation wavelength, and
ηunk and ηstd are the indexes of refraction of the respective solvents (taken to be
equal to the neat solvents in both cases). Low-temperature emission spectra
were collected using a Janis SVT-100 optical cryostat as described previously.12
Measurements were taken at 80 K in a 9:2 mixture of butyronitrile and
propionitrile, both of which were freeze-pump-thaw degassed before use.
Estimates of the zero point energy gap (E0), Huang-Rhys factor (SM), energy of
the average vibrational mode coupling the ground and excited states ( ωM), and
spectral bandwidth (Δν̅0,1/2) were determined by a single mode fit of the steadystate emission spectra to Equation 5-2 as described by Claude and Meyer.13
The correction of Parker and Rees was applied to all spectra when converting
from wavelength to energy units.14
5

I (¯ =
!)
!M=0

E0 " ! M #M
E0

exp " 4(ln 2)

3

!M !

! " E 0 + ! M #M
¯
$ ! 0 ,1/ 2
¯

155

5-2

!
S MM
2

Nanosecond time-resolved emission and transient absorption experiments
were carried out using a Nd:YAG laser spectrometer that has been described
previously.15,16 Time-resolved emission was collected on the same samples
used to acquire the room-temperature steady-state emission spectra (vide
supra). Samples for time-resolved absorption measurements were prepared with
an absorbance in the range of 0.3-0.5 at the excitation wavelength, 510 nm, and
sealed under an Ar atmosphere in 1 cm quartz cuvettes. Data corresponds to a
15 shot average (0.2 Hz) of the signal and baseline as well as background
sample emission with 1-3 mJ of power at the sample. The baseline and emission
were subsequently subtracted from the signal and the data analyzed using a
program of local origin. All data were checked for linearity with respect to pump
power. In addition, absorption spectra were measured before and after all timeresolved absorption experiments to ensure the integrity of the sample. Laser
power was periodically monitored to ensure constant pump power over the
course of the experiment. Data acquired at each probe wavelength was fit to a
single exponential kinetic model; the amplitudes for each of these fits were
plotted to produce the differential absorption spectra reported herein.

5.3 Results and Discussion
I have synthesized and characterized the cyano-substituted terpyridine
complexes that served as synthetic inspiration for the cyano-substituted
bipyridine complexes discussed in the previous chapters. The properties of the
long lived 3MLCT state in these terpyridine complexes, shown in Figure 5-3, are
investigated and compared to the related bipyridine complexes. The focus of this
chapter is the similarities and differences between the two families of molecules
in the ground state and long-lived excited state.

156

5.3.1 Synthesis
Hanan, Campagna, and coworkers originally synthesized these complexes in
the hope that the coordinating ability of the cyanide group would make them
useful building blocks for more complex supermolecular assemblies.17

The

authors used the unusual chemistry-on-the-complex approach to synthesize
these complexes because, at the time, there was not a high yielding method
available to synthesize the free 4'-cyano-2,2':6',2"-terpyridine (CN-tpy) ligand.
Pre-coordinating the starting chloro-substituted ligand to the ruthenium center
made the terpyridine more reactive to substitution and prevented it from
coordinating to and deactivating the Pd catalyst. The starting Cl-tpy ruthenium
complexes and final CN-tpy ruthenium products were synthesized exactly as
reported in the literature, without any changes or modifications.
After recrystallization, [Ru(CN-tpy)2]2+ complex was isolated with mixture of
hexafluorophosphate and nitrate counterions.

The elemental analysis results

suggest the average formula is [Ru(CN-tpy)2](NO3)1.5(PF6)0.5.

The infrared

spectrum of the compound (Figure B-4) shows infrared stretches at 831 cm-1 and
558 cm-1 for PF6 and a broad feature at 1300 - 1400 cm-1 which is most likely
due to the nitrate ion. The identity of the counterion, however, will have no effect
on the photophysical properties of the complex.

5.3.2 Ground State Spectroscopic Properties
The

ground

state

absorption

spectra

of

the

two

complexes

and

[Ru(tpy)2](PF6)2 are shown in Figure 5-4. The electronic absorption spectra of
ruthenium terpyridyl complexes are qualitatively very similar to the ruthenium
bipyridyl complexes discussed in Chapter 3. They exhibit both intraligand as well
as charge transfer transitions, but unlike the bipyridine complexes there isn't an
obvious distinction between the absorption features associated with the CN-tpy

157

Wavelength (nm)
350
400

300

500

600

Normalized Absorbance

4

3

2

1

0

35

30

25
3

20

15

-1

Energy (x10 cm )

Figure 5-4: Normalized absorption spectra of [Ru(tpy)(CN-tpy)]2+ (1, blue) and
[Ru(CN-tpy)2]2+ (2, red) and [Ru(tpy)2]2+(black) in acetonitrile solution. Spectra
are normalized at the maximum of the MLCT absorption feature to better show
the change in shape of the MLCT absorption band across the series.

and the unsubstituted tpy.

Based on extinction coefficients as well as

comparison to the cyano-substituted ruthenium bipyridine complexes, the UV
absorption features in complexes 1-2 can be assigned as π

π* absorptions of

the terpyridine ligands whereas the somewhat weaker visible features are metalto-ligand charge transfer (MLCT) in nature. In [Ru(tpy)2](PF6)2 there are two UV
absorptions, one at 270 nm and one at 307 nm which can both be readily
assigned as π

π* absorptions of the terpyridine ligand by comparison to the

reported absorption spectrum of [Zn(tpy)2]2+.18

Complex 2 has two UV

absorptions at 273 nm and 317 nm. These are assigned to the analogous π

π*

absorptions of the CN-tpy ligand. In complex 1, the UV absorption features are
at 271 nm and 309 nm, which fall between the absorption maxima of the two
homoleptic complexes.

These features are a superposition of absorption
158

features of the tpy ligand and the CN-tpy ligand; it is not possible to differentiate
the contributions from the different ligands as in the bipyridine complexes.
Assignments within the charge transfer band are not straight forward but can
be clarified using electrochemistry. As discussed in Chapter 3, the energy of a
metal-to-ligand charge transfer band can be thought of in terms of the energy
required to oxidize the metal and reduce the ligand.19 For two ligands bound to
the same metal, the energy of the MLCT states will therefore correlate with the
reduction potential of the ligands. The electrochemical properties of complexes 1
and 2 have been investigated by cyclic voltammetry (CV) and differential pulse
voltammetry (DPV), the results of which are shown in Figure 5-5 and listed in
Table 5-1.
The electron-withdrawing nature of the cyano group is apparent in the trends
in the ruthenium oxidation potential across the series, although the effect is not
as dramatic as in the bipyridine complexes.

Starting with [Ru(tpy)2]2+, each

successive replacement of 2,2':6',2"-terpyridine by CN-tpy in the coordination
sphere systematically shifts the oxidation potential of the metal center positive by
ca. 100 mV (compared to 130 mV for the bipyridine complexes). This shift in
charge density at the ruthenium indicates the comparatively poorer electron
donating ability of the CN-tpy relative to the unsubstituted terpyridine ligand.
To assign the ligand reductions, first compare the reduction potentials of
complex 2 and [Ru(tpy)2](PF6)2. In these cases, there is no question of which
ligand is being reduced. The CN-tpy ligand in complex 2 is more easily reduced
than the tpy ligand in [Ru(tpy)2]2+ by 380 mV which is significant, but less than
the 480 mV seen in the bipyridine complexes. Next, compare the heteroleptic
complex 1 to the two homoleptic complexes; the first reduction potential is much
closer in energy to that of complex 2, whereas the second reduction potential is
much closer to the [Ru(tpy)2]2+. This, combined with the larger energy splitting

159

0.5 µA
-1.2

-1.6

-2.0

1.2 1.0 0.8

1.5

1.0

0.5

0.0
-0.5
Potential (V)

-1.0

-1.5

-2.0

Figure 5-5: Cyclic voltammogram of [Ru(tpy)(CN-tpy)](PF6)2 (1) in CH3CN with
0.1 M TBAPF6 as the electrolyte. Spectrum was collected using the standard
three electrode setup with a platinum working electrode, graphite counter
electrode, and a Ag/AgCl reference electrode and is referenced to the ferrocene/
ferrocenium redox couple, which was added as an internal standard. The
differential pulse voltammogram is shown in the insets

Table 5-1: Electrochemical Data for Complexes 1-2 and [Ru(tpy)2](PF6)2a
Complex

E1/2ox

E1/2red1

E1/2red2

ΔE (eV)b

[Ru(tpy)2](PF6)2

+0.92

-1.63

-1.88

2.55

[Ru(tpy)(CN-tpy)](PF6)2

+ 1.03

-1.34

-1.81

2.37

[Ru(CN-tpy)2](PF6)2

+1.13

-1.25

-1.53

2.38

aPotentials

are reported in V versus the ferrocene/ferrocenium couple in
acetonitrile solution as described in the experimental section.
bΔE = (E ox -E red1).
1/2
1/2
160

between the first and second reductions show that the first reduction potential is
the reduction of the CN-tpy ligand and the second reduction is of the
unsubstituted tpy ligand.
With the electrochemical results, assigning the electronic absorption features
is now simplified. First, start with the two homoleptic complexes, complex 1 and
[Ru(tpy)2]2+. In complex 2 the narrow charge transfer absorption feature at 491
nm is solely due to metal-to-ligand charge transfer transitions to the CN-tpy
ligand and in [Ru(tpy)2]2+ the MLCT transition to the terpyridine ligand is blueshifted to 475 nm due in part to the higher reduction potential of the terpyridine

420

440

Wavelength (nm)
460 480 500
550
485

1.0

600
490

495

650
500

Normalized Absorbance

1.0
0.9

0.8

0.8

0.6

0.7
20.8 20.6 20.4 20.2 20.0

0.4
0.2
0.0
24

22

20

18
3

16

-1

Energy (x10 cm )
Figure 5-6: Electronic absorption spectra of [Ru(CN-tpy)2]2+ (2) in methanol (red
line, λmax= 488 nm), acetonitrile (orange line, λmax= 491 nm), acetone (green
line, λmax= 491 nm), nitromethane (blue line, λmax= 492 nm), DMF (purple line,
λmax= 494 nm), and DMSO (black line, λmax= 495 nm). The inset shows an
expanded view of the absorption maxima.
161

ligand. Complex 1 will have both MLCT transitions to CN-tpy and to tpy. Unlike
the bipyridine complexes, there is no clear distinction between the two absorption
features. Comparison of the band shapes in complexes 1 and 2 does suggest
some differentiation between the CN-tpy and tpy absorption features. The MLCT
absorption feature in complex 1 has a definite shoulder on the blue edge of the
absorption band that is not present in the homoleptic complex. This broadening
of the charge transfer manifold in complex 1 is due to the addition of tpy charge
transfer transitions to the absorption feature.
Solvatochromism is an important feature of some cyanide-containing
compounds.

In ruthenium(II) cyanide complexes such as [Ru(bpy)CN4]2- the

absorption spectrum can shift over 6800 cm-1 (~150 nm) by changing
solvents.20,21

Molecular

donor-acceptor

systems

that

incorporate

[Ru(bpy)CN4]2- can change the direction of energy transfer by changing the
solvent.22-24 While the solvatochromatic response for complexes 1 and 2 is not
nearly as dramatic as the ruthenium cyanide complexes (vide supra) there is a
definite shift in the MLCT absorption maximum as a function of solvent, which
can be seen in Figures 5-6 and B-6. There have been many different scales
(donor number, acceptor number, dipole moment, and Z-values for example)
used to quantify the effect a solvent has on electronic absorption.25-27
The trends seen in solvatochromism can also give insight to the electronic
structure of the ground state and/or the excited state and how they interact with
solvent. The solvatochromatic shift in the ruthenium cyanide complexes is a
function of the solvent acceptor number, which is the electron accepting strength
of the solvent. Solvents that are good electron acceptors shift the absorption
maximum to higher energy and solvents that are poor electron acceptors shift the
absorption maximum to lower energy.21 This trend shows that the solvent is
specifically interacting with the molecule and accepting electron density,

162

MLCT Absorption Maximum
3
-1
(x10 cm )

20.40

methanol

20.35

acetone

acetonitrile
nitromethane

20.30
20.25

DMF

20.20

DMSO

20.15
1.5

2.0

2.5

3.0

3.5

4.0

Dipole Moment (D)

Figure 5-7: Plot of MLCT absorption maximum vs solvent dipole moment for
[Ru(CN-tpy)2]2+ in various solvents (see Figure 5-6). The plot shows a
correlation between the absorption maximum and the solvent dipole moment,
suggesting the solvatochromatic response is caused by the charged excited state
being stabilized by solvent dipole interactions.

therefore stabilizing it. The fact that the strongest electron acceptors give the
highest energy absorption features indicates that the solvent interaction is
strongest in the ground state and that the ground state is stabilized more than the
excited state. As seen in Figure 5-7, the solvatochromatic response for complex
2 is correlated with the solvent dipole moment; solvents with a large dipole shift
the spectrum lower in energy and solvents with a small dipole moment shift the
spectrum to higher energy. This indicates that, unlike the ruthenium cyanide
complexes, the solvent has no specific interactions with the molecule and the
solvatochromatic shift is due to a dipole interaction. The absorption maximum is

163

shifted to lower energy in more polar solvents; this suggests that the solvent is
primarily stabilizing the excited state and lowering its energy. This phenomenon
is seen in charge transfer excited states with and without cyanide groups. This
behavior is another confirmation that the initial excited state has a significant
dipole, and is therefore localized on a single ligand.28

5.3.3 Steady-State and Time-Resolved Emission Spectroscopies
The room temperature emission spectra for the two complexes are plotted in
Figure 5-8a. The emission maxima for complexes 1 and 2 are at 696 nm and
688 nm respectively, following the same trend as the ruthenium oxidation
potentials across the series. Radiative quantum yields for complexes 1-2 were
determined relative to both the [Ru(bpy)3](PF6)29 and [Os(bpy)3](PF6)210,11
standards, and are listed in Table 5-2.

The quantum yield values for these

complexes may seem small in comparison to the ruthenium bipyridine complexes
in Chapter 3, but when compared to other ruthenium terpyridine complexes, they
are actually quite large; for example, [Ru(ph-tpy)2]2+ (ph-tpy = 4'-phenyl-2,2':6'2"tepyridine) has a quantum yield of 4 x 10-5 and [Ru(tpy)2]2+ has a quantum yield
of less than 5 x 10-6.29
To obtain more quantitative insights into the 3MLCT excited states of this
system, I carried out a single mode spectral fitting analysis using Equation 5-2
as described by Meyer and coworkers and the procedure described in Chapter
3.13,30 Briefly, the low temperature emission spectra (Figure 5-8b) was fit to
yield well defined values for the four parameters (E0, SM, ωM, and Δν̅0,1/2). The
value of

ωM was assumed to be temperature independent, so the room

temperature spectra were fit using the low temperature value of

ωM as a fixed

value. Using this approach, well defined (<10% variation) values for the 3MLCT
zero-point energy (E0), Huang-Rhys factor (SM), and the spectral bandwidth

164

(Δν̅0,1/2) could be determined for the room-temperature spectra. Spectral fitting
parameters for complexes 1 and 2 are listed in Table 5-3.

Wavelength (nm)
550

600

650 700

800

Normalized
Emission Intensity

1.0

(a)

0.8
0.6
0.4
0.2
0.0
1.0

Normalized
Emission Intensity

900 1000

(b)

0.8
0.6
0.4
0.2
0.0
18

16

14

12
3

10

-1

Energy (x10 cm )
Figure 5-8: Steady-state emission spectra of [Ru(tpy)(CN-tpy)](PF6)2 (1, blue
triangles) and [Ru(CN-tpy)2](PF6)2 (2, red circles) (a) Room-temperature spectra
acquired in deoxygenated CH3CN solution. The solid lines correspond to fits to
an asymmetric double sigmoidal function – see text for further details. (b)
Emission spectra acquired in a 9:2 butyronitrile/propionitrile glass at 80 K

165

Table 5-2: Photophysical Data for Complexes 1 and 2
kobs a
(x107 s-1)

kr a,b
(x104 s-1)

knr a,c
(x107 s-1)

Complex

λem
(nm)

Φa
(x10-3)

[Ru(tpy)(CN-tpy)]2+

696

1.9 ± 0.2

1.25 ± 0.03 2.38 ± 0.24 1.25 ± 0.03

[Ru(CN-tpy)2]2+

688

1.7 ± 0.2

1.82 ± 0.09 3.09 ± 0.37 1.82 ± 0.09

a Error bars determined by the standard
bk =Φ•k
c
r
r obs. knr = kobs- kr.

deviation of multiple measurements.

The fitted values for E0 for all three complexes are very close to the observed
emission maxima, indicating that the ν* = 0

ν = 0 transition is the dominant

contribution to the observed emission spectrum and that the ground state and
excited state potential surfaces are fairly nested. The Huang-Rhys factor is a
measure of the vibrational overlap between the excited state and the ground
state and is proportional to (ΔQ)2, which makes SM a useful parameter for
gauging the amount of structural distortion in the excited state relative to the
ground state.15

As with the bipyridine complexes, is difficult to draw any

conclusions from the differences in the room temperature SM value since the
variations in SM are within the uncertainties of the fits.

The values for the

terpyridine are very similar to those of the bipyridine complexes.

When

compared to [Ru(bpy)3]2+ (SM = 1.01 ± 0.08), the Huang-Rhys factors for the
terpyridine complexes are significantly smaller, which indicates an increase in
delocalization.

This increased delocalization could be due to the addition of the

cyanide group but it just as easily be due to the additional pyridine ring in the
terpyridine ligand. Unfortunately the more informative comparison to [Ru(tpy)2]2+
is not possible because [Ru(tpy)2]2+ is not measurably emissive in a room
temperature solution.

166

The lifetimes of complexes 1 and 2 were also determined using nanosecond
time-resolved emission (Figure 5-9). Radiative and non-radiative decay rates for
both complexes were calculated from the lifetimes and quantum yields (Table
5-2).

Table 5-3. Spectral Fitting Results for Terpyridine Complexes

Complexes

Low Temperaturea,b
E0 (cm-1)

SM

ωM (cm-1) Δν̅0,1/2(cm-1)

[Ru(tpy)(CN-tpy)]2+

15045 ± 50

0.55 ± 0.04

1135 ± 40

910 ± 60

[Ru(CN-tpy)2]2+

15274 ± 60

0.67 ± 0.05

1161 ± 40

1000 ± 60

Room Temperatureb,c
E0 (cm-1)

SM

ωM (cm-1)d Δν̅0,1/2(cm-1)

[Ru(tpy)(CN-tpy)]2+

14445 ± 70

0.68 ± 0.1

1135

1510 ± 100

[Ru(CN-tpy)2]2+

14820 ± 80

0.82 ± 0.12

1161

1795 ± 140

a Fitting results for 80 K emission spectra in 9:2 butyronitrile:propionitrile.
b Error bars represent an approximate range of visually equivalent fits. c Fitting

results for room temperature emission spectra in CH3CN, using a fixed value of
ωM based on the low temperature data. See text for details.

The results in Table 5-2 clearly show that the deactivation of the 3MLCT state
in the terpyridine complexes is dominated by non-radiative decay, which is very
different from what is seen in the bipyridine complexes.

In ruthenium(II)

terpyridine complexes, the major deactivation pathway is through the lowest
energy ligand field state(s).

Terpyridine is a weaker field ligand than the

bipyridine, mostly due to its unfavorable bite angle. This means that the ligand
field states in terpyridine complexes are lower in energy and therefore closer in

167

energy to the MLCT excited states. For comparison, the energy barrier between
the lowest energy triplet state and the ligand field states in [Ru(bpy)3]2+ is
approximately 4000 cm-1 whereas in [Ru(tpy)2]2+ it is only 1500 cm-1.31

In

terpyridine complexes, the long lived 3MLCT state can thermally populate the
ligand field manifold, which decays very rapidly down to the ground state nonradiatively.

This means that the rate of non-radiative decay in terpyridine

complexes cannot be predicted by the energy gap law (i.e. the ΔE and ΔQ
between the excited state and the ground state) like it can in bipyridine
complexes.30
3MLCT

Instead, it is a function of the energy difference between the

and 3LF states.

Normalized Emission Intensity

0.0
-0.2
-0.4
-0.6
-0.8
-1.0
0

100

200

300

Time (ns)
Figure 5-9: Time resolved emission decay traces for [Ru(tpy)(CN-tpy)](PF6)2 (1,
blue triangle) and [Ru(CN-tpy)2](PF6)2 (2, red circle). Black lines are single
exponential fits of the emission decay.

168

The results in Table 5-2 show that complex 1 has a slower rate of nonradiative decay than complex 2, which suggests that the energy barrier between
the 3MLCT and 3LF states for complex 1 is larger than for complex 2. At first, it
may seem counterintuitive that a heteroleptic complex would have a higher
energy ligand field state than its corresponding homoleptic complex but work by
Maestri et. al. has shown that to be true.

The authors studied a series of

homoleptic and heteroleptic bis(terpyridyl)ruthenium(II) complexes with electron
donor and acceptor substituents and found that the complexes with the longest
lifetimes were the heteroleptic complexes with one electron donor and one
electron acceptor.29

Their results (Figure 5-10) show that electron donor

!M
*

"*
L

"M
[(tpy)Ru(tpy)]

2+

[(D-tpy)Ru(D-tpy)]
2+

[(A-tpy)Ru(A-tpy)]

2+

2+

[(A-tpy)Ru(D-tpy)]

Figure 5-10: Schematic representation of the effect of electron accepting (A) and
electron donating (D) substituents on the energy of the HOMO (π M) and LUMO
(π*L) orbitals in [Ru(tpy')2]2+ complexes. Purple arrows represent ligand field
transitions and green arrows represent metal-to-ligand charge transfer
transitions. Reproduced from reference 29.

169

substituents primarily destabilize the HOMO (π M) orbital, lowering both the
MLCT and LF energies and electron acceptor substituents primarily stabilizes the
LUMO (π L) orbital, lowering the MLCT energy and (slightly) raising the LF
energy. In a heteroleptic complex with both donor and acceptor substituents the
HOMO will be destabilized and the LUMO will be stabalized, leading to a slightly
lower LF state and a much lower MLCT state. Experimentally, the heteroleptic
complexes in the study had the largest energy barrier between the MLCT and LF
states and the longest lifetimes.
The values for the radiative decay rate (kr) in complexes 1 and 2 are similar
to those seen in the bipyridine complexes in Chapter 3.

The difference in kr

between the two complexes can be explained by radiative decay theory, which
specifies a proportionality between kr and the product μ2 • E03 where μ is the
transition dipole moment and E0 is the energy gap.32 This means the higher
radiative decay rate in complex 2 can be explained by the larger energy gap
between the ground state and the excited state.

5.3.4 Time-Resolved Electronic Absorption Spectroscopy.
Nanosecond transient absorption spectroscopy is another important tool for
characterizing the lowest energy excited state in transition metal complexes.33
The transient absorption spectra of complexes 1 and 2 acquired following 1A1
1MLCT

excitation at 510 nm are shown in Figure 5-11.

Like the bipyridine

complexes in Chapter 3, the most striking aspect of Figure 5-11 is the similarity
between the differential absorption spectra for the two complexes; this is in
agreement with the notion anticipated from the electrochemistry data that the
character of the long-lived 3MLCT excited state is the same in both complexes.
There are three principle features observed in the spectra: a strong net
absorption in the near-UV, a strong bleach in the mid-visible, and a weak

170

absorption extending into the red.

This overall pattern is typical of what is

observed for the excited state spectra of RuII polypyridyl complexes.
By comparison to the bipyridine complexes in Chapter 3 the spectra can be
tentatively assigned as follows: the strong absorption in the near UV is primarily
due to the absorption of the terpyridyl radical anion in the excited state, the
bleach in the mid visible is due to the loss of the ground state MLCT absorption,
and the weaker absorption in the red is due to a combination of absorption of the
terpyridyl radical anion and an excited state LMCT transition.

350

Wavelength (nm)
400 450 500 600 700
(a)

Change in
Absorbance

0.1
0.0
-0.1

Change in
Absorbance

-0.2
(b)

0.1
0.0
-0.1
-0.2
-0.3

28

24

20
3

16
-1

Energy (x10 cm )
Figure 5-11: Nanosecond time-resolved differential absorption spectra acquired
in room-temperature CH3CN solution for (a) [Ru(tpy)(CN-tpy)](PF6)2 (1) and (b)
[Ru(CN-tpy)2](PF6)2 (2). The individual points correspond to the amplitudes of fits
of the kinetics data to single-exponential decay models; a smoothed solid line
has been included in both plots to guide the eye.

171

CN
NC

N
N

N

N

Ru
N

CN

N
N
Ru

N

N

N

N

N

4'-cyano-2,2':6'2"-terpyridine
CN-tpy

4,4"-dicyano-2,2':6'2"-terpyridine
CN2-tpy

CN
NC

CN

N
N

N
Ru

N

N
N

4,4',4"-tricyano-2,2':6'2"-terpyridine
CN3-tpy

Figure 5-12: A series of ruthenium(II) terpyridyl incorporating an increasing
number of electron withdrawing cyanide groups.

5.4 Concluding Comments and Future Directions
The study of complexes 1 and 2 discussed in this chapter has laid the
groundwork for a wider study of the photophysical properties of ruthenium(II)
terpyridyl complexes.

While there have been some systematic studies of

ruthenium(II) terpyridyl complexes, in general they are less studied and not as
well understood as the related bipyridine complexes. Systematically increasing
the number of cyanide groups on the terpyridine ligands, as shown in Figure
5-12, will increase the electronic asymmetry in the complexes. This increased

172

asymmetry should, according to the model developed by Maestri et. al., lengthen
the lifetime and increase the quantum yield of these complexes by increasing the
energy gap between the 3MLCT and 3LF states. Measuring the 3MLCT lifetime
and quantum yield for the expanded series of complexes as well using variable
temperature emission to determine the energy barrier between the 3MLCT and
3LF

states will give additional experimental evidence that increasing asymmetry

increases the energy difference between the 3MLCT and 3LF states in a series of
complexes where the nature of the substituent does not change across the
series.
The meridional binding in bis terpyridine complexes places the two terpyridine
ligands orthogonal to each other. NMR spectroscopy shows that the identity of
one terpyridine has no effect in the 1H or 13C shifts for the other terpyridine
ligand in a heteroleptic complex, meaning the NMR spectrum of a heteroleptic
ruthenium(II) bis terpyridal complex is identical to the superposition of the NMR
spectra of the respective heteroleptic complexes. This suggests that there is little
communication between the two ligands in in the ground state.6 It is not clear if
this is the case in the excited state as well.

Harriman and coworkers have

identified a ~single picosecond rate of interligand electron transfer in a porphyrinsubstituted osmium(II) bis terpyridine complex.34 This relatively rapid electron
transfer suggests that there is communication between the two terpyridine
ligands in the excited state of the osmium complex. While there are quite a few
similarities between osmium(II) and ruthenium(II) polypyridyl complexes, the
ligand field states in an osmium complex are higher in energy than the analogous
ruthenium complex. Since the ligand field states play a very important role in the
excited state dynamics of ruthenium(II) terpyridine complexes, the excited state
behavior of an osmium(II) terpyridine complex is not necessarily a good predictor
of the behavior in a ruthenium(II) complex.

173

With the series of complexes in Figure 5-11, it may be possible to measure
the interligand electron transfer rate, which has not yet been done for a
bis(terpyridyl)ruthenium(II) complex.

The increased asymmetry in these

complexes should increase the separation between the two metal-to-ligand
charge transfer manifolds and hopefully allow for the selective excitation of the
higher energy, unsubstituted terpyridine ligand. The interligand electron transfer
and/or excited state delocalization can be monitored by excited state absorption
anisotropy and infrared spectroscopy as described in Chapter 4. Understanding
the communication (or lack there of) between terpyridine ligands in ruthenium(II)
bis terpyridine complexes is key for incorporating these types of complexes into
light harvesting antenna arrays, molecular wires, or other super-molecular donoracceptor systems.

174

REFERENCES

175

5.5 References
(1) Benniston, A. C.; Chapman, G.; Harriman, A.; Mehrabi, M.; Sams, C. A.
Inorg. Chem. 2004, 43, 4227–4233.
(2)

Indelli, M. T.; Bignozzi, C. A.; Scandola, F.; Collin, J. Inorg. Chem. 1998, 37,
6084–6089.

(3)

Hammarström, L.; Johansson, O. Coord. Chem. Rev. 2010, 254,
2546-2559.

(4)

Abrahamsson, M.; Jager, M.; Kumar, R. J.; Osterman, T.; Persson, P.;
Becker, H. C.; Johansson, O.; Hammarström, L. J. Am. Chem. Soc. 2008,
130, 15533-15542.

(5)

Schramm, F.; Meded, V.; Fliegl, H.; Fink, K.; Fuhr, O.; Qu, Z.; Klopper, W.;
Finn, S.; Keyes, T. E.; Ruben, M. Inorg. Chem. 2009, 48, 5677-5684.

(6)

Constable, E. C.; Cargill Thompson, A. M. W.; Tocher, D. A.; Daniels, M. A.
M. New J. Chem. 1992, 16, 855–867.

(7)

Wang, J.; Fang, Y.; Hanan, G. S.; Loiseau, F.; Campagna, S. Inorg. Chem.
2005, 44, 5-7.

(8)

Richardson, D. E.; Taube, H. Inorg. Chem. 1981, 20, 1278-1285.

(9)

Suzuki, K.; Kobayashi, A.; Kaneko, S.; Takehira, K.; Yoshihara, T.; Ishida,
H.; Shiina, Y.; Oishi, S.; Tobita, S. Phys. Chem. Chem. Phys. 2009, 11,
9850-9860.

(10) Johnson, S. R.; Westmoreland, T. D.; Caspar, J. V.; Barqawi, K. R.; Meyer,
T. J. Inorg. Chem. 1988, 27, 3195-3200.
(11) The reported literature value was corrected to account for the change in
[Ru(bpy)3](PF6)2 quantum yield from 0.062 to 0.095.
(12) Damrauer, N. H.; McCusker, J. K. Inorg. Chem. 1999, 38, 4268–4277.
(13) Claude, J. P.; Meyer, T. J. J. Phys. Chem. 1995, 99, 51-54.
(14) Parker, C. A.; Rees, W. T. Analyst (London). 1960, 85, 587-600.
(15) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K. J. Chem.
Soc. 1997, 119, 8253–8268.

176

(16) Picraux, L. B.; Smeigh, A. L.; Guo, D.; McCusker, J. K. Inorg. Chem. 2005,
44, 7846–7865.
(17) Borgstrom, M.; Ott, S.; Lomoth, R.; Bergquist, J.; Hammarström, L.;
Johansson, O. Inorg. Chem. 2006, 45, 4820-4829.
(18) Albano, G.; Balzani, V.; Constable, E. C.; Maestri, M.; Smith, D. R. Inorg.
Chim. Acta. 1998, 277, 225-231.
(19) Vlček, A. A.; Dodsworth, E. S.; Pietro, W. J.; Lever, A. B. P. Inorg. Chem.
1995, 34, 1906-1913.
(20) Bignozzi, C. A.; Chiorboli, C.; Indelli, M. T.; Scandola, M. A. R.; Varani, G.;
Scandola, F. J. Am. Chem. Soc. 1986, 108, 7872–7873.
(21) Timpson, C. J.; Bignozzi, C. A.; Sullivan, B. P.; Kober, E. M.; Meyer, T. J. J.
Phys. Chem. 1996, 100, 2915–2925.
(22) Indelli, M. T.; Ghirotti, M.; Prodi, A.; Chiorboli, C.; Scandola, F.;
McClenaghan, N. D.; Puntoriero, F.; Campagna, S. Inorg. Chem. 2003, 42,
5489–5497.
(23) Easun, T. L.; Alsindi, W. Z.; Deppermann, N.; Towrie, M.; Ronayne, K. L.;
Sun, X. Z.; Ward, M. D.; George, M. W. Inorg. Chem. 2009, 48, 8759-8770.
(24) Simpson, N. R. M.; Ward, M. D.; Morales, A. F.; Barigelletti, F. J. Chem.
Soc. Dalton Trans. 2002, 2449-2454.
(25) McRae, E. G. J. Phys. Chem. 1957, 61, 562-572.
(26) Kosower, E. M. J. Am. Chem. Soc. 1958, 80, 3253-3260.
(27) Kamlet, M. J.; Abboud, J. L. M.; Abraham, M. H.; Taft, R. W. J. Org. Chem.
1983, 48, 2877-2887.
(28) Kober, E. M.; Sullivan, B. P.; Meyer, T. J. Inorg. Chem. 1984, 23,
2098-2104.
(29) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; Cargill Thompson,
A. M. W. Inorg. Chem. 1995, 34, 2759–2767.
(30) Kober, E. M.; Caspar, J. V.; Lumpkin, R. S.; Meyer, T. J. J. Phys. Chem.
1986, 90, 3722-3734.

177

(31) Sauvage, J. P.; Collin, J. P.; Chambron, J. C.; Guillerez, S.; Coudret, C.;
Balzani, V.; Barigelletti, F.; De Cola, L.; Flamigni, L. Chem. Rev. 1994, 94,
993-1019.
(32) Kestner, N. R.; Logan, J.; Jortner, J. J. Phys. Chem. 1974, 78, 2148-2166.
(33) McCusker, J. K. Acc. Chem. Res. 2003, 36, 876–887.
(34) Benniston, A. C.; Harriman, A.; Pariani, C.; Sams, C. A. J. Phys. Chem. A.
2007, 111, 8918-8924.

178

APPENDICES

179

"0#(

'0$!
'0$1

(01(

'0'(
'0&!

(0$)

(

&011

'0%

$01&
!0(%
#0))
$0!(

!0)$
$0!(
)

(01
*!+,--./

(01(
'0'(
'0&$
'0&!
'0$!
'0$1

(0%

(0&)
(0$)

(0&)

Appendix A: Additional Figures for Chapter 3

'

&

%
*!+,--./

$

#

"

!

Figure A-1: 1H-NMR of [Ru(bpy)2(CN-Me-bpy)](PF6)2 in CD3CN. Inset shows
an expanded view of the aromatic region.

180

Wavelength (nm)
300

350

400

500

600

Normalized Absorbance

6
10 min
20 min
25 min
30 min
35 min
37 min
40 min
42 min
45 min
46 min

5
4
3
2
1
0

35

30

25
3

20
-1

Energy (x10 cm )

Figure A-2: Electronic absorption spectra of the conversion of [Ru(bpy)(Cl-Mebpy)2]2+ to [Ru(bpy)(CN-Me-bpy)2]2+. The inset indicates the time following
initiation of the reaction that each aliquot was taken. See text for further details.

181

'

!0#*
!0"&

&0$'
&0$&
&0#!
&0$

*!011

'01
)*+,--./

#0*'
*0'#
*0((

*0('

#01&
*0(&
(

'0*!
&0%'

'0&1
'0%'
'0$1
'0#(
&0%'
&0%#
&0$'
&0$&
&0#!

'0*!

'0$1
'0#(

'0&1

'0$

&

%

$
)*+,--./

#

"

!

Figure A-3: 1H-NMR of [Ru(bpy)(CN-Me-bpy)2](PF6)2 in CD3CN. Inset shows
an expanded view of the aromatic region.

182

1 !A

-1.2

-2.0

1.4 1.2 1.0

1.0

0.0

-1.0
Potential (V)

-2.0

Figure A-4: Cyclic voltammogram of [Ru(bpy)(CN-Me-bpy)2](PF6)2 in CH3CN
with 0.1 M TBAH as the electrolyte. Spectrum was collected using the standard
three electrode setup with a platinum working electrode, graphite counter
electrode, and a Ag/AgCl reference electrode and is referenced to the ferrocene/
ferrocenium redox couple, which was added as an internal standard. The
differential pulse voltammogram is shown in the insets.

183

Wavelength (nm)
300

350

400

500

Normalized Absorbance

7

600

Start
15 min
30 min
40 min
45 min
55 min
60 min
65 min
70 min
75 min
78 min
80 min

6
5
4
3
2
1
0

35

30

25
3

20
-1

Energy (x10 cm )

Figure A-5: Electronic absorption spectra of the conversion of [Ru(Cl-Mebpy)3]2+ to [Ru(CN-Me-bpy)3]2+. The inset indicates the time following initiation
of the reaction that each aliquot was taken. See text for further details.

184

)

"0$!

'0%)

(0'!
'0%)

(0'!

'0%

(

&011

(0%
(01
*!+,--./

!0'!

!0)'

)01

'

&

%
*!+,--./

$

#

"

!

Figure A-6: 1H-NMR of [Ru(CN-Me-bpy)3](PF6)2 in CD3CN. Inset shows an
expanded view of the aromatic region.

185

2 !A
-1.2

-1.6

1.4 1.2 1.0

1.6

1.2

0.8

0.4

0.0

-0.4

-0.8

-1.2

-1.6

Potential (V)

Figure A-7: Cyclic voltammogram of [Ru(bpy)3](PF6)2 in CH3CN with 0.1 M
TBAH as the electrolyte. Spectrum was collected using the standard three
electrode setup with a platinum working electrode, graphite counter electrode,
and a Ag/AgCl reference electrode and is referenced to the ferrocene/
ferrocenium redox couple, which was added as an internal standard. The
differential pulse voltammogram is shown in the insets.

186

Change in
Absorbance

Change in Molar Absortivity
Absorbance (x103 M-1 cm-1)

350

Wavelength (nm)
400
500 600 700

15

(a)

10
5
0
0.4

(b)

0.010

0.2

0.000
16

0.0

14

12

-0.2
(c)

0.10
0.05
0.00

Change in
Absorbance

-0.05
(d)

0.1
0.0
-0.1
28

24
20
16
3
-1
Energy (x10 cm )

Figure A-8: (a) Ground state absorption spectrum for [Ru(bpy)(CN-Mebpy)2](PF6)2 (2) in room-temperature CH3CN solution. (b) Oxidative difference
spectra acquired at an applied potential of +1.68 V versus Ag/AgCl. The inset
corresponds to an expanded view of the low-energy portion of the spectrum. (c)
Reductive difference spectra acquired at an applied potential of -0.980 V versus
Ag/AgCl. (d) Time-resolved differential absorption spectrum of compound 1
following ca. 10 ns excitation at 500 nm.

187

Wavelength (nm)

Change in
Absorbance

Molar Absortivity
3 -1
-1
(x10 M cm )

350

400

20
15
10
5
0
0.4

500

600 700
(a)

0.010
0.005
0.000

0.2

(b)
14

0.0

12

-0.2

Change in
Absorbance

Change in
Absorbance

0.2

(c)

0.1
0.0
-0.1
0.05

(d)

0.00
-0.05
-0.10

28

24

20
3

16
-1

Energy (x10 cm )
Figure A-9: (a) Ground state absorption spectrum for [Ru(CN-Me-bpy)3](PF6)2
(3) in room-temperature CH3CN solution. (b) Oxidative difference spectra
acquired at an applied potential of +1.85 V versus Ag/AgCl. The inset
corresponds to an expanded view of the low-energy portion of the spectrum. (c)
Reductive difference spectra acquired at an applied potential of -0.900 V versus
Ag/AgCl. (d) Time-resolved differential absorption spectrum of compound 1
following ca. 10 ns excitation at 500 nm.

188

-1

Wavenumber (cm )
2250

Absorbance

2150

2100

v(CN)*

(a)

0.020
0.015

0.2
0.010
0.1

0.005

0.0

Change in Absorbance

v(CN)

0.3

Change in Absorbance

2200

0.000

v(CN)

0.020

250 ns
300 ns
500 ns
750 ns
1000 ns
1250 ns
1500 ns
2000 ns

0.015
0.010

(b)

0.005
0.000
2250

2200

2150

2100
-1

Wavenumber (cm )

Figure A-10: Steady-state and time-resolved infra-red absorption data for 2 in
CH3NO2 solution. (a) Comparison of the ground-state (blue line) and step-scan
infrared differential absorption data acquired at a time delay of 150 ns following
ca. 10 ns excitation at 540 nm (red line). The ~40 cm-1 red-shift in the CN
stretching frequency reflects the presence of an electron in the π* orbital of the
CN-Me-bpy ligand in the 3MLCT excited state. (b) Nanosecond step-scan
infrared spectra as a function of time following ca. 10 ns excitation at 540 nm.

189

0.25

v(CN)

v(CN)*

0.006

0.15

0.004

0.10
0.002
0.05
0.000

0.00
-0.05

Change in Absorbance

Absorbance

0.20

v(CN)

2250

2200

2150

2100

-1

Energy (cm )

Figure A-11: Steady-state and time-resolved infra-red absorption data for 3 in
CH3NO2 solution. Comparison of the ground-state (blue line) and step-scan
infrared differential absorption data acquired at a time delay of 150 ns following
ca. 10 ns excitation at 500 nm (red line). The ~40 cm-1 red-shift in the CN
stretching frequency reflects the presence of an electron in the π* orbital of the
CN-Me-bpy ligand in the 3MLCT excited state. Low signal to noise is primarily
due to the lower solubility of complex 3.

190

Appendix B: Additional Figures for Chapter 5

Figure B-1: 1H NMR spectrum of [Ru(tpy)(CN-tpy)](PF6)2 in CD3CN.

191

80

1602
1449

% Transmission

2240

3116

60

40

558

20
835

4000

3500

3000

2500

2000

1500

1000

500

-1

Energy (cm )

Figure B-2: Infrared spectrum of [Ru(tpy)(CN-tpy)](PF6)2 (1) in a KBr pellet

192

Figure B-3: 1H NMR spectrum of [Ru(CN-tpy)2]2+ in CD3CN.

193

80

1603

50

558

1353
1426

% Transmission

1625

60

2240

3121

70

40
30
831

4000

3500

3000

2500
2000
-1
Energy (cm )

1500

1000

500

Figure B-4: Infrared spectrum of [Ru(CN-tpy)2](NO3)1.5(PF6)0.5 (2) in a KBr
pellet

194

0.2 !A
-1.2

-1.6

1.2 1.0

1.5

1.0

0.5

0.0
-0.5
Potential (V)

-1.0

-1.5

-2.0

Figure B-5: Cyclic voltammogram of [Ru(CN-tpy)2]2+ (2) in CH3CN with 0.1 M
TBAH as the electrolyte. Spectrum was collected using the standard three
electrode setup with a platinum working electrode, graphite counter electrode,
and a Ag/AgCl reference electrode and is referenced to the ferrocene/
ferrocenium redox couple, which was added as an internal standard. The
differential pulse voltammogram is shown in the insets

195

420

440

460

Wavelength (nm)
480 500

Normalized Absorbance

1.0

550

600

480 485 490 495 500
1.0

0.8
0.9

0.6

0.8
0.7

0.4

21.0

20.5

20.0

0.2
0.0
24

23

22

21

20

19
3

18

17

16

-1

Energy (x10 cm )

Figure B-6:
Electronic absorption spectra of [Ru(tpy)(CN-tpy)](PF6)2 in
acetonitrile (red line, λmax= 488 nm), acetone (orange line, λmax= 488 nm),
nitromethane (green line, λmax= 490 nm), DMSO (blue line, λmax= 492 nm), and
pyridine (purple line, λmax= 492 nm). The inset shows an expanded view of the
absorption maxima.

196

Appendix C: Instructions for the Collection Step Scan IR Data on the
Bruker IFS66/S
C.1 Instrument Warm Up Procedure
1. If it isn't on already, turn on the step scan instrument and globar source on
the front of the SSIR power supply. A green light should light up on the MIR
source button. The globar will need about an hour or two to warm up.
2. Turn on the laser by turning the key on the front of the laser power supply
and turning on the switch on the back of the laser. Open the OPOTEK
control software and wait for the temperature to reach 35°C (temperature can
be checked on the remote or under laser status in the software). This will
take ~15-20 minutes. Once the correct temperature is reached turn on the
flashlamps by clicking the lamp button. Choose your wavelength and click
the tune button. Set Laser energy to 100 and turn on the Q switch by clicking
the Laser button (make sure the beam block is in front of the laser). Let the
laser warm up ~30-45 min before using.

C.2 Collection of the Ground State Spectrum
1. Float the table by opening the rightmost valve all the way and increasing the
pressure to ~70 psi. Adjust the flow to the SSIR (left valve) so there is
sufficient flow to move the interferometer mirror (the light on the front of the
instrument will be blinking green). The table will take a couple minutes to
pressurize and float.
2. Open the cover to the detector compartment and fill the detector with liquid
nitrogen. Close the cover and connect the DC output on the cover to the
detector X input on the SSIR.

197

3. Open the OPUS 5.5 software (the password is OPUS) and then open the
Advanced Measurement dialog box from the Measure menu and fill in all the
necessary information (See below for example list of settings)
4. Go to the Check Signal tab.

You should see the detector signal in the

window. Adjust the three knobs on the parabolic mirror focusing light into the
detector (it can be accessed from the sample compartment) to maximize the
signal. It is easiest to change the display from interferogram to counts to do
this.
5. One the signal is maximized close the sample compartment and purge the
sample chamber with N2 using the middle valve and purge the rest of the
instrument by increasing the purge flow to 1000-1500 L/hr on the instrument
regulator. After purging 5-10 min the signal intensity will stabilize. Run the
Auto Align sequence to make adjustments to the mirrors in the
interferometer.
6. Fill the cell with your solvent of choice and put it in the sample holder facing
the interferometer.
7. Vary the aperture setting (Optic tab) to make the signal as large as possible
without saturating the detector.

The detector is saturated when the

interferogram is asymmetric (positive peak larger or smaller than the
negative one). Collect a solvent scan with the sample single channel button.
8. Remove the cell and replace the solvent with the sample solution. Place the
cell back in the sample holder and purge the sample compartment for 5
minutes.
9. Collect a molecule scan with the sample single channel button.

198

10. Calculate the absorbance spectrum by:
• Divide sample spectrum by solvent spectrum with the spectrum calculator
(in toolbar or Manipulate menu) to get the transmission spectrum.
• Convert the transmission to absorbance using the AB<->TR conversion. If
it's not automatically selected, just drag the transmission block into the
files to be converted box.
• The absorbance of the vibrations of interest should be ~0.3 to 0.7.

C.3 Collection of the DC Coupled (Ground State) Step-Scan Spectra
1. The step scan experiment is noise and vibration sensitive, so turn off any
music and speak quietly while the experiment is in progress.
2. Open the Advanced Measurement box and go to the check signal tab.
Connect the DC output of the detector to the oscilloscope. Set divisions to
500 mV and 200 us.
3. The signal going to the external digitizer has to be smaller than ±1V. Vary the
aperture to make sure the DC signal is within ±1V. Change the scope to
ground mode to find where 0 V is. If the signal is less than 2 V peak-to-peak
but is offset from zero use a small screwdriver in the hole on top of the
detector to adjust the bias voltage (and the offset of the signal).
4. Connect the DC output of the detector to channel A of the digitizer.
Terminate the AC output of the detector and channel B of the digitizer.
5. Turn off the purge to the sample compartment and turn the flow for the
instrument purge down to ~500 L/hr
6. Open the Time Resolve Step Scan dialog box in the Measure menu

199

7. Fill in the appropriate settings.

Most settings should transfer from the

Advanced Measurement dialog box. See below for example settings. Note
the interferogram size. This is the number of mirror positions needed to
collect the spectra. The scan progress bar at the bottom of the screen will
count number of mirror positions.
8. Collect the DC spectrum by clicking the Start Step Scan Time Resolved
Measurement button
9. The resulting spectra should show an identical interferogram at all time
points. Sometimes the 3D viewing window will only show the positive half of
the interferogram.

C.4 Collection of the AC Coupled (Excited State) Step-Scan Spectra
1. Purge the instrument and sample compartment for 5-10 minutes while setting
up.
2. Using the power meter, tune the laser with the two knobs on the back to get
the maximum power.
3. Measure the power as close to the sample as possible. The desired power
will vary by molecule, but usually ~3-4 mJ per pulse at the sample is a good
power. If the power is too high adjust it using the laser energy slider in the
OPOTEK control software.
4. Turn on the trigger and connect it to the oscilloscope. The trigger signal
needs to be 4-5 V in intensity and at least 200 ns wide. Adjust the trigger
position to meet these requirements.

Disconnect from the scope and

connect to the trigger channel of the digitizer.

200

5. Connect the DC output of the detector to channel B of the digitizer. Connect
the AC output of the detector to the amplifier and connect the amplifier output
to channel A of the digitizer. Turn on the amplifier.
6. Unblock the laser and verify that the laser is going through the glass window
and hitting the center of the sample.
7. Turn off the sample compartment purge and return the instrument flow to
~500 L/hr
8. Collect the AC spectrum by clicking the Start Step Scan Time Resolved
Measurement button
9. The progress bar at the bottom will keep track of the number of mirror
positions measured. A typical scan will take 60-90 min but that will depend
on the number of mirror positions and the number of laser shots per mirror
position.
10. Depending on how strong the signal is, 4-6 or more spectra will need to be
averaged together to get reasonable signal to noise. Be sure to purge 5-10
minutes between spectra and refill the detector with liquid nitrogen after 6-8
hours (It has an ~8 hour hold time and will automatically shut off to protect
the electronics if the liquid nitrogen runs out). Also monitor the laser power
before collecting each spectra to ensure the power stays constant.
11. The DC spectrum collected in the second channel is a way to monitor the
sample stability. Any changes in in the DC spectrum over time indicate some
change in the sample like decomposition or evaporation.

201

C.5 Data Analysis
The OPUS software will sometimes crash if data analysis and data collection
happen at the same time.

The OPUS software is also installed on the

nanosecond computer, so its safer to transfer files to the nanosecond computer
and analyze the data there if you want to collect and analyze data at the same
time.
1. Open the DC step scan spectrum and right click in the 3D display window to
open the extract spectra dialog box
• In the Select files tab fill in the file name and save location
• In the Extraction range tab select from beginning of file to end of file and
uncheck use block list
• In the Extraction mode tab select Store Average block and Load extracted
files
• Click the extract button
2. Select the averaged DC interferogram in the 2D display window (it should
now have both the negative and positive portion of the signal)
3. Open the Interferogram to spectrum dialog box from the toolbar or the
manipulate menu
• In the Select files tab drag the file icon into the box if it is not already listed
in the box
• In the Store tab the first frequency is 5000 and the last is 0. Select the
store phase box
• In the Apodization tab select the Blackman-Harris 3-term function and 8 as
a zerofilling factor

202

• In the Limit data tab uncheck both boxes and set direction and datapoints
to both
• In the Phase correction tab select Mertz as the phase correction mode
• In the Non Linearity tab uncheck the box for non linearity correction
• In the Peak search tab choose absolute largest value as the peak search
mode. Put 0 as the number of positions to test and choose automatic as
the symmetry of the interferogram
• Calculate the single channel spectrum by clicking the convert button
4. Open the AC step scan spectra in the 3D display window
5. Open the Interferogram to spectrum dialog box from the toolbar or the
manipulate menu
• Keep settings the same as for the DC spectrum EXCEPT:
– Uncheck store phase in Store tab
– Select Mertz/Stored phase option in the Phase correction tab and drag
the phase block of the DC spectrum into the box
– Select Take from Stored phase as the peak search mode in the Peak
search tab
• Click the convert button to calculate the single channel spectrum
6. Use the spectrum calculator to calculate the change in absorbance spectrum
from the AC and DC single channel spectra
• To get the change in transmission: ((AC/50)+DC)/DC
• Use the AB<->TR conversion tool to convert to the change in absorbance
spectrum
7. Repeat for all AC step scan spectra

203

8. In order to improve signal to noise the AC ΔA spectra need to be averaged
together, but in the software only 2D spectra can be averaged.

To get

around this:
• Right click in the 3D display to open the Extract spectra dialog box
• Fill in a file name and path
• Choose the beginning and end of file as the extraction range. Check Use
block list box and fill in the block you want to extract (for 10 ns steps block
# x10 = time). It is best to do one at a time because the software will
sometimes crash or extract the wrong blocks when extracting multiple
blocks.
• Choose series of single blocks as the extraction mode
• Repeat for all desired times and for all AC spectra
9. Use Averaging (from the toolbar or the manipulate menu) to average all of
the AC spectra at the same time delay together.
• Choose the select by symbol option and drag all of the extracted AC
spectra from the first time point into the box and click the average button
• The file will by default be named Av.# and saved in the OPUS work folder.
Use save as to rename and change save location
• Repeat for all time points
10. In order to open the files in another program (Excel, Igor, Origin, etc) they
must be saved as text files. In the Mode tab of the Save as Dialog box
choose Data point table and in the Data point table tab make sure the save
all data points option is selected.

204

C.6 Instrument Shut Down Procedure
1. Turn off the laser with the OPOTEK control software. Click the laser and the
lamp buttons to turn them off and then exit the program. Leave the power
supply on for ~5-10 min to allow the water to cool. Move the beam block in
front of the laser.
2. Turn off the amplifier and disconnect it from the instrument
3. Open the Advanced Measurement box and click on the check signal tab to
switch the instrument back to scan mode (the light on the front of the
instrument should go from red to blinking green) and exit the OPUS program
4. If the instrument will be used again within a few weeks or a month it is better
to leave the globar on, but if the instrument won't be used again soon shut it
off with switch on the power supply and turn off the water circulator by
unplugging it from the power strip
5. Turn the nitrogen pressure down to ~20 psi and close the right and middle
valves.

Open the left valve all the way to keep nitrogen purging in the

instrument
6. Turn off the trigger, scope and power meter and shut down the computer.

C.7 Example Advanced Measurement Settings
1. Basic
• Fill in appropriate sample information
2. Advanced
• Fill in file name and path
• Resolution : 4 or 6

205

• Sample scan time: 30 scans
• Background scan time: 1 scan
• Save data from: Depends on region of interest, but MUST be 0 at ends.
For 5100 nm LP filter use 2600 to 1000. For the 4000 nm LP filter use
2250 to 1000
• Result spectrum: Transmittance
• Data blocks to be saved: Single channel and Sample inteferogram
3. Optic
• Source: MIR
• Beamsplitter: KBr
• Aperture setting: start with 4 mm. Adjust as necessary depending on
sample.
• Measurement channel: Front
• Background meas. channel: Front
• Detector setting: Ex3
• Scanner velocity: 8; 40.0 KHz
• Sample signal gain: 1
• Background signal gain: 1
• Switch gain: ON
• Delay after device change: 0
• Delay before measurement: 0
• Window in points: 350
4. Acquisition
• Wanted high and low frequency limit: use the same as Save data from

206

• Low pass filter: Open
• Acquisition mode: Single sided
• Correlation mode: No
5. FT
• Phase resolution: 8 x Resolution
• Phase correction mode: Mertz
• Apodization function: Blackman-Harris 3-Term
• Zerofilling factor: 8

C.8 Example Step Scan Time Resolved Measurements Settings
1. Basic
• Same as advanced measurement. Most settings should transfer over
2. Recorder setup
• Device: PAD82A (external 10 ns digitizer)
• Time resolution: 10 ns
• Number of timeslices: Depends on the lifetime of the molecule. 200 will
give a 1990 ns window
• Timebase: Linear timescale
• Input range: ± 1V
• Repetition/coadd count: 30
• Trigger mode: Internal for DC and External Positive Edge for AC
• Pre/Post trigger: 0 points
• Experiment recovery time: 0
• Stabilization delay after stepping: 40 ms
• Second Channel: Unused for DC and Save for AC
• Second Channel input range: ± 1V (AC only)

207

3. Advanced
• Fill in file name and path
• Resolution: Same as Advanced Measurement
• Save data from: Same as Advanced Measurement
• Data blocks to be saved: Sample inteferogram
4. Optic
• Source: MIR
• Beamsplittler: KBr
• Aperture: Start with 4mm, adjust as necessary
• Measure channel: Front
• Detector Setting: Ex3
• Scanner Velocity: 8; 40.0 KHz
• Sample signal gain: 1
• Delay after device change: 0
• Delay before measurement: 0
5. Acquisition
• Wanted High and Low frequency: Same as save data from
• Acquisition mode: Single sided
6. FT
• Settings don't matter because FT will be done separate from data
collection

208

Appendix D: Cartesian Coordinates for Optimized Geometries of CN-Mebpy Ruthenium(II) Complexes

D.1 [Ru(bpy)2(CN-Me-bpy)](PF6)2 Optimized Ground State Geometry
N
Ru
N
N
N
N
N
H
C
C
C
C
C
C
H
H
H
H
C
C
C
C
C
C
H
H
H
H
H
H
H
H
C
C
C
C
C

1.09287500
-0.51128700
-2.01409000
-0.65890200
-0.63482600
-2.02639500
1.10046500
-2.53819400
-1.79044500
0.11005500
-1.60747700
-0.99643100
-0.02981700
-4.06509400
0.88526700
2.34678200
0.84244000
2.36911200
-3.42435800
-2.68463700
-2.38529000
-4.10352300
-3.72460400
-3.37671700
0.65854800
2.45552600
-2.36760500
-2.38408200
-4.86117300
-3.64328900
-4.21305200
-1.12996200
3.50862600
1.00691900
2.34524400
3.39898000
2.11874600

1.26210500
0.00061600
-1.42715100
0.90551800
-0.90745800
1.41234300
-1.24774400
3.38615000
2.60085900
0.60296800
1.90031200
2.28377600
1.26728500
-3.33799400
-5.05640200
5.51781200
-0.18643600
-5.49649100
3.10869400
1.59746200
2.16381800
3.29299600
2.52650600
-3.14811500
-0.98371000
-4.99478400
0.99454600
-1.00554300
-4.07794700
-3.74324200
-2.67621900
2.81997500
-1.45190200
-2.54420200
-0.69433500
-2.79270100
-3.37289500

0.41525600
0.00202000
-0.32194800
-1.88557100
1.89131400
0.33677600
-0.42178800
-3.27365800
-3.19533000
-2.96742100
-1.98939900
-4.30949800
-4.19314600
-0.50568200
-1.60684100
0.67420600
-2.84824700
-0.73776300
-0.68225500
1.51435000
-0.76171900
0.53306900
1.65119400
0.70349000
5.02502900
-2.43100800
2.35609500
-2.33641700
-0.57256000
1.57336600
-2.58144300
-5.24781900
-0.41506900
-0.82405200
-0.21547500
-0.82651400
-1.04271700

209

C
C
C
H
H
C
C
C
C
C
H
H
H
C
H
C
C
C
C
C
C
C
C
H
H
H
H
H
N
N
H
H

-2.35091200
4.59006000
1.94288200
0.00679300
-4.22017500
0.01928300
-1.56580400
0.13750100
-0.92809200
-1.72619800
0.85482000
-1.04386300
0.60774100
-3.70931600
4.48674200
3.50027400
0.99219600
2.34108100
3.38318400
2.09959300
-2.68268100
4.57018800
1.91626400
-0.01031400
-4.91017700
4.49248800
-3.70869200
-2.45918800
5.56230000
5.53921700
2.42053200
0.85698900

-2.18841200
-3.57484600
-4.80680800
-2.92848800
2.63908300
-1.26920600
-1.91850300
-0.59624800
-2.30349000
-2.62865300
0.20652500
-2.84721800
0.98876000
-2.56068400
1.04150600
1.47592800
2.56319700
0.71235300
2.82074500
3.39940000
-1.61605100
3.60915000
4.83986600
2.94588700
4.02093400
-1.01542000
3.69610000
-3.42751900
-4.22529400
4.26466700
5.04570500
5.08881700

0.77667500
-1.02514400
-1.47844600
-0.97850900
2.61340600
4.19197300
1.99829900
2.96866100
4.31077000
3.20182000
2.84755700
5.24710800
-5.02986500
-1.62383900
0.26455900
0.41874100
0.80048800
0.21882100
0.81448600
1.01330300
-1.49314500
1.01181900
1.42318500
0.94321200
0.60477900
-0.25044300
-1.55178000
3.28247000
-1.19071900
1.17624000
2.37654200
1.53848900

210

D.2 [Ru(bpy)2(CN-Me-bpy)](PF6)2 Optimized Lowest Energy Triplet State
Geometry
N
Ru
N
N
N
N
N
H
C
C
C
C
C
C
H
H
H
H
C
C
C
C
C
C
H
H
H
H
H
H
H
H
C
C
C
C
C
C
C
C
H

-1.06879600
0.53024400
2.08412800
0.59167000
0.62255900
2.05967700
-1.05257900
2.35225100
1.62551000
-0.23223200
1.51903500
0.78024800
-0.16167700
4.15657800
-0.82469800
-2.39137700
-0.94166400
-2.33158500
3.36678900
2.76988800
2.34516300
4.09561300
3.79153400
3.43192200
-0.77721500
-2.37539800
2.51144800
2.51261600
4.96261800
3.68185200
4.36592300
0.85566600
-3.47548300
-0.95315700
-2.32054400
-3.35667600
-2.04634200
2.39300100
-4.53687100
-1.88217000
0.04979400

-1.23035100
-0.00187900
1.37408800
-1.02924400
1.02419000
-1.40703800
1.25141900
-3.63588200
-2.82944200
-0.77819500
-2.04641800
-2.56638600
-1.52474300
3.23683400
5.08093200
-5.47409000
0.03302600
5.51265500
-3.20367700
-1.50615600
-2.24455000
-3.31089000
-2.45142800
3.14482800
1.30750100
4.99401000
-0.83099800
0.79002500
3.96295300
3.79931600
2.42434000
-3.16731700
1.47532900
2.56561800
0.69316200
2.80120100
3.38562300
2.20511000
3.58293700
4.81842100
2.94749000

0.47965100
-0.00164500
-0.37387200
-1.82587500
1.82163500
0.36201900
-0.47091900
-3.12617400
-3.06287200
-2.87949900
-1.90164300
-4.15441300
-4.06244600
-0.62877900
-1.65730700
0.87424000
-2.77244800
-0.82505200
-0.59103600
1.51837600
-0.69450000
0.60560900
1.67782400
0.57144800
4.89078100
-2.51442900
2.32605200
-2.34063100
-0.72506600
1.40299200
-2.64497200
-5.05966400
-0.46805800
-0.86947300
-0.24385900
-0.89351200
-1.10203200
0.68123800
-1.11392500
-1.54847200
-1.01983700

211

H
C
C
C
C
C
H
H
H
C
H
C
C
C
C
C
C
C
C
H
H
H
H
H
N
N
H
H

4.33194000
-0.10489900
1.57183500
-0.19961500
0.86046900
1.70386000
-0.92870500
0.95551400
-0.83395600
3.82962200
-4.47531300
-3.49409600
-0.98673300
-2.32926700
-3.39275500
-2.09050200
2.78978500
-4.58285200
-1.94541700
0.01083100
4.88772900
-4.46263700
3.59942700
2.44843500
-5.50596700
-5.56018900
-2.45325800
-0.89188600

-2.50332000
1.52986500
2.02117900
0.78864700
2.54998900
2.79754500
-0.00542300
3.14613600
-1.28974600
2.38340300
-0.99177200
-1.42891600
-2.54110900
-0.66034800
-2.75095100
-3.34628400
1.45923300
-3.51732900
-4.77467100
-2.93205700
-4.05276500
1.04718300
-3.86246900
3.58774200
4.23858300
-4.16032700
-4.93337500
-5.04525900

2.62083400
4.06459100
1.89386700
2.87998400
4.15357700
3.05709200
2.77518200
5.06010800
-4.88515600
-1.69908300
0.28167600
0.45977100
0.89335200
0.23967500
0.90120800
1.12799000
-1.53428500
1.12268000
1.59488900
1.05544000
0.69719600
-0.30119600
-1.42418000
3.11762200
-1.29994700
1.30976600
2.55610500
1.72135100

212

D.3 [Ru(bpy)(CN-Me-bpy)2](PF6)2 Optimized Ground State Geometry
N
Ru
N
N
N
N
N
H
C
C
C
C
C
H
H
H
H
C
C
C
C
C
H
H
H
H
H
H
H
C
C
C
C
C
C
C
C
H
C
C
C
C

0.60851200
-0.00317700
-0.41959900
-0.59387600
0.39326700
-2.05126600
2.04458700
-3.51578500
-2.45322400
0.23828200
-1.93768900
-0.19272000
-1.00428000
4.03520100
-0.73905300
1.28487000
5.49376000
-4.14470400
-2.72104400
-2.75240800
-4.11192600
-0.54938600
1.46927000
5.30802700
-2.12270300
-0.96972900
-1.23192600
-0.42790100
-1.50806100
4.14880800
2.70309200
2.75684300
4.81971700
4.09321700
-0.26021700
6.24850300
4.76400100
2.09611900
1.12086000
0.22064400
0.83370600
0.95179500

1.04407900
-0.44699600
-2.06875400
1.05501900
-2.07607000
-0.33862700
-0.36119400
2.60466900
2.36588700
1.75467100
1.35571100
2.78473700
-4.37803400
-2.62807700
4.58635500
1.47911500
-2.65714400
0.67994200
-1.11092800
0.55985300
-1.05392000
-4.48232300
-2.99973200
-1.41342800
-1.79604100
-0.97732600
-5.27567100
-5.46425300
-2.97389000
0.64561000
-1.14267300
0.53461400
-0.17089600
-1.09748300
-3.31502300
-0.06608300
-1.99941800
-1.82611900
-3.12142300
-3.31923200
-1.98706000
-4.39480500

1.33300300
0.00260800
-1.26635200
-1.32464400
1.26822400
0.39627800
-0.39694500
-2.12932600
-2.13733800
-2.14109600
-1.31177600
-2.99322200
-2.75309600
-3.01890600
3.65310400
-2.11402900
-2.00795900
-0.25210600
1.29361600
-0.37732900
1.48303600
-1.42840100
4.34839300
-3.25081600
1.88005800
-2.96604000
-3.32622000
-0.97759700
-4.34767200
0.23317800
-1.29464200
0.36940600
-0.69545100
-1.49225800
-0.69892900
-0.83200600
-2.49841200
-1.87402200
3.32462400
0.69806800
2.55388200
2.75021900

213

C
H
H
C
H
C
C
C
C
C
C
C
C
H
H
H
H
N
N
H
H
C
H
H
H
C
H
H
H
C
N
C
N
C
C

0.49690600
0.95339200
1.16971200
-1.16017300
3.54780800
2.48359400
-0.21154400
1.95440800
1.62518100
0.23382900
-0.86018500
2.16386700
-0.73108500
-1.26017500
4.71411500
-4.70096500
0.36546300
7.42091800
2.59648300
-0.44491400
-1.75094900
-4.79462600
-5.33743600
-5.52729600
-4.07315300
0.78546300
1.80241700
0.51091500
0.79513900
-6.25569900
-7.42813700
-2.10736500
-2.52859300
-4.82720500
-1.58264100

-4.49121600
-0.99438400
-5.29604800
-3.10165400
2.57038400
2.33946600
1.74513700
1.33537300
3.05785100
2.76868000
-1.97210400
4.09226300
3.51627700
1.47637100
1.35115000
1.38749600
-5.47074700
0.01223900
4.94297000
3.42747700
3.13524000
-1.94339300
-1.34764500
-2.59891400
-2.57389900
3.53133900
3.14554000
3.44733300
4.60049100
-0.01132000
0.07609200
4.12194200
4.97743500
-0.12739600
3.08156800

1.42511600
2.97034000
3.32144300
-3.32448200
2.11921500
2.13488900
2.16068900
1.31044500
2.98681900
3.01337400
-2.55134500
3.83017000
3.89952300
2.14165400
0.84058500
-0.86542500
0.97186500
-0.95152400
4.52598800
4.95569300
3.78988800
2.49224900
3.23782700
2.00315000
3.02069000
-3.86570800
-3.74592500
-4.92536400
-3.61548100
0.80465800
0.91753900
-3.82420200
-4.52115200
0.67606500
-2.97935800

214

D.4 [Ru(CN-Me-bpy)3](PF6)2 Optimized Ground State Geometry
N
Ru
N
N
N
N
N
H
C
C
C
C
H
H
H
H
C
C
C
C
C
H
H
H
H
C
C
C
C
C
C
C
C
H
C
C
C
C
H
C
H
C

-1.31324000
0.00279700
1.46240900
-1.64997500
1.73407700
-0.41949900
0.19985200
-3.74888300
-3.32008500
-2.23450700
-2.18388900
-3.37431200
2.88242200
-3.62630700
-1.77660900
1.58325100
-1.84019500
0.29031300
-1.48340600
0.00233800
3.80642700
2.62904300
1.11659800
0.19881700
4.82444400
-0.48187500
1.03288800
-0.55893700
0.38298600
1.17232100
2.74715900
0.46991000
2.11877900
1.60796300
2.96969900
2.90152500
1.78119300
4.13230900
0.83628800
2.22707400
-2.40020000
-2.30835600

1.23583100
0.00032700
-1.06381700
-0.72399900
0.51076800
-1.76080900
1.80359500
-3.39302800
-2.46466200
-0.10734100
-1.90056000
-0.60272800
4.17969700
1.80242300
0.81364800
3.76350000
-3.71017700
-2.23160100
-2.49044100
-3.44303200
-1.63151600
2.48690900
-1.61960400
-1.94118800
-1.53977700
4.11840700
2.01346200
2.84706800
4.32454700
3.25392200
-0.93553800
5.62824400
3.42385300
1.16112400
1.70552600
-0.04021800
1.35278200
0.25909300
1.75884000
-2.60760700
4.45162100
3.43323400

1.05759400
-0.00136100
-1.05999700
-1.06328800
1.05912900
1.04744700
-1.04719600
-0.80967700
-1.18520200
-2.12549900
-0.58705200
-2.78167300
-3.68198500
4.82772400
-2.46294900
-4.80496600
1.16077000
2.10758900
0.56700600
2.75926300
-1.18162700
-4.15124000
2.44685400
-2.46318800
-0.80752800
-1.15474700
-2.10145400
-0.56682600
-2.24500300
-2.74725800
-0.58041100
-2.84806000
-3.90859400
-2.44019700
2.78704300
0.57986300
2.12592200
1.18298000
2.46409300
-2.78449200
0.80181500
1.17741700

215

C
C
C
C
C
C
C
H
H
H
H
N
N
H
H
C
H
H
H
C
H
H
H
C
N
C
N
C
C
C
H
H
H
C
H
H
H
C
N
C
N
C
C

-2.07621700
-1.42189800
-3.09418000
-2.98938600
1.22513600
-4.01041200
-3.81138600
-1.95420300
-1.07841000
-2.67932200
5.06094000
0.54842200
-4.76496100
-3.58160500
-4.88433700
2.95420200
1.93782300
3.37815800
3.56011900
1.89682900
0.83133400
2.46465600
2.15906100
5.43166900
6.46218300
4.63400100
5.51667300
4.16977500
3.54978800
0.83186300
1.62557600
0.20945900
1.29889100
-3.97978800
-3.41605800
-4.00165200
-5.01652400
-1.45852700
-1.74643200
-5.08858100
-6.04930300
-1.09849800
-3.91795300

0.84743700
2.52340700
3.02079100
1.69326900
-1.87493800
3.94998100
1.19767600
-0.17447200
4.94736900
-4.29243100
-0.16913600
6.69424300
4.70545800
0.15481400
1.26889800
2.64903100
2.97635500
2.17044200
3.54093100
-3.49539200
-3.45608800
-3.19658100
-4.53969100
1.44241200
1.70396400
-3.18973300
-3.78596100
1.13117700
-2.46858800
-3.91288100
-3.19947800
-4.04728200
-4.88249700
0.13246800
1.03898700
-0.50360500
0.42256800
-5.44304500
-6.47082400
-2.38915900
-2.85021100
-4.18902500
-1.81716000

2.11474800
0.58253000
2.26906900
2.76712800
-2.12560000
2.87568800
3.93012700
2.45036100
-0.77570900
0.78261300
0.81001100
-3.34985700
3.37989400
4.16844500
3.70749400
3.96308900
4.20176100
4.85574800
3.75524100
-3.95757400
-4.20317300
-4.84852700
-3.74168600
2.90387800
3.41760200
-2.89529000
-3.40486500
2.28636900
-2.28279500
3.92752900
4.16842900
4.82203400
3.70938200
-3.95058400
-4.19084600
-4.84530700
-3.73412700
2.86362800
3.36867400
-2.89282300
-3.40149400
2.25629200
-2.28144700

216