ULTRAFAST SPECTROSCOPIC STUDIES OF MOLECULAR
INTERACTIONS AND VIBRATIONAL ENERGY RELAXATION
DYNAMICS IN BINARY SOLVENTS AND LIPOSOMES
By
Chen Qiu

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemistry – Doctor of Philosophy
2015

ABSTRACT
ULTRAFAST SPECTROSCOPIC STUDIES OF MOLECULAR INTERACTIONS AND
VIBRATIONAL ENERGY RELAXATION DYNAMICS IN BINARY SOLVENTS AND
LIPOSOMES
By
Chen Qiu
Achieving a fundamental understanding of intermolecular interactions and energy
transfer processes in fluid systems is key for studies on chemical reaction mechanisms, material
properties and the biological dynamics. There is a well established body of work describing
transient molecular organization and interactions between dissimilar molecules in neat liquids,
the studies in multi-component liquid remain a challenge due to the transient and complex nature
of solvent-solvent and solute-solvent interactions.
To gain insight into the solution phase heterogeneity and molecular scale organization,
picosecond laser technologies and time resolved spectroscopic approaches have been applied.
We selected the ethanol/cyclohexane and n-butanol/cyclohexane binary solvent mixtures as
model systems. The polycyclic aromatic hydrocarbon (PAH) perylene has been chosen as the
probe molecule to examine local organization in binary solvent systems. The perylene ring
breathing mode is nearly degenerate with the ethanol and n-butanol terminal methyl group
rocking modes. Steady-state spectroscopic data show that there is a discontinuous dependence of
the spectroscopic origin on the binary solvent systems examined. From both orientational and
vibrational energy relaxation dynamics measurements of perylene as a function of solution
composition, we observed molecular scale heterogeneity in both binary solvent systems. For the
ethanol/cyclohexane system, both rotational diffusion and vibrational population relaxation time

constants show a clear discontinuity between 5% and 7.5% (v/v) ethanol, suggesting a
discontinuous change in the organization of the chromophore local environment.

For n-

butanol/cyclohexane system, the rotational diffusion results show that perylene reorients as an
oblate rotor in neat n-butanol and cyclohexane, but as a prolate rotor in all binary mixtures, and
the vibrational population relaxation data show that perylene experiences an n-butanol
dominated environment when the n-butanol concentration is 5% (v/v) or above.

Taken

collectively, both studies demonstrate a non-uniform distribution of alcohol in the binary
mixtures, providing experimental evidence on the existence of composition-dependent nano
meter scale local organization in these systems.
Another question we addressed in this dissertation using picosecond spectroscopy is the
long-term stability of phospholipid vesicles in an aqueous environment, which is an important
issue for studies that use phospholipid vesicles. Unilamellar vesicles containing 1,2-dimyristoylsn-phosphatidylcholine (DMPC), with and without cholesterol, formed by extrusion in aqueous
buffer solution (pH 8) were shown to remain dimensionally stable for periods in excess of
hundreds hours by dynamic light scattering (DLS) measurements. The rotational diffusion
dynamics of perylene confined in the vesicle acyl chain region revealed structural evolution that
was dependent on vesicle composition. We also found that the re-extrusion of the vesicles
caused no change in the average diameter or size distribution, but did give rise to diminished
organization in the lipid acyl chain region for these vesicles. These findings provide new insight
from both macroscopic and microscopic perspectives on the structural stability of phospholipid
vesicles.

To my parents,
for their unconditional love and support…

iv

ACKNOWLEDGMENTS

My time in graduate school has been a real journey filled with enjoyable adventure. At
this point, my sincere thanks and appreciation goes to many people that inspired and helped me
through the past few years.
First and foremost I want to thank my advisor, Dr. Gary J. Blanchard, for his guidance
and encouragement through in my entire research work. I still remember at that time when I was
joining Blanchard group, I barely know anything about laser spectroscopy, and he is the one who
open this window for me and let me see a beautiful world behind it. I really appreciate the time
he spent with me together in the dark laser lab, setting up the “optics LEGO” and playing with
our “laser toys”, which is the best memory I have ever had in graduate school. Besides my
research, the most valuable thing I learn from him is the passion to work and the positive attitude
to no matter what. I am really grateful for everything that I learned from him. Thank you, Gary,
for always being patient, inspiring and supportive, for being a wonderful academic advisor more
than I expected.
I also want to thank my second reader, Dr. Dana Spence, and my committee members,
Dr. Heedeok Hong and Dr. John McCracken. I appreciate all the comment and suggestion you
made to my dissertation work and especially the inspiring discussion during my final defense. In
the meantime, I want to thank all the faculty members that have helped me on my courses,
research, and teaching, Dr. Greg Swain, Dr. Merlin Bruening, Dr. Tom Carter and Dr. Kathy
Severin.
I would like to thank everyone in Blanchard Group, both current and previous members,
Dr. Douglas Gornowich, Dr. Christine Hay, Dr. Iwan Setiawan, Dr. D.J Osborn, Stephen

v

Baumler, Fredy Pratama, Xiaoran Zhang, Hannah Mize, and Ke Ma. I am thankful for all of your
help and kindness during my years in graduate school. Thank you for sharing your brilliant ideas
on both science and life. I will remember the fun time we had together. It is a real pleasure to
work with you guys and I will miss you all.
Next I would like to thank my dear friends, Chao Cheng, Xingyi Yang, Zhe Jia, Yuling
Xie, Jinlan Dong, Chen Zhang, and Yan Zhu. Thank you for being such wonderful friends that
made my life in graduate school so colorful and enjoyable.
Last but definitely not the least, I want to express my deepest thanks and genuine
appreciation to everyone in my family, especially my dear parents, for giving me the opportunity
to explore the world that I want, for being supportive behind me no matter what happened, for
believing in me even I am starting to question myself, for picking me up when I was at my
lowest, for the unconditional love, understanding, and encouragement. I am so blessed and
proud to have such incredible and inspirational parents, love you.

vi

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... ix
LIST OF FIGURES ....................................................................................................................... xi
CHAPTER 1 BACKGROUND AND MOTIVATION .................................................................. 1
1.1 Introduction........................................................................................................................... 1
1.2 Rotational Dynamics Studies in Liquid................................................................................. 4
1.3 Vibrational Energy Relaxation (VER) Studies in Liquid ...................................................... 5
1.3.1 Theoretical studies on VER in liquid ............................................................................. 8
1.3.2 Experimental studies on VER in liquid........................................................................ 10
1.4 Organization of the Dissertation ........................................................................................ 18
REFERENCES ......................................................................................................................... 19
CHAPTER 2 TIME-RESOLVED PUMP-PROBE SPECTROSCOPY ....................................... 26
2.1 Introduction......................................................................................................................... 26
2.2 Instrument Schematic .......................................................................................................... 28
2.2.1 Light Source ................................................................................................................. 30
2.2.2 Modulation ................................................................................................................... 35
2.2.3 Time-delay and Polarization Control .......................................................................... 39
2.2.4 Optical Flow-cell and Sample Pumping ...................................................................... 40
2.2.5 Detection Electronics and Signal Processing .............................................................. 35
2.3 Molecular Reorientation Measurements............................................................................. 42
2.4 Vibrational Relaxation Measurements ............................................................................... 46
2.4.1 Molecular Contribution to the Experimental Signal ................................................... 47
2.4.2 Kinetic Model of the Time-Domain Signal .................................................................. 49
REFERENCES ......................................................................................................................... 52
CHAPTER 3 ORIENTATIONAL AND VIBRATIONAL RELAXATION DYNAMICS OF
PERYLENE IN CYCLOHEXANE-ETHANOL BINARY SOLVENT SYSTEM................. 56
3.1 Introduction......................................................................................................................... 56
3.2 Experimental ....................................................................................................................... 60
3.3 Results and Discussion ....................................................................................................... 62
3.4 Conclusions ......................................................................................................................... 83
REFERENCES ......................................................................................................................... 84
CHAPTER 4 SPECTROSCOPIC STUDIES ON PREFERENTIAL SOLVATION IN THE
CYCLOHEXANE / N-BUTANOL BINARY SOLVENT SYSTEM ..................................... 90
4.1 Introduction......................................................................................................................... 90
4.2 Experimental ....................................................................................................................... 92
4.3 Results and Discussion ....................................................................................................... 95
4.4 Conclusions ....................................................................................................................... 117
REFERENCES ....................................................................................................................... 118
vii

CHAPTER 5 PHOSPHOLIPID VESICLES STABILITY AND TEMPORAL VARIATIONS IN
ACYL CHAIN ORGANIZATION ........................................................................................ 123
5.1 Introduction....................................................................................................................... 123
5.2 Experimental ..................................................................................................................... 125
5.3 Results and Discussion ..................................................................................................... 129
Conclusions ............................................................................................................................. 151
REFERENCES ....................................................................................................................... 152
CHAPTER 6 SUMMARY AND FUTURE WORK .................................................................. 157
6.1 Summary of the Dissertation Work ................................................................................... 157
6.2 Future Directions .............................................................................................................. 161
REFERENCES ....................................................................................................................... 163

viii

LIST OF TABLES

Table 1.1. Selected physical properties of ethanol, n-butanol, and cyclohexane at room
temperature.. ............................................................................................................... 15
Table 2.1. Selected perylene vibrational modes and assignments. ............................................... 34
Table 3.1. Reorientation times and zero-time anisotropies for perylene in cyclohexane-ethanol
mixed solutions. The data are the best fit result of the data to the function of
R(t)=R1(0)exp(-t/1) or R(t)= R1(0)exp(-t/1) + R2(0)exp(-t/2). Times are given in
ps and the uncertainties listed here are standard deviations (±1) based on 6-9
measurements for each sample. .................................................................................. 69
Table 3.2. Cartesian components of the rotational diffusion constant, D, and reorientation time,
OR, for perylene in cyclohexane-ethanol mixed solutions. ........................................ 72
Table 3.3. Vibrational population relaxation times, T1, for perylene in cyclohexane-ethanol
mixed solutions.. ......................................................................................................... 79
Table 4.1. Reorientation times and zero-time anisotropies for perylene in cyclohexane-butanol
mixed solutions ......................................................................................................... 104
Table 4.2. Cartesian components of the rotational diffusion constant, D, and reorientation time,
OR, for perylene in cyclohexane/n-butanol mixed solutions.................................... 105
Table 4.3. Dependence of perylene fluorescence lifetime on n-butanol concentration .............. 108
Table 4.4. Vibrational population relaxation times, T1 in ps, for perylene in cyclohexane-butanol
mixed solutions. Calculated T1 times are based on the r-8 dependence of T1
relaxation, assuming a statistical distribution of solvents. ........................................ 116
Table 5.1. Dynamic light scattering (DLS) data for extruded DMPC and DMPC/cholesterol
vesicles: Measured mean vesicle diameter (d) and distribution width (w) (nm). ..... 135
Table 5.2. Dependence of calculated vesicle diameter on refractive index. Uncertainties in the
diameters are ±1 nm. ................................................................................................ 137
Table 5.3. Reorientation time constant and Cartesian components of the rotational diffusion
constant and aspect ratio Dz/Dx as a function of time for DMPC vesicles. .............. 148
Table 5.4. Reorientation time constant and Cartesian components of the rotational diffusion
constant and aspect ratio Dz/Dx as a function of time for 80% DMPC / 20%
cholesterol vesicles. .................................................................................................. 149

ix

Table 5.5. Reorientation time constant and Cartesian components of the rotational diffusion
constant and aspect ratio Dz/Dx as a function of time for 70% DMPC / 30%
cholesterol vesicles. .................................................................................................. 150

x

LIST OF FIGURES

Figure 1.1. Time scale of selected physical and chemical processes in solution phase. ................ 3
Figure 1.2. Schematic of (a) electronic energy transfer (b) vibrational energy transfer between
two dissimilar molecules. There are two pathways to excite the donor: I. direct
absorption (showing in blue) and II. stimulated Raman scattering (showing in red) ... 7
Figure 1.3. Different pump-probe schemes for the measurement of ground state vibrational
energy relaxation. Pump and probe laser pulses are shown with solid and dashed
lines, respectively. Non-radiative processes are shown using curvy arrows. ............ 12
Figure 1.4. The structure of perylene and axis assignment in the molecular frame. .................... 14
Figure 1.5. Liquid-vapor equilibrium phase diagram of ethanol/cyclohexane binary mixture at
T=293.15K. x represents the liquid mole fraction, and y represents the vapor mole
fraction of ethanol. (Data for the plot is adopted from Dortmund Data Bank Version
2011, URL: http://www.ddbst.com/ddbst.html). ........................................................ 17
Figure 2.1. Schematic of the time resolved pump-probe instrument, showing light source
components (red), detection components (green), and beam paths ............................ 29
Figure 2.2. Schematic of laser light source system. ...................................................................... 31
Figure 2.3. Excitation and emission spectra of perylene in ethanol. ............................................ 32
Figure 2.4. Noise spectrum of dye laser output, showing the flicker noise (low frequency) and
shot noise (high frequency) limited regions................................................................ 36
Figure 2.5. Electro-optic (EO) and mechanical hybrid modulation signal encoding scheme used
in the pump-probe spectroscopy system. .................................................................... 38
Figure 2.6. Two ellipsoid rotor types. Cartesian components of the rotational diffusion constant
(D) upon the assignment of axes and rotor ellipsoid shapes. ...................................... 44
Figure 2.7. Schematic three level system...................................................................................... 48
Figure 3.1. (a) Excitation and emission spectra of perylene in cyclohexane. (b) Excitation and
emission spectra of perylene in ethanol. Intensities have been normalized. ............. 64
Figure 3.2. Dependence of perylene 0-0 transition wavelength on concentration of ethanol in
ethanol-cyclohexane binary solvent system. All spectra have been normalized and the
0-0 transition wavelength results are averaged from at least three measurements. .... 65
Figure 3.3. (a) Experimental time scans I||(t) (blue) and I (t) (red) for perylene in 75 %
ethanol/25 % cyclohexane, along with the instrumental laser cross-correlation
xi

function. (b) Anisotropy decay function, R(t), for the data shown in panel (a). The
decay is fitted to the function R(t)=R(0)exp(-t/OR). Results for theses data are
presented in Table 3.1. ................................................................................................ 68
Figure 3.4. Reorientation time constants as a function of solvent system viscosity. The number
listed above each point indicates the amount of ethanol (%v/v) in the solvent system.
..................................................................................................................................... 70
Figure 3.5. (a) Dependence of perylene Dx and Dz on ethanol concentration in the binary solvent
system. (b) Ethanol concentration dependence of perylene Dz/Dx ratio for the same
solvent system. ............................................................................................................ 73
Figure 3.6. Vibrational population relaxation data (probe polarized at 54.7°with respect to the
pump) and instrument response function for perylene in 99.0 % cyclohexane-1.0 %
ethanol solution. The data and response function are normalized for presentation.
These data are typical of the T1 measurements reported here. .................................... 78
Figure 3.7. Vibrational population relaxation times, T1, for the perylene 1375 cm-1 symmetric
ring breathing mode as a function of solution ethanol content. .................................. 82
Figure 4.1. Excitation and emission spectra of perylene in 5.0 n-butanol / 95.0% cyclohexane
(v/v). Intensities have been normalized. .................................................................... 94
Figure 4.2. (a) Perylene 0-0 transition wavelength and (b) fluorescence Stokes shift as a function
of n-butanol concentration in n-butanol-cyclohexane binary solvent system. ........... 97
Figure 4.3. Absorption and emission band maxima as a function of binary solvent system
composition. ................................................................................................................ 98
Figure 4.4. (a) Experimental time scans I||(t) and I(t)for perylene in 75% n-butanol / 25%
cyclohexane, along with the instrumental laser cross-correlation function. The signal
I||(t) is blue, I(t) is red, and the instrument response function is black. (b)
Anisotropy decay function, R(t)=R(0)exp(-t/OR). The fitted curve is shown in red.
The fitted results are given in Table 4.1. .................................................................. 100
Figure 4.5. (a) Experimental OR values (•) as a function of solvent system viscosity. The
calculated DSE reorientation times as a function of solvent system viscosity are
shown as lines for the stick limit (f = 1) and slip limit (fobl = 0.081, fpro = 0.085). (b)
Frictional interaction coefficients, f, determined from the experimental data as a
function of solvent system composition. The f values are all greater than the
theoretical slip limit values ....................................................................................... 106
Figure 4.6. (a) Fluorescence lifetime data for perylene in 50/50 (%v/v) butanol/cyclohexane. (b)
Perylene fluorescence lifetime as a function of n-butanol concentration (%v/v). .... 107
Figure 4.7. Vibrational population relaxation data (probe polarized at 54.7°with respect to the
pump) and instrument response function for perylene in 97.5% cyclohexane / 2.5% nxii

butanol solution. The data and response function are normalized for presentation.
These data are typical of the T1 measurements reported here. ................................. 114
Figure 4.8. Vibrational population relaxation times, T1, for the perylene 1375 cm-1 symmetric
ring breathing mode as a function of solution n-butanol content (%v/v). ................ 115
Figure 5.1. Time-Correlated Single Photon Counting (TCSPC) instrument schematic ............. 128
Figure 5.2. Structures of 14:0 DMPC, perylene and cholesterol. ............................................... 131
Figure 5.3. (a) Measured mean size of vesicles from DLS with time. Inset: DLS size
distribution data for DMPC vesicles as a function of time after formation. (b)
Measured distribution width from DLS of vesicles with time. For both plots, solid
circles represent 100% DMPC vesicles, open circles represent 80% DMPC / 20%
cholesterol vesicles, and solid triangles represent 70% DMPC / 30% cholesterol
vesicles. ..................................................................................................................... 134
Figure 5.4. (a) Experimental size distributions for nominal 100 nm diameter vesicles comprised
of DMPC (solid line), 80% DMPC / 20% cholesterol (dashed line) and 70% DMPC /
30% cholesterol (dotted line), all acquired within one hour after extrusion to form the
vesicles. (b) Experimental correlation functions associated with the same three
samples. ..................................................................................................................... 136
Figure 5.5. Anisotropy decay data for perylene in 100-nm diameter DMPC vesicles at 293 K,
showing decay that is best fit by two exponential components. The fitted line is
indicated as the solid line and the residuals of the fit are shown centered around the
zero line. The fitted function is R(t) = R1(0)exp(-t/1) + R2(0)exp(-t/2), where R1(0)
= 0.08 ±0.01, 1= 421 ± 45 ps, R2(0) = 0.20 ±0.01, 2= 2358 ±139 ps... ................ 141
Figure 5.6. (a) Dependence of perylene Dx and Dz on time elapsed after formation for nominal
100 nm diameter DMPC vesicles at 293 K. (b) Dependence of Dz/Dx ratio for these
vesicles. Vesicles were re-extruded through 100 nm diameter pore filter at 370 hours.
................................................................................................................................... 142
Figure 5.7. (a) Dependence of perylene Dx and Dz on time elapsed after formation for nominal
100 nm diameter 80% DMPC / 20% cholesterol vesicles at 293 K. (b) Dependence of
Dz/Dx ratio for these vesicles. Vesicles were re-extruded through 100 nm diameter
pore filter at 250 hours. ............................................................................................. 146
Figure 5.8. (a) Dependence of perylene Dx and Dz on time elapsed after formation for nominal
100 nm diameter 70% DMPC / 30% cholesterol vesicles at 293 K. (b) Dependence of
Dz/Dx ratio for these vesicles. Vesicles were re-extruded through 100 nm diameter
pore filter at 250 hours. ............................................................................................. 147

xiii

CHAPTER 1
BACKGROUND AND MOTIVATION

1.1 Introduction
Despite the fact that chemistry is a discipline that encompasses a broad range of subjects,
which continues to expand, a central part of much work in chemistry lies in some aspect of
understanding and controlling chemical reactions or energy transfer processes. The key to
developing a molecular-scale understanding of chemical reaction mechanisms and material
properties is elucidating the interactions between dissimilar molecules, especially with respect to
energy exchange.1,2
Intermolecular interactions occur in all phases, and it is the strength of these interactions
that determine the phase. In each phase, a different timescale is relevant due to the distance and
orientational freedom that exists between molecules. In the gas phase, molecules are relatively
distant from one another, and essentially most intermolecular interactions take place through
binary collisions, with the associated relaxation process being described in terms of state-to-state
collision cross sections. In the solid state, molecules exist with fixed locations and spatial
relationships, where energy relaxation involves the scattering of (harmonic) lattice phonons from
localized molecular vibrational states. In both cases, molecular elastic and inelastic interactions
can be treated separately. The inelastic process requires that the total amount of energy in the
system doesn't change due to the collision, but a transfer of energy has taken place; the elastic
process does not involve the transfer of energy between molecules but does shift the phase of
(vibrational) wave function as a result of the collision. Several measurement techniques can be
used to gain detailed information on structure and intermolecular interaction in the gas and solid
1

states, including mass spectrometry,3 X-ray diffraction and multi-dimensional NMR
spectroscopy.4
The liquid phase however is different from the gas or solid phases. Liquids are more
complicated because of its characteristically strong intermolecular interactions over relatively
long distances, as well as the short persistence time of any molecular organization within the
system. Rapidly changing interactions characterized by a wide range of spatial relationships
between interacting species pose a substantial experimental challenge.

The long-range

interactions in the liquid state refer to both the electronic-electronic and vibrational-vibrational
energy transfer over nanometer length scales in solution. The characteristic time scale for
intermolecular interactions in the liquid phase ranges from several femtoseconds (collisions) to
hundreds of picoseconds (resonant energy transfer). The time scales of molecular rotational and
vibrational motion in solution are shown in Figure 1.1, with comparison to other chemical and
physical processes that are commonly studied.5

Despite the experimental complexity, the

investigations of solute-solvent molecular interactions and the local organization of the solvent
are extremely important because the solution phase is the medium used for the majority of
chemical syntheses and analyses.

Gaining insight into local organization and energy flow

between dissimilar molecules in the liquid phase, and then relating this knowledge to the
macroscopic properties of the liquid medium attracts more and more attention but remains a
challenge.

2

Figure 1.1. Time scale of selected physical and chemical processes in solution phase.

For dilute solutions, where the solute is present at low concentration, the solvent
surrounding the solute organize in some manner to minimize the energy of interaction. The
number of solvent configurations within kBT of the energetic minimum may be very large and
rapid exchange between these configurations is facile.

The examination of solvent-solute

interactions will thus elucidate trends or broad patterns rather than specific, fixed structures. To
examine those solvent-solute interactions, the most widely used experimental approach is to
interrogate a probe molecule in some manner, potentially disrupting the solvent organization in
the process, and then monitor changes in probe molecule response.

Molecular rotational

dynamics6,7 and energy relaxation8-13 are two phenomena used widely in the examination of

3

solvent-solute interactions, which both can be quantitated using a single experimental apparatus
in the Blanchard labs. A detailed description of this instrument is presented in Chapter 2.

1.2 Rotational Dynamics Studies in Liquids
In the solution phase molecules execute diffusional motion that can be either translational
or rotational in nature.

When the solute molecule is a chromophore that can be excited

selectively, its rotational motion can be characterized. The rotational motion of the solute is
confined by its solvent environment, with the details of the confinement depending on the
intermolecular interactions between solvent molecules as well as between solvent and solute.
The rate of solute rotation about this Cartesian axes defined by its excited and emitting transition
dipole moments will differ according to the confinement imposed by the solvent local
environment. One method used to characterize solute rotational motion is to excite a subset of
the chromophores in solution with a pulse of polarized light, which photoselects an
orientationally non-random subset of the total chromophore population. The process by which
this photoselected population evolves to a random orientational distribution is monitored through
its time-dependent orientational relaxation. These measurements provide insight into how solute
molecular motion is affected by inter-molecular interactions with surrounding solvent molecules.
The effects of these interactions on chromophore rotational dynamics have been classified into
three broad categories. These are 1) short-range repulsive forces, which dominate molecular
collisional processes,14 2) long-range electrostatic interactions between charged or dipolar
solutes and polar solvents,15-17 and 3) site-specific solvent-solute interactions, such as hydrogen
bonding.18-20 To understand solute rotational dynamics, the binary collision approximation21-23
and Debye-Stokes-Einstein hydrodynamic theory24-27 are applied in most cases.
4

Binary solvent mixtures are used commonly to modify solvent system properties for use
in chemical reactions and separations. The properties of binary solvent systems typically do not
vary smoothly with composition. Solvent system properties and the molecular-scale nature of
solvent-solvent and solvent-solute interactions typically reflect nanoscale heterogeneity in binary
systems, and the details of heterogeneity on such short length scales in highly dynamic systems
remain to be understood. The properties of some binary solvent systems have been studied
recently, using both theoretical calculations and anisotropy decay measurements to demonstrate
unexpected features.

28,29

Studies on water/ionic liquid and water/alcohol binary systems found

composition dependent excess viscosity, which resulted from specific intermolecular interactions
associated with local structural heterogeneity.30-33

The rotational diffusion dynamics of

coumarin 153 in water/1-propanol solutions showed that the orientational relaxation and
solvation time constants depend sensitively on solvent 1- propanol content at alcohol low mole
fractions, and this behavior was explained in terms of preferential solvation of the chromophore
due to the formation of 1-propanol clusters.34 The rotational diffusion dynamics data reported in
Chapters 3 and 4 focus on alcohol/cyclohexane binary solvents, which also exhibit molecular
scale compositional heterogeneity.

1.3 Vibrational Energy Relaxation (VER) Studies in Liquids
Vibrational energy relaxation (VER) is an important phenomenon to understand because
it is the dominant mechanism by which thermal energy dissipates in matter. The dissipation of
energy determines bulk material properties such as thermal conductivity, for example.

In

addition, the study of VER can be used to investigate local organization in condensed phases.
Getting a comprehensive understanding of vibrational energy transfer process, including VER

5

pathways, rates, and detailed mechanisms over wide ranges of system types and chemical
conditions is a sought-after goal because such information would allow better predictability and
control over chemical reactions35-37 and dynamics in biological systems.11,38,39
To examine the transfer of energy between dissimilar molecules in the liquid phase, a
simplified electronic and vibrational energy transfer donor-acceptor model is used, shown in
Figure 1.2. When a ground state donor molecule is excited to either an electronic or vibrational
excited state, the energy transfer process is initiated. The excess energy in the system is
dissipated non-radiatively and transferred to the surrounding bath. The details of the transfer
process depend on a number of factors. The efficiency of this process is enhanced if the bath
contains a (vibrational) resonance that is degenerate with the excited vibrational resonance of the
chromophore. Such through-space relaxation processes do not require molecular collisions and
their efficiency depends on the extent to which the donor and acceptor (vibrational) resonances
are degenerate as well as the orientation of the donor with respect to the acceptor.

6

S1

S1
(a)

S0

S0
donor

acceptor

S1
S1
(b)
S0

I. II.

S0
acceptor

donor

Figure 1.2. Schematic of (a) electronic energy transfer (b) vibrational energy transfer
between two dissimilar molecules. There are two pathways to excite the donor: I. direct
absorption (showing in blue) and II. stimulated Raman scattering (showing in red).

The investigation of energy transfer has been active since the first observation in the
1920s by Franck and Cario.40 Despite the long history of the field, there remain open questions
regarding the efficiency and dominant mechanism(s) of energy transfer in any given molecular
system. Understanding VER in the liquid phase is a challenge because of the transient nature of
any organization that exists in liquids and the potential for competition between collisional and
7

resonant non-collisional energy transfer processes. Theoretical investigations and computational
modeling have been the key to understanding experimental measurements. A number of models
have been developed to describe VER in condensed phases.41,42,43,44 From an experimental
standpoint, beyond the classic methods like ultrasonic attenuation45-47 and infrared
spectroscopy,48 time-resolved49-51 and multi-dimensional spectroscopies52,53 have provided
significant insight into energy relaxation behavior in liquid systems.

1.3.1 Theoretical studies on VER in liquid
As described above, the liquid phase is too dense to treat in the context of a binary
collision model (gas), and too disordered for a harmonic phonon model (solid). For this reason
numerous theoretical models have been introduced to describe the VER processes in the liquid
phase. The most widely used classical model for VER in liquids is the isolated binary collision
(IBC) model, introduced in the 1920s, which was proposed originally for gas phase studies.54 In
this model the molecular interactions in the liquid system are simplified by assuming that the
collision which determines the vibrational relaxation time (T1) can be approximated by isolated
binary encounters and thus treated in the same manner as gas phase scattering theory.54-56
According to this model, the VER rate is determined by the product of the probability of energy
transfer per collision and the number of collision events per second in the liquid. The former
quantity is estimated from low-density gas phase scattering theory calculations.

There are

several criticisms of this approach, all based on the limitations of IBC theory to address the
inherent complexity of the liquid phase. Fixman57 has pointed out that, due to the interactions of
the colliding pair with surrounding molecules, the energy transfer probability should be density
dependent. Work by Herzfeld56 has suggested that the use of the hard-sphere assumption in IBC

8

can greatly overestimate the rate of VER in liquid. It is clear that more complex treatments of
this problem are required.
Modern theoretical treatments of VER in the liquid phase can be divided into two broad
categories; 1) the simulation of liquid phase systems, and 2) descriptive explanations of VER in
liquid.9 Molecular dynamics (MD) simulation is the main approach by which liquid phase
systems have been modeled. This method is based on simulating the physical movements of
atoms and molecules in a dynamic system under the influence of different types of physical
forces. The limitation of this approach is the complexity associated with the examination of
large molecules in liquids due to their structure and molecular motions, including rotations,
translations, and vibrations, in combination with the large number of intermolecular interactions
that contribute to the system dynamics.13,58
To fulfill the dynamic picture of VER in condensed phase and further explain these
processes, several theories have been developed.41 Instantaneous-Normal-Mode (INM) theory42
suggests that during the initial stages of VER (hundreds of femtoseconds), there is not enough
time for organizational change in the solvent, so the solute and solvent molecules can be treated
as two fixed molecules, and each degree of freedom can be described using normal modes. INM
theory provides specific dynamical information on the VER process. For example, Garberoglio
and Vallauri applied INM theory to liquid methanol system to illustrate the presence of linear
hydrogen bond network. They also found that due to the strong coupling of the chain motion to
other diffusive processes in the liquid, the relationship between the chain dynamics and INM
mode is less evident in methanol than hydrogen fluoride.44 Based upon INM theory, Stratt and
coworkers further proposed an instantaneous pair (IP) theory,43 which suggests the VER in
liquids is controlled primarily by the solute/solvent nearest pair. For small solute molecules

9

during a short time after initial excitation, the major contribution to VER should be from the
closest atomic pairs; for large solutes, although there many surrounding solvent molecules, the
strongest interaction is also from the few atoms in closest proximity solvent molecule moving
along the direction of the specific solute vibrational mode, which makes the largest contribution
to VER. IP theory helps visualize the molecular scale dynamics of VER, and provides the
opportunity for direct comparison with experimental results.
Theoretical studies and computational simulations of VER are important in understanding
fast relaxation processes in the liquid phase. Each approach, however, has its own limitations.
For example, molecular dynamics (MC) simulation is suitable only for one-component liquid
solvent, and INM and IP theories can only describe equilibrium processes. At the present time,
the interpretation of experimental data on complex systems remains limited. More sophisticated
theories that will be able to provide insight into compositionally and structurally complex
systems are needed. Despite that limitation, experimental VER data can provide useful insight
into short range, transient organization in the liquid phase.

1.3.2 Experimental studies on VER in liquids
The information on energy dissipation both within the probe molecule and between the
probe molecule and the surrounding medium can be gathered from a variety of energy relaxation
measurements, including excitation transport, transient spectral shift, and vibrational population
relaxation. State-of-art spectroscopic techniques are elucidating the dynamics and mechanisms
that mediate VER.
As mentioned previously, the early stage experimental studies of VER in condensed
phases were primarily frequency-domain experiments, such as ultrasonic attenuation
measurements and infrared absorption, which provided only the overall (relatively slow)
10

relaxation time for certain vibrational modes in small molecules.47 The development of modern
short pulse spectroscopies has revolutionized time-domain VER measurements. One of the most
powerful spectroscopic methods for VER studies in the liquid phase is the pump-probe
technique.59 In this type of measurement the probe solute molecule is excited by an ultrafast
pump laser, and then the transfer of energy from the vibrationally excited probe to its
surrounding bath returns the probe to its ground state, and the temporal evolution of some aspect
of this process is interrogated by a second (probe) laser pulse. By controlling the time delay
between pump and probe laser pulses, the characteristic time constant(s) of VER can be obtained.
In Figure 1.3, four different types of pump-probe schemes are summarized. The probe molecule
in ground state could be excited directly by absorption of resonant infrared photons, as in (a)60
and (d),12 or by stimulated Raman scattering, as in (b)61 A special case is shown in (c)62, the
indirect depopulation is generated by exciting the molecule to a short-lived electronic state,
followed by rapid internal conversion to the ground state. After the excitation process, the
kinetics of population redistribution within the ground state are monitored by the probe laser, via
the bleaching of a vibrational absorption (a), Anti-Stokes Raman scattering (b), or double-step
transitions to the S1 state (c) and (d). The specific pump-probe measurement scheme and
instrumentation used in this work differs from those illustrated in Fig. 1.3 and is described in
detail in Chapter 2.

11

pump

probe

(a)
pump

(b)

probe

IFL
pump

probe

probe (c)

(d)
pump

Figure 1.3. Different pump-probe schemes for the measurement of ground state vibrational
energy relaxation. Pump and probe laser pulses are shown with solid and dashed lines,
respectively. Non-radiative processes are shown using curvy arrows.

In most experimental studies, the VER rate is the primary information of interest and it is
this quantity that is related to molecular scale organization. Numerous systematic studies have
been conducted to understand vibrational energy relaxation in liquid systems.18,35,50,51,63 Among
the experimental studies, the most commonly used approach is to examine trends in the
relaxation behavior of a single probe molecule in a series of different solvent systems, or to
examine several structurally similar or related molecules in a single environment. The systems

12

studied include neat liquids and binary solvent systems, along with some more complicated,
heterogeneous fluid systems.
In studies of neat liquids, a series of related solvent molecules are chosen in order to
obtain the dependence of vibrational relaxation on some particular structural aspect of the bath.
For example, previous work from the Blanchard group focused on the vibrational energy
relaxation of the perylene and methylperylene ring breathing modes in normal alkanols18 and
alkanes51,63 ranging from C1 to C10, as well as selected aliphatic aldehydes and ketones,50 to
probe local organization. These studies showed that solvent-dependent reorientation behavior
was in many instances correlated with solvent-dependent vibrational relaxation dynamics.
Applying a similar experimental approach to binary solvent mixtures, Hochstrasser and
coworkers64 measured the vibrational population relaxation of a cyanide ion in H2O and D2O
using an IR pump-probe experiment. In that work the isotopic composition of the ion was used
to control the oscillation frequency of the CN- stretch.

They observed a decrease in the

relaxation time with increasing CN- vibrational frequency and found a significant correlation
between VER and IR absorption cross section, providing experimental evidence for the primary
contribution to VER in this system.
The choice of probe molecule can play an important role in determining the information
content of a series of measurements. The probe molecule used in this dissertation is perylene, a
polycyclic aromatic hydrocarbon of D2h point group symmetry. This chromophore possesses a
center of inversion, and its molecular structure and axis assignments are shown in Figure 1.4.
Due to the existence of the inversion center, perylene does not have a permanent dipole moment,
the 1375 cm-1 ring breathing mode modulates the molecular quadrupole moment. In the work
presented here, we examine the binary solvent systems cyclohexane/ethanol (Chapter 3) and

13

cyclohexane/n-butanol (Chapter 4). Previous studies have shown that there is efficient coupling
between the perylene 1375 cm-1 ring breathing mode and the terminal methyl group rocking
mode (ca. 1370 cm-1) of n-alcohols and other species containing aliphatic chains.65,66 The IRactive terminal methyl group rocking mode of alkanols and other systems couples to the perylene
ring breathing mode by means of quadrupole-dipole coupling, which exhibits a distance
dependence of with r-8.67

The steady-state spectra and detailed vibrational mode analysis of

perylene is shown in Chapter 3 and Chapter 4.

z
x

y
Figure 1.4. The structure of perylene and axis assignment in the molecular frame.

The long term goal of this project is to gain a fundamental understanding on the
intermolecular interaction in complex fluid systems. Multi-component solvent becomes the main
focus in the early stage of this study. Therefore, choosing solvent system is crucial. Binary
solvent systems have been chosen because they serve as the most simplified model for multicomponent solvent with the potential of providing useful information of the complex solventsolute and solvent-solvent interactions. Previous studies in our group had revealed the solventsolute interactions in a variety of alkanols by incorporating the fluorescence probe molecular and
14

using time-resolved spectroscopic methods, which is the starting point of this study.

We

introduced the second solvent element into the system and formed two model binary solvent
systems: cyclohexane/ethanol and cyclohexane/n-butanol.
Both cyclohexane/ethanol and cyclohexane/n-butanol binary systems have been studied
extensively in terms of bulk physical properties from a thermodynamic standpoint.68-71 The
major physical properties72 (at room temperature) of each solvent are summarized in Table 1.1.
Despite these differences in physical properties, cyclohexane and ethanol or n-butanol are
miscible over a wide concentration range,73 with bulk properties, such as viscosity, varying
monotonically with the solvent composition.71
Table 1.1. Selected physical properties of ethanol, n-butanol, and cyclohexane at room
temperature.
Ethanol

n-Butanol

Cyclohexane

Molar mass (g/mol)

46.07

74.12

84.16

Viscosity (cp)

1.095

2.544

1.020

Boiling point (ºC)

78.4

117.4

80.7

Dipole moment (D)

1.69

1.66

0

Binary mixtures formed from two volatile liquids exhibit a range of boiling behavior,
from ideal, with a simple continuous change in boiling point with composition, to non-ideal,
showing the presence of an azeotrope, characterized by either a maximum or minimum boiling
point.

One of the most interesting facts about both ethanol/cyclohexane and n15

butanol/cyclohexane binary solvent mixtures is that they both form azeotropes.68 For a given
ideal binary solvent mixture, the composition of the vapor and liquid are governed by Raoult’s
and Dalton’s Law, with the partial pressure of each component in the vapor phase being
proportional to the mole fraction of the same component in the liquid. For non-ideal mixtures
such as ethanol/cyclohexane, this relationship is not linear, and the liquid-vapor equilibrium
phase diagram is shown in Figure 1.5. In conceptual terms, a liquid with a higher vapor pressure
has a lower boiling point and, as a result, for both binary solvent systems examined,
ethanol/cyclohexane and n-butanol/cyclohexane, a minimum boiling point azeotrope is formed.
Taking ethanol/cyclohexane as an example, the minimum boiling point, 64.8ºC, occurs when the
mole fraction of ethanol reaches x=0.43. In these azeoptrope mixtures, the tendency for the
solvent molecules to escape from the mixture is different from pure liquids. The intermolecular
interactions between ethanol and cyclohexane molecules are weaker than they are in the pure
liquids, and the enthalpy change of this mixing process is endothermic.

16

14

P[KPa]

12

10

8

6
0.0

0.2

0.4

0.6

0.8

1.0

x,y (Ethanol) [mol/mol]

Figure 1.5. Liquid-vapor equilibrium phase diagram of ethanol/cyclohexane binary
mixture at T=293.15K. x represents the liquid mole fraction, and y represents the vapor
mole fraction of ethanol. (Data for the plot is adopted from Dortmund Data Bank Version
2011, URL: http://www.ddbst.com/ddbst.html)

The formation of an azeotrope plays an essential role in the industrial distillation
processes, however, the detailed molecular level picture for these systems are still not clear.74-77
By using ultrafast laser spectroscopy, we are able to examine these non-ideal binary solvent
mixtures from a molecular point of view, which can possibly provide information on the
molecular level organization and interactions and thus a better understanding of the factors
responsible for non-ideal binary solvent behavior.

17

1.4 Organization of the Dissertation
The work presented in this dissertation has been focused on understanding intermolecular
interactions in heterogeneous fluid environments, in terms of molecular-scale organization and
vibrational energy relaxation behavior.
Chapter 2 of this dissertation describes a newly developed time-resolved pump-probe
spectroscopic instrument used in this work, as well as a brief discussion of the principles of
anisotropy decay and vibrational population relaxation measurements as they apply to the
stimulated emission based method we use.

In Chapters 3 and 4 we present the rotational

diffusion dynamics and vibrational energy relaxation results for the nonpolar fluorescence probe
molecule perylene in two binary solvent mixtures systems; ethanol/cyclohexane and nbutanol/cyclohexane. By varying the composition of the binary solvent systems we were able to
observe system-specific trends in local organization and molecular-scale compositional
heterogeneity. In addition to the study of binary solvent systems, we are also interested in
understanding the long term persistence of structural organization in more complex solution
phase systems. One such system is vesicles formed by lipids in aqueous medium. Chapter 5 of
this dissertation deals with the long-term stability and temporal variations in the acyl chain
region of phospholipid vesicles, providing evidence of the dimensional and morphological
stability of unilamellar 1,2-dimyristoyl-sn-phosphatidylcholine (DMPC) vesicles. The broad
conclusions of this work, as well as the future directions of this work are summarized in Chapter
6.

All the information gathered from this work is for the purpose of increasing our

understanding of molecular interactions and vibrational energy transfer processes in complex
chemical environments. Such an understanding is a requirement for further work aimed at
predicting and tailoring energy flow in complex liquid phase systems.

18

REFERENCES

19

REFERENCES

(1)

Harris, C. B.; Smith, D. E.; Russell, D. J. Chemical Reviews 1990, 90, 481.

(2)

Secrest, D. Annual Review of Physical Chemistry 1973, 24, 379.

(3)
May, J. C.; Goodwin, C. R.; Lareau, N. M.; Leaptrot, K. L.; Morris, C. B.; Kurulugama,
R. T.; Mordehai, A.; Klein, C.; Barry, W.; Darland, E.; Overney, G.; Imatani, K.; Stafford, G. C.;
Fjeldsted, J. C.; McLean, J. A. Analytical Chemistry 2014, 86, 2107.
(4)

Kogler, G.; Mirau, P. A. Macromolecules 1992, 25, 598.

(5)

Zewail, A. H. The Journal of Physical Chemistry A 2000, 104, 5660.

(6)

Savitzky, B. H.; Stratt, R. M. The Journal of Physical Chemistry B 2008, 112, 13326.

(7)
Khara, D. C.; Kumar, J. P.; Mondal, N.; Samanta, A. The Journal of Physical Chemistry
B 2013, 117, 5156.
(8)

Benjamin, I. The Journal of Physical Chemistry B 2006, 110, 9375.

(9)

Egorov, S. A.; Skinner, J. L. The Journal of Chemical Physics 1996, 105, 7047.

(10)

Fujisaki, H.; Stock, G. The Journal of Chemical Physics 2008, 129.

(11) Fujisaki, H.; Straub, J. E. Proceedings of the National Academy of Sciences of the United
States of America 2005, 102, 6726.
(12)

Laubereau, A.; Seilmeier, A.; Kaiser, W. Chemical Physics Letters 1975, 36, 232.

(13)

Skinner, J. L. Theor Chem Acc 2011, 128, 147.

(14)

Ratner, M. A. International Journal of Quantum Chemistry 1987, 31, 989.

20

(15)

Barbara, P.; Jarzeba, W. In Advances in Photochemistry; John Wiley & Sons, Inc.: 2007.

(16)

Maroncelli, M. Journal of Molecular Liquids 1993, 57, 1.

(17)

Van der Zwan, G.; Hynes, J. T. The Journal of Physical Chemistry 1985, 89, 4181.

(18)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.

(19)

Dutt, G. B.; Ghanty, T. K. The Journal of Chemical Physics 2003, 118, 4127.

(20) Greiner, A. J.; Pillman, H. A.; Worden, R. M.; Blanchard, G. J.; Ofoli, R. Y. The Journal
of Physical Chemistry B 2009, 113, 13263.
(21)

Widom, B. The Journal of Chemical Physics 1960, 32, 913.

(22)

Chandler, D. The Journal of Chemical Physics 1974, 60, 3500.

(23)

Evans, G. T. The Journal of Chemical Physics 1988, 88, 5035.

(24)

Journal of the Society of Chemical Industry 1929, 48, 1036.

(25)

Zwanzig, R.; Harrison, A. K. The Journal of Chemical Physics 1985, 83, 5861.

(26)

Chuang, T. J.; Eisenthal, K. B. The Journal of Chemical Physics 1972, 57, 5094.

(27)

Youngren, G. K.; Acrivos, A. The Journal of Chemical Physics 1975, 63, 3846.

(28)

Dutt, G. B. The Journal of Chemical Physics 2000, 113, 11154.

(29)

Dutt, G. B.; Doraiswamy, S. The Journal of Chemical Physics 1992, 96, 2475.

(30)

Rodrí
guez, H.; Brennecke, J. F. Journal of Chemical & Engineering Data 2006, 51, 2145.

(31)

Mukherjee, A.; Bagchi, B. The Journal of Physical Chemistry B 2001, 105, 9581.
21

(32)

Mukherjee, A.; Srinivas, G.; Bhattacharyya, S.; Bagchi, B. J Chem Sci 2001, 113, 393.

(33)

Stevenson, S. A.; Blanchard, G. J. The Journal of Physical Chemistry A 2006, 110, 3426.

(34)

Shirota, H.; Castner, E. W. The Journal of Chemical Physics 2000, 112, 2367.

(35)

Elsaesser, T.; Kaiser, W. Annual Review of Physical Chemistry 1991, 42, 83.

(36) Suhling, K.; French, P. M. W.; Phillips, D. Photochemical & Photobiological Sciences
2005, 4, 13.
(37) Tesar, S. L.; Kasyanenko, V. M.; Rubtsov, I. V.; Rubtsov, G. I.; Burin, A. L. The Journal
of Physical Chemistry A 2012, 117, 315.
(38)

Miller, R. J. D. Annual Review of Physical Chemistry 1991, 42, 581.

(39)

Bu, L.; Straub, J. E. The Journal of Physical Chemistry B 2003, 107, 12339.

(40)

Cario, G.; Franck, J. Z. Physik 1923, 17, 202.

(41)

Sun, X.; Stratt, R. M. The Journal of Chemical Physics 2013, 139.

(42)

Keyes, T. The Journal of Physical Chemistry A 1997, 101, 2921.

(43)

Richard, M. S. In Ultrafast Infrared And Raman Spectroscopy; CRC Press: 2001.

(44)

Garberoglio, G.; Vallauri, R. The Journal of Chemical Physics 2001, 115, 395.

(45)

Shiwa, Y.; Kawasaki, K. Progress of Theoretical Physics 1981, 66, 406.

(46)

Onuki, A. Physical Review E 1997, 55, 403.

(47)

Madej, D.; Hornowski, T. Journal of Physics: Condensed Matter 2002, 14, 13429.

22

(48) Chang, H. C.; Lin, J. C.; Wu, J. Y.; Chen, K. H. The Journal of Physical Chemistry 1995,
99, 11081.
(49) Nienhuys, H.-K.; Woutersen, S.; van Santen, R. A.; Bakker, H. J. The Journal of
Chemical Physics 1999, 111, 1494.
(50)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 6785.

(51)

Jiang, Y.; Blanchard, G. J. The Journal of Physical Chemistry 1994, 98, 6436.

(52) King, J. T.; Ross, M. R.; Kubarych, K. J. The Journal of Physical Chemistry B 2012, 116,
3754.
(53)

Loparo, J. J.; Roberts, S. T.; Tokmakoff, A. The Journal of Chemical Physics 2006, 125.

(54)

Chandler, D. W.; Ewing, G. E. The Journal of Physical Chemistry 1981, 85, 1994.

(55)

Krueger, H.; Weitz, E. Israel Journal of Chemistry 1989, 29, 487.

(56)

Herzfeld, K. F. The Journal of Chemical Physics 1962, 36, 3305.

(57)

Fixman, M. The Journal of Chemical Physics 1961, 34, 369.

(58) Jas, G. S.; Larson, E. J.; Johnson, C. K.; Kuczera, K. The Journal of Physical Chemistry
A 2000, 104, 9841.
(59)

Bagchi, B. Molecular Relaxation in Liquids; Oxford University Press, USA, 2012.

(60)

Graener, H.; Ye, T. Q.; Laubereau, A. The Journal of Chemical Physics 1989, 91, 1043.

(61)

Laubereau, A.; Kaiser, W. Reviews of Modern Physics 1978, 50, 607.

(62) Wild, W.; Seilmeier, A.; Gottfried, N. H.; Kaiser, W. Chemical Physics Letters 1985, 119,
259.

23

(63)

Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 9411.

(64) Kim, Y. S.; Hochstrasser, R. M. Proceedings of the National Academy of Sciences of the
United States of America 2005, 102, 11185.
(65)

Pillman, H. A.; Blanchard, G. J. The Journal of Physical Chemistry B 2011, 115, 3819.

(66)

Pillman, H. A.; Blanchard, G. J. The Journal of Physical Chemistry B 2010, 114, 13703.

(67) Gray, C. G.; Gubbins, K. E. Theort of Molecular Fluids, Vol. 1. Fundamentals; Oxford
Science, 1984.
(68)

Scatchard, G.; Satkiewicz, F. G. Journal of the American Chemical Society 1964, 86, 130.

(69) Bich, E.; Hensen, U.; Michalik, M.; Wandschneider, D.; Heintz, A. Physical Chemistry
Chemical Physics 2002, 4, 5827.
(70)

Durov, V. A.; Tereshin, O. G. The Journal of Physical Chemistry B 2006, 110, 8441.

(71) Stokes, R. H. Journal of the Chemical Society, Faraday Transactions 1: Physical
Chemistry in Condensed Phases 1977, 73, 1140.
(72) CRC Handbook of Chemistry and Physics, 71st Edition; Lide, D. R., Ed.; CRC Press:
Boca Raton, FL, 1990-1991.
(73)

Analytical and Bioanalytical Chemistry 2004, 378, 26.

(74)

Luyben, W. L. Industrial & Engineering Chemistry Research 2014, 53, 5590.

(75) Li, W.; Shi, L.; Yu, B.; Xia, M.; Luo, J.; Shi, H.; Xu, C. Industrial & Engineering
Chemistry Research 2013, 52, 7836.
(76) Rév, E.; Lelkes, Z.; Varga, V.; Stéger, C.; Fonyó, Z. Industrial & Engineering Chemistry
Research 2002, 42, 162.

24

(77)

Furzer, I. A. Industrial & Engineering Chemistry Research 2001, 40, 990.

25

CHAPTER 2
TIME-RESOLVED PUMP-PROBE SPECTROSCOPY

2.1 Introduction
In this chapter we present the instrumentation developed in our lab for two complementary
time-resolved stimulated emission measurements.

The first is anisotropy decay and the second

is vibrational population relaxation, both with ca. 5 ps time resolution. Comparing these
techniques to the time correlated single photon counting (TCSPC) instrument operating in our
lab, which is capable of 35 ps time resolution at best, this stimulated emission system is able to
provide the requisite sensitivity and stability to allow information to be extracted from complex
systems, such as multi-component solvents, micelles and vesicles, lipid bilayers and light
harvesting protein systems.

It is important to note that this pump-probe instrument is capable of

detecting both stimulated emission, which depends on S1 population, and ground state recovery,
which depends on S0 population, whereas TCSPC senses spontaneous emission from the S1 state
of the sample of interest.

By combining the power and stability of the light source with shot

noise limited detection sensitivity,1 this time-resolved spectroscopic instrument is suitable for
both anisotropy decay and vibrational population relaxation measurements on a variety of
chemical and biochemical systems.

For the investigation of binary solvent systems and lipid

bilayer structures, we incorporate selected probe molecules in the sample and monitor the time
domain response of this probe to gain information on its immediate environment. We measure
rotational diffusion and vibrational resonance energy transfer under degenerate donor-acceptor
26

conditions.

Knowledge of both types of complementary information provides useful

information on the chromophore local environment.

In this chapter we discuss the instrument

and its sub-systems, including the light source and the high S/N detection methodology used in
this setup, as well as the principles of each measurement.

27

2.2 Instrument Schematic
The overall schematic of the time-resolved pump-probe spectroscopic system is shown in
Figure 2.1.

The purple line represents the pump laser which is used to excite the sample of

interest, and the blue line represents the probe laser which is used to investigate the signal
generated from the sample as a function of time after excitation.
divide the system into five major sub-systems:

Light source, modulation, time delay and

polarization control, sample and detection electronics.
sub-system are discussed in detail.

28

In the following discussion we

The parameters and functions of each

Figure 2.1. Schematic of the time resolved pump-probe instrument, showing light source
components (red), detection components (green), and beam paths.

29

2.2.1 Light Source
Source laser.

The pump laser for this system is a mode-locked Nd:YVO4 laser (Spectra

Physics Vanguard), which produces 13 ps pulses at 80 MHz repetition rate, with 2.5 W average
power output at both 532 nm (second harmonic of the 1064 nm fundamental wavelength) and
355 nm (third harmonic).

The stability of this source laser is characterized by less than 1% drift

in average power over 12 hours after 15 minutes of thermal equilibration time.

For the

pump-probe measurements discussed here, only the third harmonic light is used to excite the
cavity dumped dye lasers.
Dye laser.

Synchronously pumped, cavity dumped dye lasers (Coherent 702) have been

used as the light sources in this work.

This technology has been used widely to produce

picosecond laser pulses for a broad range of time- and frequency-domain spectroscopies.2

The

primary reason for using synchronously pumped cavity dumped dye lasers is their wide
wavelength tunability, moderately high peak power, which is important for both fast anisotropy
and vibrational relaxation measurements, and control over the repetition rate of the output pulses.
We use picosecond rather than femtosecond pulses in this work because the spectroscopic
features of interest are molecular vibrational bands with linewidths of 20 cm-1 or less for the
vibrational relaxation measurements, which require the light sources to be characterized by a
several cm-1 linewidth for the reasons of spectroscopic resolution.

Based on the Fourier

relationship between the time and frequency domains, the time width of the light pulses have to
be no shorter than a few picoseconds to meet this frequency domain linewidth constraint.3
30

cavity dumping electronics
tunable output
532 nm
Coherent 702 dye laser

Nd:YVO4 laser

355 nm

Coherent 702 dye laser

545 – 900 nm
(275 – 450 nm by SHG)
430 – 545 nm

Figure 2.2. Schematic of laser light source system.
The green light indicates second harmonic output and the blue line indicates third
harmonic output from the Nd:YVO4 laser. Cavity dumping electronics are used on the
dye lasers to control the repetition rate of their output. The wavelengths of the dye lasers
depend on the dyes and optics used. In the configuration used in this work, the third
harmonic output of the Nd:YVO4 laser excites both dye lasers.

The schematic of the laser light source system is shown in Figure 2.2.

The repetition rate

of each dye laser is controlled by cavity dumping with a cavity dumper driver (Gooch and
Housego 64389.5–SYN–9.5–1).

The output of each dye laser is characterized by ca. 5 ps

pulses with a repetition rate ranging from 80 MHz to 80 kHz based on the setting of cavity
dumping electronics.

The dye laser pulse repetition rate has been used in this work is 8 MHz

(125 ns inter-pulse spacing) with ~100 mW average output power. The output wavelength of
the dye laser is tunable and depends on the dye used, the optics in the dye laser cavity and the

31

excitation wavelength.

Generally, the dye laser output can be tuned from 430 nm to 545 nm

using 355 nm (third harmonic) excitation and from 545 to 900 nm using 532 nm (second
harmonic) excitation.

For the experiments reported in this dissertation, Stilbene 420 laser dye

(Exciton) has been excited at 355 nm and tuned from ca. 430 nm to ca. 480 nm depending on the
system under investigation.

The Stilbene 420 dye solution is prepared by dissolving 2 grams of

dye in 1.2 L of ethylene glycol solvent.

The excitation and emission spectra of perylene, the

Linear response (a.u.)

chromophore in this work are shown in Figure 2.3.

300

350

400

450

500

550

600

Wavelength (nm)
Figure 2.3. Excitation and emission spectra of perylene in ethanol.
Because the pulses used to excite the dye lasers in our system are derived from a single
source laser, the pulsed outputs of the dye lasers are synchronized tightly in time, which is
critical for our measurements.

In practice, the synchronization jitter between the dye laser

pulses limits the time resolution of the measurement to ca. 5 ps.
32

The “pump” laser pulse

(purple line) is used to excite the sample of interest at a wavelength determined by origin
transition energy of chromophore in the particular solvent system.

The “probe” laser pulse

(blue line) interrogates the perturbation to the sample caused by pump pulse as a function of time
and polarization differences between the pulses.

The probe pulse wavelength is offset from the

pump pulse wavelength to access a specific vibrational mode of chromophore.

In the work

presented here, the probe pulse is shifted 1375 cm-1 to the red from the perylene spectral origin.
The 1375 cm-1 mode is the symmetric ring breathing mode of perylene.

By varying the

wavelength of probe pulse, the vibrational population relaxation of perylene from selected
vibrational modes into degenerate or near-degenerate bath modes can be evaluated.
specific donor vibrational modes of perylene are indicated in Table 2.1.

33

The

Table 2.1. Selected perylene vibrational modes and assignments.

Wavenumber (cm-1)

Vibrational motion of perylene

1375

v7(ag) ring breathing

1298

v9(ag) naphthalene rings rocking in plane

1141

v26(ag) CH bending

979

v12(ag) symmetric ring deformation

549

v14(b3g) asymmetric ring deformation

357

v16(ag) butterfly mode

The modes need to be chosen carefully because only the modes with near degeneracy to a bath
mode can be used to provide complementary information on energy relaxation.

The unique

aspect of these measurement is the ability to probe the coupling between the chromophore and
specific acceptor modes of its immediate environment, both as a function of mode identity and
length-scale of coupling (r-8 vs. r-6, vide infra).4 It is the ability to tune the laser wavelengths
precisely and achieve ~ 5 ps time resolution that makes this light source well suited to the
experiments described in this dissertation.
The same pump and probe laser wavelengths were used for vibrational population relaxation
and orientational relaxation measurements. The exact wavelengths of both pump and probe
pulses used in the measurements will be specified in the experimental sections of Chapters 3 and 4.
34

2.2.2 Modulation
The magnitude of the signal recorded with this instrument is on the order of T/T ~ 10-5 due
to the presence of only small amounts of fluorescent chromophores in the sample. The low
frequency flicker noise of the lasers is typically several percent of the overall output, which is at
least two orders of magnitude larger than the expected experimental signal.

In order to detect the

small T/T signals characteristic of these experiments, the detection frequency needs to be shifted
into a shot noise limited region of the dye laser output.

The output of the dye laser is

characterized by a noise spectrum that decays as 1/f, reaching the shot noise floor at frequencies
above ~1 MHz, as schematized in Figure 2.4.

35

Figure 2.4. Noise spectrum of dye laser output, showing the flicker noise (low frequency)
and shot noise (high frequency) limited regions.

In our system, the shift in detection frequency is accomplished using a hybrid electro-optic
and mechanical modulation signal encoding scheme.5-8

In this scheme, a different radio

frequency electro-optic modulation is applied to the pump and probe pulse trains.

The

electro-optic modulations are capable of producing 100% depth of modulation with 50%
throughput of average power for a sinusoidal modulation waveform.1
In its simplest form, the signal can be visualized as a single modulation experiment, where
the pump beam is modulated at a frequency, pu, and interactions between the pump and probe
36

beams in the sample will produce a transfer of modulation from the pump beam to the probe
beam.

The amplitude of the transferred modulation is proportional to the interaction of both

beams with the sample, and detecting the modulation on the probe beam at frequency pu yields
the signal of interest.

Unfortunately, a single modulation is not sufficiently detectable in

practice because of optical scattering and electronic pick-up effects in the detector housing that
produce a large and variable background.

To achieve improved electronic and physical

isolation of the signal, the pump and probe beams are modulated at two different RF frequencies,

1 and 2. The experimental signal is the result of the mixing of two modulation frequencies
by the sample, which is manifested as the sum and difference of the two modulation frequencies,
as described by Eq. 2.1.

1
1
cos 1  cos 2  cos(1  2 )  cos(1  2 )
2
2

(2.1)

When both laser pulses are co-focused into the sample, the sample is functioning as a
molecular frequency mixer, and the efficiency of mixing is determined by the extent to which the
sample interacts with electric fields of both the pump and probe beams.

The experimental

signal is detected at frequency1 + 2, which is located in a frequency range characterized by
shot noise limited background, as shown in Figure 2.4. Noise spectrum of dye laser output,
showing the flicker noise (low frequency) and shot noise (high frequency) limited regions..
While this additional modulation frequency enhances the robustness of data acquisition it is still
susceptible to some electronic and optical interference effects.

To further discriminate between

the electronic broadcast and nonlinear frequency mixing effects associated with the photodiode
37

detector, an audio frequency modulation is applied mechanically to the pump beam by
introducing a chopper in to the system.

The addition of the audio frequency component to the

modulation has the effect of adding sideband modulations at AF to the sum and difference
frequency terms on the right hand side of Eq. 2.1.
in Figure 2.5.

This overall effect from modulation is shown

For the work presented here, the pump and probe laser pulse trains are amplitude

modulated at 3 MHz and 2 MHz respectively, using electro-optic modulators (Thor Labs Model
EO-AM-R-X-C4, 400-600 nm, where X is the modulation frequency, in MHz) driven by
synthesized function generators (Stanford Research Systems, Model DS340) and mechanical
modulation is applied to the pump pulse train by a chopper (HMS Light Beam, Model 220) at 50
Hz.

Figure 2.5 Electro-optic (EO) and mechanical hybrid modulation signal encoding scheme
used in the pump-probe spectroscopy system
38

2.2.3 Time-delay and Polarization Control
The time delay between pump and probe pulses is controlled by a precision motorized
long-travel linear translational stage (Parker Daedal 110708R0175), on which is mounted a
broadband corner-cube retro-reflector (Newport UBBR 2.5-1, 63.5mm, 73%, 1 arc sec, 400-700
nm).

The position and movement of the stage-mounted retro-reflector is controlled by a

program written-in-house using National Instruments LabVIEW® programming language. The
effective delay time range using this translation stage is in excess of 1.5 ns, and the
step-resolution typically used in this work is 1 ps.
The polarization of probe beam is controlled by a broadband polarization rotator mounted on
a motorized high-resolution rotational stage (Newport, Model URM100CCHL) connected with a
universal high-performance motion controller driver (Newport, Model ESP100).

For

orientational anisotropy decay measurements, the polarization of the probe beam is set to 0°and
90° with respect to the polarization of the pump beam for the acquisition of parallel and
perpendicular time-resolved data.

For vibrational population relaxation measurements, the

polarization of probe beam is set to 54.7° with respect to that of the pump to remove
contributions from molecular motion to the experiment signal.

39

2.2.4 Optical Flow-cell and Sample Pumping
The pump and probe laser pulses are temporally and spatially overlapped in the sample of
interest.

To minimize the effects of thermal lensing induced by a temperature gradient

generated by the heating of the sample by the pump laser pulse, we use a flow cell (quartz,
48ES1, Precision Cells Inc.) and associated pumping equipment.

The sample flow is

perpendicular to the axis of the incident laser beam, with the flow rate controlled by a
magnetically driven gear pump (MICROPUMP, GA series).

During the measurements, the

sample solution flows in a closed circuit from a water-jacketed, temperature controlled 4-mL
reservoir, through the micro-pump and the flow cell.
The advantages of using this sample flowing setup include: 1) Small amount of sample
solution needed (≤ 5mL total); 2) controllable sample flow rate (between 9.5 L/min and 42
mL/min); 3) wide operating temperature range; 4) wide sample range (organic, inorganic and
biological) due to the high chemical and abrasion resistance of cell and pumping gear materials;
5) easy and fast pre-/post- experimental sample cleaning.

2.2.5 Detection Electronics and Signal Processing
After the probe passes through the sample, the transmitted signal intensity is detected using
a large area photodiode (RCA, Model C30956E), coupled with a multi-step synchronous
demodulation detecting system.

Detection of the signal component modulated at 5 MHz (1 +
40

2) is accomplished by a radio frequency lock-in amplifier (Stanford Research Systems Model
SR844).

The output of this device is further demodulated by an audio frequency lock-in

amplifier (Stanford Research Systems Model SR510), at the chopping frequency AF (50 Hz), to
extract the experimental signal T.

The 5 MHz reference signal is generated electronically

using a double balanced mixer (Mini Circuits ZAD-1W) and a Butterworth filter (Drake
Electronics) to remove the difference frequency component.

The quantity T is acquired by

lock-in detection of the probe intensity at modulation frequency 2 (2 MHz).

For this detection

method, the ultimate sensitivity is T/T ~ 10-7, depending on the intensity of the probe laser
pulse at the detector (the detection limit is taken as 3x the shot noise on the probe laser pulse).
Such sensitivity is sufficient to measure the vibrational relaxation dynamics in lipid bilayers with
a chromophore loading of 1 chromophore per 1250 lipid molecules.
The resulting signals, T and T, are sent to a computer (PC) for processing and storage.
The instrument is controlled (translation stage position and movement, probe beam polarization
angle, acquisition and storage of experimental T and T signals) by a program written in-house
using National Instruments LabVIEW® programming language.

Each reported time constant is

based on the average of at least six individual data acquisitions, and each acquisition is the
average of at least three time-scans.
In the next section, we will discuss the two types of measurements that have been conducted
using the pump-probe instrument, as well as the information content of the data observed.

41

2.3 Molecular Reorientation Measurements
There are two physical processes that could contribute to anisotropy decay data: rotational
diffusion9-11 and resonant energy transfer12-17 between like molecules.

However, in our case,

the anisotropy decay signal collected from the instrument is dominated by rotational diffusion
due to the low concentration of chromophore present in the sample (typically ~10-5 M), with the
distance of ca. 500nm between solute molecules.

Based on the instrumentation described

above, we quantitate the molecular reorientation behavior of the chromophore using fluorescence
anisotropy decay measurements.
To elucidate the intrinsically anisotropic nature of chromophore molecular motion, Chuang
and Eisenthal derived a series of equations that relate the spectroscopic and dynamical properties
of the solute molecules to the functional form of the experimental data.11 Polarized stimulated
emission transients are acquired for polarizations parallel, I||(t), and perpendicular, I(t), to the
vertically-polarized excitation pulse. These raw signals are combined to produce the induced
orientational anisotropy function, R(t ) , (Eq. 2.2).
R(t ) 

I (t )  I  (t )
I (t )  2 I  (t )

(2.2)

The decay of R(t) is a result of the re-randomization of a non-random orientational
distribution of chromophores selected by the pump pulse.

The R(t) function can contain up to

five exponential components, however, experimentally only one- or two-component anisotropy
decays are observed for perylene and 1-methylperylene.7,18

42

Relating the decay of R(t) to the Cartesian components of the rotational diffusion constant is
accomplished through a series of equations derived by Chuang and Eisenthal,11 who showed that
the functional form of R(t) depends on the shape of the volume swept out by the reorienting
chromophore molecule.

This shape is modeled as an ellipsoid and the Cartesian components of

the rotational diffusion constant are related to rotor shape as follows.

We assign the planar

chromophore π system as the x-y plane, with the x-axis being the long in-plane axis and the
y-axis being the short in-plane axis.

The z-axis is normal to the chromophore π (x-y) plane.

For an oblate rotor, Dz>Dy=Dx and for a prolate rotor, Dx>Dy=Dz.
two ellipsoidal rotor shapes.

43

Figure 2.6 below shows the

Figure 2.6. Two ellipsoid rotor types. Cartesian components of the rotational diffusion
constant (D) upon the assignment of axes and rotor ellipsoid shapes.

The number of decay components in R(t) indicates the relative orientation of the excited and
probed transition dipole moments and the magnitude of R(0) is related to the angle between the
excited and probed transition dipole moments.

For a typical S1←S0 transition, where the

excited and emitting transition moments are nominally parallel to one another, if the transition
dipole moments are polarized along the chromophore x (long) axis, reorientation as a prolate
rotor gives rise to a single-exponential anisotropy decay functionality,

R(t )  0.4 exp(6Dz t )

44

(2.3)

where the only dynamic information available is for rotation about the axis (z), perpendicular to
the observation (x) axis.

Reorientation as an oblate rotor gives rise to a two exponential

anisotropy decay,

R(t) = 0.1exp(-(2Dx + 4Dz )t) + 0.3exp(-6Dx t)

(2.4)

where information is available on rotation about two different axes (x, z).
Conversely, if the transition moments accessed are short-axis (y) polarized, we will see for a
prolate rotor,

R(t) = 0.1exp(-(4Dx + 2Dz )t) + 0.3exp(-6Dx t)

(2.5)

and for an oblate rotor,

R(t) = 0.4exp(-(2Dx + 4Dz )t)

(2.6)

For perylene it is known that the S1 ← S0 transition is long (x) axis polarized, so Eqs. 2.3 and 2.4
are operative.
The determination of rotational diffusion constant (D) is based on measuring the
experimental Cartesian components, where,

D  13  Dx  Dy  Dz 

(2.7)

D is related to its immediate surroundings through Eq. 2.8, the modified Debye-Stokes-Einstein
(DSE) equation, 19-23
D

k BTS
6Vf

(2.8)

Where is the solution bulk viscosity, V is the solute hydrodynamic volume (225 Å3 for
perylene)24, f is a boundary condition term to account for frictional contributions to solvent-solute
45

interactions, which can range from near zero in the slip limit to unity in the stick limit depending
on the effective shape of the rotating ellipsoid,21,22 T is the solution temperature, kB is the
Boltzmann constant, and S is a shape factor that accounts for the non-spherical shape of the
solute.20 For perylene, S = 0.69.25
Distinguishing between one and two-component exponential decays in R(t) can be a
challenge in practice, since it requires both high time resolution and high signal-to-noise ratio
data to resolve multiple components.

Making this distinction for anisotropy decay

measurements relies on the ability to resolve time constants that are in the 20 ps to 100 ps time
range in most cases, a time-window that is difficult to access with conventional TCSPC
instrumentation.

However, the pump-probe instrument described here provides the time

resolution to resolve fast anisotropy decay components, and thus the ability to extract the (fast)
Cartesian components of D, providing a new level of detail to understanding molecular-scale
interactions in a variety of multi-component and biomolecular systems.

2.4 Vibrational Relaxation Measurements
The detailed process of thermal energy dissipation from molecules into their surrounding
bath is an essential and unexplored issue for many liquid phase and biological systems,
especially with regard to bilayer and protein dynamics.

The pump-probe spectrometer we

developed is designed to measure the vibrational energy transfer from probe chromophores to the
surrounding bath in fluid systems.

Vibrational energy transfer is the molecular process that

ultimately gives rise to macroscopic thermal conductivity, and the rate of vibrational relaxation is
46

related to the thermal conductivity of the medium surrounding the chromophore.26,27

From

these measurements on vibrational mode-specific relaxation of probe molecules in liquid and
lipid bilayers of controlled composition, we ultimately hope to develop systematic knowledge of
how chemical composition, fluidity and spatial heterogeneity are related to vibrational energy
flow within these systems.

2.4.1 Molecular Contribution to the Experimental Signal
As mentioned above, the information content of the T/T signal depends on the pump and
probe wavelengths and their relative polarizations.

For vibrational relaxation measurements,

the probe wavelength is shifted to the red of the pump wavelength by the vibrational resonance
of the chromophore, for example, 1375 cm-1 for the perylene ring breathing mode.6 To eliminate
the contributions from molecular motion to the experimental signal, the probe beam polarization
is set to 54.7°with respect to that of the pump.

47

S10

S0v
S00
Figure 2.7. Schematic three level system.

The coupled three level system (Figure 2.7) is a well established simplified model of the
probe molecule, where the pump pulse excites the chromophore S10  S00 (0-0) transition and the
probe is set to the wavelength of the S0v  S10 transition.

Both stimulated emission and

absorption occur between levels coupled by an electric field (S00  S10 and S0v  S10 transitions).
For perylene used in this work, the S0v  S10 transition and S00  S10 transition have similar
transition cross section. The population difference between the two states represents the relative
contribution of absorption and stimulated emission to the overall observed signal. The chemical
information content of the pump-probe data is determined by the time evolution of the populations
of each state involved.

The relationship derived by Einstein shows

Bij  B ji

and

Aij  (8 hv3 / c3 ) , where Bji is the rate constant for absorption, Bij is the rate constant for
stimulated emission and Aij is the rate constant of the spontaneous emission.
i is taken to be the excited state and j is taken to be the ground state.
48

In this description,

As shown in Eq. 2.9,

dS j
dSi

  Aij Si  Bij v ( S j  Si )
dt
dt

(2.9)

v is the power density of the incident electric field at frequency v, and the terms Si and Sj refer to
two states coupled by the electric fields showing in Figure 2.7 respectively.

It is the time

evolution of Si and Sj that contain the information of interest.

2.4.2 Kinetic Model of the Time-Domain Signal
S10

k12
k 21

S0v 
 S00 (scheme I)
k 23

We can treat the population dynamics of this system in the context of scheme I for
vibrational states S0v substantially higher in energy than kBT.

In the experiment, the probe laser

pulse senses both absorption and stimulated emission for the S0v  S10 transition. A set of
coupled differential equations has been introduced to model the time evolution of the populations
of all states. 28,29
S10  t  

In Equation 2.10, the solution of S10(t) and S0v(t) are shown,

 2  3
 2  2
exp  3k12t  
exp  2k12t 
3  2  3 
2  2  3 

1
S t  
exp  3k12t   exp  2k12t  
2  3 

(2.10)

v
0

where the terms  are the normalized rate constants (1 = k21/k12 and 2 = k23/k12) and  are the
positive and negative roots of the solution to the coupled differential equations.

Here the rate

constant k23, which is used to describe S0v population relaxation, is typically much larger than the
rate constant k12 which is the sum of the spontaneous and stimulated emission rate constants
from S10. Under these conditions, λ2 ≈ 1 and λ3k12 ≈ k23.30
49

The experimental signal detected by the probe laser beam is the difference between gain
from S10  S0v stimulated emission and loss from S0v  S10 absorption. Generally, the time
evolution of S10  S0v gain will be different from that of S0v  S10 loss as the initial populations
of these two states are different.

The signal is virtually all gain at the time near zero, because

S0v(0) ≈ 0 and intuition suggests a larger signal compared to times later when both gain and loss
contribute to the detected signal.

Once absolute changes in the number of photons resulting

from the interaction of the probe laser pulse with the sample were detected, this form of signal
could be generated. However, after applying modulations to the pump and probe pulse trains to
encode the experimental signal, the form of signal changed.30

The expected form of the signal

(T/T) is that of a difference between two exponential functions (Eq. 2.11),
S t  

1

1
 exp  3k12t   exp  2k12t  
2
 2  3   3


2

(2.11)

with the vibrational population relaxation time (v = T1 = k23-1) appearing to be a build-up in the
signal after excitation. The vibrational relaxation time constant T1 characterizes the relaxation
of population from specific chromophore vibrational modes into its immediate environment.
For the three-modulation detection scheme we used, the expected form of the fitted signal
becomes,
S (t )

k (exp( t /  elec ) 

T1

 elec

exp( t / T1 ))

(2.12)

where elec is the decay time constant associated with the depopulation of the excited electronic
state according to the spontaneous and stimulated emission, and T1 is the vibrational population
relaxation time constant.
50

Both anisotropy decay data and vibrational population relaxation data obtained from the
pump-probe instrument are used in subsequent Chapters of this dissertation to understand local
organization in structurally heterogeneous systems.

51

REFERENCES

52

REFERENCES

(1)

Blanchard, G. J.; Wirth, M. J. Analytical Chemistry 1986, 58, 532.

(2)

Ratner, M. A. International Journal of Quantum Chemistry 1987, 31, 989.

(3)
990.

Sala, K.; Kenney-Wallace, G.; Hall, G. Quantum Electronics, IEEE Journal of 1980, 16,

(4)
Gray, C. G.; Gubbins, K. E. Theort of Molecular Fluids, Vol. 1. Fundamentals; Oxford
Science, 1984.
(5)

Fleming, G. R. In Advances in Chemical Physics; John Wiley & Sons, Inc.: 2007, p 1.

(6)

Bado, P.; Wilson, S. B.; Wilson, K. R. Review of Scientific Instruments 1982, 53, 706.

(7)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.

(8)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 6785.

(9)

Lipari, G.; Szabo, A. Biophysical Journal, 30, 489.

(10)

Szabo, A. The Journal of Chemical Physics 1984, 81, 150.

(11)

Chuang, T. J.; Eisenthal, K. B. The Journal of Chemical Physics 1972, 57, 5094.

(12) Suhling, K.; French, P. M. W.; Phillips, D. Photochemical & Photobiological Sciences
2005, 4, 13.
(13)

Baumann, J.; Fayer, M. D. The Journal of Chemical Physics 1986, 85, 4087.

53

(14) Ediger, M. D.; Domingue, R. P.; Fayer, M. D. The Journal of Chemical Physics 1984, 80,
1246.
(15)

Ediger, M. D.; Fayer, M. D. The Journal of Chemical Physics 1983, 78, 2518.

(16)

Ediger, M. D.; Fayer, M. D. Macromolecules 1983, 16, 1839.

(17) Keller, L.; Hussey, D. M.; Fayer, M. D. The Journal of Physical Chemistry 1996, 100,
10257.
(18) Qiu, C.; Blanchard, G. J. Spectrochimica Acta - Part A: Molecular and Biomolecular
Spectroscopy 2013, 110, 383.
(19)

Zwanzig, R.; Harrison, A. K. The Journal of Chemical Physics 1985, 83, 5861.

(20)

Perrin, F. J. Phys. Radium 1936, 7, 1.

(21)

Hu, C. M.; Zwanzig, R. The Journal of Chemical Physics 1974, 60, 4354.

(22)

Youngren, G. K.; Acrivos, A. The Journal of Chemical Physics 1975, 63, 3846.

(23)

Journal of the Society of Chemical Industry 1929, 48, 1036.

(24)

Edward, J. T. Journal of Chemical Education 1970, 47, 261.

(25)

Jiang, Y.; Blanchard, G. J. The Journal of Physical Chemistry 1994, 98, 6436.

(26)

Callaway, J. Physical Review 1959, 113, 1046.

(27)

Nicolazzi, W.; Pillet, S.; Lecomte, C. Physical Review B 2008, 78, 174401.

(28)

Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 9411.

54

(29)

Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 9417.

(30)

Blanchard, G. J. Review of Scientific Instruments 1996, 67, 4085.

55

CHAPTER 3
ORIENTATIONAL AND VIBRATIONAL RELAXATION DYNAMICS OF
PERYLENE IN CYCLOHEXANE-ETHANOL BINARY SOLVENT SYSTEM*

3.1 Introduction
Understanding molecular scale organization and heterogeneity in liquid phase systems
continues to be a challenge because of the transient nature of intermolecular interactions.
Achieving a fundamental understanding of solution phase heterogeneity is a prerequisite for
gaining predictive control over bulk solution phase properties, such as thermal conductivity,
viscosity or density.1-4 For neat solvents there is a well-established body of work that has
provided significant insight into the interactions between dissimilar molecules that mediate
molecular motion and intermolecular energy transfer.5-16 Such understanding is more limited for
solute-solvent systems where the solvent is comprised of more than one constituent. In the
studies of neat solvents, the modified Debye-Stoke-Einstein (DSE) model provides a reasonable
means of predicting the chromophore rotational diffusion.17

However, in multi-component

solvent systems, due to the fact that the DSE theory doesn't take into account the details of the
coupling between solute and solvent(s) or any solution phase heterogeneity in composition,

*

The work in this chapter has been published entitled “ Orientational and Vibrational Relaxation
Dynamics of Perylene in the Cyclohexane–Ethanol Binary Solvent System” on The Journal of
Physical Chemistry B 2014, 118, 10525.
56

experimental reorientation data can deviate significantly from model predictions.18 Achieving a
detailed understanding of intermolecular interactions in these systems is an area of active
investigation because of the broad utility of complex solvent systems. For several binary solvent
systems, the notion of one solvent being distributed uniformly in the second solvent is frequently
a poor assumption at the molecular scale.

One significant limitation that exists from an

experimentalist’s standpoint is the ability to interrogate any (non)uniformity within a binary
solvent system, primarily because of the inability to locate a probe chromophore in the
immediate proximity of the heterogeneity of interest.
Because the investigation of binary solvent systems is significantly more challenging
than that for neat solvents, it is important to choose systems carefully. Binary solvent systems
typically contain solvents of reasonably similar polarity for reasons of miscibility.

The

miscibility of cyclohexane and ethanol is anomalous in that respect. Ethanol is a polar solvent,
characterized by extensive hydrogen bonding, a permanent dipole moment of ca. 1.7 D and a
dielectric constant of ca. 24.19 Cyclohexane has a dipole moment of ca. 0 D and a dielectric
constant of ca. 2.20 Despite these differences in solvent properties, ethanol and cyclohexane are
miscible over a wide range of concentrations.21 Further, properties such as bulk viscosity and
dielectric constant for this binary mixture vary monotonically with the mole fraction of ethanol
in cyclohexane.22,23 To evaluate whether or not the correspondence between concentration and
bulk properties extends to the molecular level, we have chosen perylene as the probe molecule,
and examined the rotational diffusion and vibrational population relaxation dynamics of this
57

chromophore in the ethanol-cyclohexane binary system.

Our data show that local

organization/heterogeneity does indeed exist in this system, and that it changes in a manner that
is consistent with a discontinuous reorganization of the perylene immediate environment with
increasing ethanol concentration.
We are interested in understanding local organization in multi-component solvent systems
for reasons of quantitating and controlling vibrational energy flow, and providing insights on
tailoring the solvent system for “desired properties”. There have been several studies on the
rotational diffusion dynamics of polar and nonpolar molecules in binary solvent systems by using
steady-state and time-resolved spectroscopic methods, theoretical treatments and molecular
dynamics simulations.24-33

We report here on the vibrational population relaxation and

orientational relaxation behavior of perylene in the ethanol-cyclohexane binary system. We have
measured both the orientational relaxation time of perylene and the vibrational population
relaxation time of the perylene 1375 cm-1 ring breathing mode because these quantities sense
different aspects of the immediate environment of the chromophore. The reorientation time is
sensitive to solvent-solute frictional interactions and confinement imposed on the rotating
chromophore by the solvent surrounding it. The vibrational population relaxation measurement
reports on the coupling efficiency of the perylene chromophore ring breathing mode and the
ethanol terminal methyl group.

Because perylene is centrosymmetric, the 1375 cm-1 ring

breathing mode modulates the molecular quadrupole moment and the characteristic length scale
of the perylene-ethanol vibrational energy transfer process scales as r-8.34 Despite the continuous
58

trends seen in the bulk properties of the ethanol-cyclohexane mixed solvent, our data reveal that
the perylene probe experiences composition-dependent local organization within this binary
system. Both experimental results show the discontinuity occurs at the same ethanol
concentration, indicating the different solvent molecular environments in solutions >7.5 %, (v/v)
ethanol and <5 % (v/v) ethanol. These data demonstrate that, from a microscopic perspective,
perylene experiences a qualitative change in its local environment over a relatively narrow
solution composition range. Such data cannot be accounted for by means of a homogeneous
binary solvent system.

59

3.2 Experimental
Laser System. A pump-probe laser spectrometer was used for both the time-resolved
anisotropy decay and vibrational population relaxation measurements reported here.

The

detailed instrumentation and measurement principles have been described in Chapter 2. For the
experiments on perylene in ethanol-cyclohexane binary solvents reported in this chapter, the
pump laser was set to specific wavelengths between 435.4 nm and 436.7 nm, depending on the
0-0 transition energy of perylene in the particular solvent mixture under investigation. The probe
laser wavelength was set to specific wavelengths between 463.2 nm and 464.6 nm, shifted 1375
cm-1 to the red of the pump pulse to access the perylene symmetric ring breathing mode. The
same pump and probe laser wavelengths were used for vibrational population relaxation and
orientational relaxation measurements. Each reported time constant is based on the average of at
least six individual data acquisitions, and each acquisition is the average of at least three timescans.
Steady-State Spectroscopy. The steady-state excitation spectra of the chromophore in the
ethanol-cyclohexane binary solvent system were obtained using a Jobin-Yvon Fluorolog 3
spectrometer, exciting the sample over the range of 300 nm to 450 nm with the emission
collection wavelength set to 460 nm. Spontaneous emission spectra were obtained using the
same instrument, exciting the sample at 410 nm and collecting emission from 420 nm to 600 nm.
For all measurements the excitation and emission monochromators were set to 0.1 nm resolution.
The steady state spectroscopic data were used to determine the appropriate pump and probe
60

wavelengths for perylene in each ethanol-cyclohexane binary solvent system. The excitation and
emission spectra for perylene in ethanol and in cyclohexane are shown in Figure 3.1.
Chemicals and Sample Handling. Perylene (99+ %, Sigma-Aldrich, Milwaukee, WI)
was used without further purification. Ethanol (99+ %, A.C.S. Spectrophotometric grade) and
cyclohexane (99+ %) were purchased from Jade Scientific in the highest purity grade available.
All solutions (10-5 M in perylene) were made fresh before each set of measurements. All pumpprobe measurements were performed with the sample housed in a quartz flow cell connected to a
magnetically driven external gear pump (Micropump®, IDEX Health & Science, LLC) to
eliminate thermal lensing contributions to the signal.

The flow rate was controlled by a

switching DC power supply (BK Precision). All measurements were performed at 20°±1°C.

61

3.3 Results and Discussion
The primary goal of this work is to understand the solvent-solute interactions between
perylene and the ethanol-cyclohexane binary solvent system. Placing this result in context with
earlier work from different groups focusing on the intermolecular interactions of the same probe
in neat solvents5-8,10,35-37 provides some insight into the factors mediating the energy transfer
process in this binary system. We show here that both the orientational anisotropy decay and
vibrational population relaxation data indicate strong solvent-solute molecular scale interactions
and organizational heterogeneity in the binary solvent system. Before detailing the time-domain
results, we consider the steady-state optical properties of perylene in the ethanol-cyclohexane
system. The subtle, solvent-dependent variations in the steady state optical response of perylene
correlate with the time-domain data, pointing to persistent local organization in this binary
solvent system.
Steady-State Spectroscopy. As a widely used fluorescent probe molecule, the absorption
and emission properties of perylene are well understood. Perylene is characterized by a high
fluorescence quantum yield and it exhibits no anomalous spectroscopic behavior.7 The quasimirror image excitation and emission spectra shown in Figure 3.1 indicate that the electronic
excited states of perylene do not exhibit intensity borrowing and that vibronic coupling plays a
minimal role in the spectral response of this molecule.38 The solvent-dependent linear optical
response of perylene in cyclohexane and ethanol is manifested in small spectral shifts and no
significant change in the band profiles. Steady state spectroscopic studies in this binary solvent
62

system have been reported by Józefowicz39 on the chromophores flouorenone and 4hydroxyflouorenone, with the observation of non-linear solvatochromic shifts as a function of
the polar component (ethanol) in the ethanol-cyclohexane binary solvent system, which can be
explained by the formation of hydrogen bonded complexes during the solvent-shell
rearrangement.
The 0-0 transition energy of perylene in each binary solvent system studied here is
obtained from the normalized excitation and emission spectra and its dependence on solvent
system composition is shown in Figure 3.1. We observe a monotonic blue shift in the 0-0
transition wavelength of perylene with increasing ethanol concentration. It is important to note
that, while this trend is monotonic with increasing ethanol concentration, it is not linear. Rather,
there are three distinct regions, each with a different slope. This finding is suggestive of the
nature of the solvent system being interrogated by perylene changing with increasing ethanol
concentration.

Understanding the reason(s) for these spectral shifts can be aided by the

acquisition of time-resolved fluorescence anisotropy decay and vibrational population relaxation
data.
Molecular Reorientation.

Molecular reorientation is one of the more widely used

properties used in the investigation of solvent-solute interactions and local organization in liquid
media.5-7,40-49

Orientational relaxation measurements exist within the context of a well-

established theoretical framework for the interpretation of experimental data. The quantity
measured is the induced orientational anisotropy decay function, R(t), which is constructed, for
63

linear response (a.u.)

a
excitation

350

400
450
500
wavelength (nm)

550

linear response (a.u.)

300

emission

300

600

b

350

400
450
500
wavelength (nm)

550

600

Figure 3.1. (a) Excitation and emission spectra of perylene in cyclohexane. (b) Excitation
and emission spectra of perylene in ethanol. Intensities have been normalized.

64

436.8

0-0 transition wavelength (nm)

436.6
436.4
436.2
436.0
435.8
435.6
435.4
435.2

0

20

40

60

80

100

ethanol concentration (% v/v)

Figure 3.2. Dependence of perylene 0-0 transition wavelength on concentration of ethanol
in ethanol-cyclohexane binary solvent system. All spectra have been normalized and the 0-0
transition wavelength results are averaged from at least three measurements.

65

the data reported here, from stimulated emission transients polarized parallel (I||(t)) and
perpendicular (I(t)) to the vertically polarized pump pulse (Eq. 3.1)

R(t ) 

I (t )  I  (t )
I (t )  2 I  (t )

(3.1)

The decay of R(t) is a result of the re-randomization of a non-random orientational distribution
of chromophores selected by the pump pulse. It is the functional form of R(t) that is related to
the chemical information of interest. R(t) can contain up to five exponential decay components,
with the most common functionalities being one- and two-component exponential decays.
Relating the decay functionality of R(t) to the Cartesian components of the rotational
diffusion constant is accomplished through a series of equations derived by Chuang and
Eisenthal.50 They showed that the number of decay components in R(t) is related to the relative
orientation of the excited and probed transition dipole moments and the shape of the ellipsoidal
volume swept out by the rotating chromophore. In this work the anisotropy decay of perylene
changes from a single-component exponential decay to a two-component exponential decay with
changes in the binary solvent composition.
For perylene, the planar chromophore π system defines the x-y plane, with the x-axis
being the long in-plane axis and the y-axis being the short in-plane axis. The z-axis is normal to
the chromophore π plane. Using these assignments, the excited and probed transition dipole
moments of perylene (S1 ← S0) are coincident with the x-axis.

The anisotropy decay time

constant(s) can be related to the Cartesian components of the rotational diffusion constant,
66

depending on the aspect ratio of the ellipsoidal volume swept out by the rotating probe. This
volume can be described as a prolate ellipsoid, with the unique rotational axis being the x-axis
(Dx > Dy = Dz), or as an oblate ellipsoid, with its unique rotational axis perpendicular to the
chromophore -system (Dz > Dy = Dx). Reorientation as a prolate rotor gives rise to a singleexponential anisotropy decay functionality,

R(t )  0.4 exp( 6Dz t )

(3.2)

where the only information available is related to rotation about the axis (z) perpendicular to the
observation (x) axis. Reorientation as an oblate rotor gives rise to a two exponential component
anisotropy decay,

R(t )  0.1exp((2Dx  4Dz )t )  0.3 exp(6Dxt )

(3.3)

For this condition, information is available on rotation about two different axes (x and z). For
ethanol concentrations of 5% and less in cyclohexane, perylene produces a two-component
anisotropy decay function (Figure 3.3), and for cyclohexane solutions containing higher
concentrations of ethanol (7.5% and above), a single exponential anisotropy decay functionality
is seen (Figure 3.4). The zero-time anisotropies, R(0), as well as the decay time constants, , are
presented in Table 3.1.

67

T/T (a.u.)

1.0

a

0.8
0.6
0.4
0.2
0.0
0

100 200 300 400 500 600 700
delay time (ps)

0.2

b

R(t)

0.1

0.0
0

100 200 300 400 500 600 700
delay time (ps)

Figure 3.3. (a) Experimental time scans I||(t) (blue) and I(t) (red) for perylene in 75 %
ethanol/25 % cyclohexane, along with the instrumental laser cross-correlation function. (b)
Anisotropy decay function, R(t), for the data shown in panel (a). The decay is fitted to the
function R(t)=R(0)exp(-t/OR). Results for these fits are presented in Table 3.1.

68

Table 3.1. Reorientation times and zero-time anisotropies for perylene in cyclohexaneethanol mixed solutions. The data are the best fit result of the data to the function of
R(t)=R1(0)exp(-t/1) or R(t)= R1(0)exp(-t/1) + R2(0)exp(-t/2). Times are given in ps and
the uncertainties listed here are standard deviations (±1) based on 6-9 measurements for
each sample.

cyclohexane
%
100
99.75
99.50
99.25
99.00
97.50
95.00
92.50
90.00
75.00
50.00
25.00
0

solvent
ethanol
%
0
0.25
0.5
0.75
1.00
2.50
5.00
7.50
10.00
25.00
50.00
75.00
100

(cp)a

R1(0)

(ps)

R2(0)

(ps)

1.020
1.020
1.021
1.022
1.022
1.025
1.030
1.036
1.041
1.073
1.131
1.193
1.260

0.22±0.01
0.06±0.01
0.14±0.09
0.17±0.06
0.13±0.01
0.11±0.08
0.29±0.12
0.15±0.01
0.16±0.01
0.21±0.01
0.21±0.02
0.18±0.02
0.21±0.01

19 ±1
10 ±1
8 ±1
8 ±4
15 ±7
12 ±8
12 ±6
21 ±3
19 ±3
23 ±2
28 ±3
29 ±1
32 ±2

0.06±0.02
0.15±0.10
0.22±0.11
0.17±0.02
0.14±0.06
0.07±0.01
0.05±0.01

240 ±16
55 ±6
35 ±4
32 ±10
50 ±8
61 ±16
99 ±15

a

The viscosities of 100% cyclohexane and ethanol are from CRC Handbook of Chemistry and
Physics, 71st ed.; Lide, D.R., CRC Press: Boca Raton, FL, 1990
The viscosities of each mixed liquid are calculated using the Refutas equation.51

69

125
1 (ps)
2 (ps)

 (ps)

100
75
50
25
0
1

10

100

ethanol concentration (% v/v)

Figure 3.4. Reorientation time constants as a function of solvent system viscosity. The
number listed above each point indicates the amount of ethanol (%v/v) in the solvent
system.

70

From these data it is clear that perylene exhibits a change in its effective rotor shape at an ethanol
concentration between 5.0% and 7.5% (v/v). We use Eqs. 3.3 and 3.4 to determine the Cartesian
components of rotational diffusion constant, D, as a function of ethanol concentration. For
solutions containing more than 7.5% ethanol, since perylene reorients as a prolate rotor, and only
one Cartesian component, Dz, is available. For ethanol concentrations below 7.5%, where there
are two exponential decay components, it is possible to extract both Dz and Dx from these
experimental data. The Dz data, and where possible, Dx along with Dz/Dx ratios are shown for
perylene in ethanol/cyclohexane, in Table 3.2 and Figure 3.5. It is instructive to place these data
in the context of analogous results for other solvent systems.
Previous work has shown that perylene exhibits a single-exponential anisotropy decay in
methanol and ethanol, but is characterized by a two component exponential decay in n-propanol
through n-decanol.7,5 A study by Jas et al.52 showed that perylene in cyclohexane and in 2propanol exhibits a two-component anisotropy decay, and the fast reorientation about the z-axis
results from confinement of the perylene molecule by the solvent. As indicated by Eqs. 3.2 and
3.3, for our experimental conditions a single exponential decay of R(t) is associated with
reorientation as a prolate rotor, and a two component exponential decay indicates reorientation as
an oblate rotor. We find that perylene reorients as an oblate rotor in neat cyclohexane and in the
ethanol-cyclohexane binary system for ethanol concentrations of less than 7.5%. With increasing
ethanol content, the effective rotor shape changes from an oblate ellipsoid to a prolate ellipsoid.

71

Table 3.2. Cartesian components of the rotational diffusion constant, D, and reorientation
time, OR, for perylene in cyclohexane-ethanol mixed solutions.
solvent
cyclohexane %
ethanol %
100
0
99.75
0.25
99.50
0.5
99.25
0.75
99.00
1.00
97.50
2.50
95.00
5.00
92.50
7.50
90.00
10.00
75.00
25.00
50.00
50.00
25.00
75.00
0
100

Dz(GHz)

Dx(GHz)

Dz/Dx

OR

13.1 ±0.4
57.4 ±4.1
34.9 ±5.5
65.3 ±7.3
49.6 ±7.5
32.2 ±3.3
24.7 ±4.8
8.2 ±1.0
8.9 ±1.0
7.4 ±0.6
6.2 ±1.4
5.8 ±0.2
5.3 ±0.3

0.5 ±0.1
5.6 ±2.9
6.1 ±3.0
5.4 ±1.7
3.6 ±1.3
4.0 ±1.2
2.4 ±1.2
-

28.1 ±1.1
9.3 ±3.2
5.9 ±2.1
10.9 ±2.2
9.9 ±2.5
7.6 ±1.5
11.4 ±3.5

36 ±1
16 ±5
13 ±4
10 ±3
16 ±5
20 ±2
21 ±5
21 ±3
19 ±2
23 ±2
28 ±3
29 ±1
31 ±2

72

80
Dz

a

Dx

Di (GHz)

60
40
20
0

1
10
ethanol concentration (%)
15

100

b

Dz/Dx

10

5

0

1
10
ethanol concentration (%)

100

Figure 3.5. (a) Dependence of perylene Dx and Dz on ethanol concentration in the binary
solvent system. (b) Ethanol concentration dependence of perylene Dz/Dx ratio for the same
solvent system.

73

We note that our ability to distinguish between a fast second component and a single decay is
limited by the temporal resolution of the instrument and the signal-to-noise ratio of the data. For
the samples ranging from 0 to 5% ethanol, changes in the local environments formed by the
binary solvent system are characterized primarily by changes in Dz and comparatively smaller
changes in Dx (Table 3.2). These results suggest that the local environment formed by the
solvent system constrains chromophore reorientation in such a way as to impede out-of-plane
rotation. This effect is most pronounced in neat cyclohexane and, with increasing ethanol
concentration, the effect is seen to plateau up to 5% ethanol.
For ethanol concentrations above 5% we observe a discontinuous change in reorientation
behavior. For solvent containing 5% ethanol or less, the Dz/Dx ratio is on the order of 10, and for
ethanol concentrations above 5%, Dz/Dx < 1. Solvent-dependent Dz/Dx ratios are associated with
changes in the nature of the interactions between solvent and solute.7,53 We consider next how
the Cartesian components of the rotational diffusion constant relate to the interactions of the
reorienting moiety with its local environment.
We can determine experimentally some of the Cartesian components of the rotational
diffusion constant, D.

D  13  Dx  Dy  Dz 

(3.4)

D is related to its immediate surroundings through Eq. 5, the modified Debye-Stoke-Einstein
(DSE) equation, 54-58

74

D

k BTS
6Vf

(3.5)

where  is the solution bulk viscosity, V is the solute hydrodynamic volume (225 Å3 for
perylene)59, f is frictional boundary condition term to account for frictional contributions to
solvent-solute interactions, which can range from near zero in the slip limit to unity in the stick
limit depending on the shape of the rotating ellipsoid,56,57 T is the solution temperature, kB is the
Boltzmann constant, and S is a shape factor that accounts for the non-spherical shape of the
solute.55 For the system under investigation here, solvent system-dependent changes in D can, in
principle be related to , f or S. For nonpolar chromophores such as perylene in solvent systems
such as ethanol-cyclohexane, V is constant, and for excitation to the S1, T is constant. Since the
bulk viscosity of the solvent system does not undergo any large-scale changes with solution
composition and the ethanol-cyclohexane binary system is well-behaved,51 S is nominally
constant for our measurements, and the only variable that can account for the observed change in
rotor shape is the frictional interaction term, f. In particular, it appears that the nature of f
changes differently for the different Cartesian axes, suggesting changes in the local organization
of the solvent environment in closest proximity to the chromophore.
In an effort to elucidate more information about the immediate environment of perylene in
the ethanol-cyclohexane binary solvent system, we have investigated the vibrational population
relaxation dynamics.

Given the mode-specificity of the initial relaxation process, such

75

measurements can provide insight into the composition and organization of the immediate
environment of the perylene chromophore in this binary solvent system.
Vibrational Population Relaxation.

Although vibrational population relaxation

measurements and molecular reorientation measurements probe fundamentally different
phenomena, they both depend sensitively on the chromophore local environment.

The

information available from the two measurements is complementary in nature. We focus on the
vibrational population relaxation of the perylene 1375cm-1 ring breathing mode.60 This mode is
degenerate with the terminal CH3 group rocking mode of ethanol bath, and there is no
cyclohexane vibrational mode in close energetic proximity. The measurement of the vibrational
relaxation time, T1, as a function of the solvent system composition can thus provide information
on the proximity of the chromophore to the ethanol terminal CH3 group.

If the ethanol-

cyclohexane binary solvent system is a homogeneous distribution of one solvent in the other then
we would expect a monotonic dependence of the measured T1 time on ethanol concentration.
However, we observe a non-linear trend in the ethanol concentration-dependence of T1 for the
perylene 1375 cm-1 mode, with the discontinuity occurring between 5% and 7.5% ethanol in
cyclohexane. It is interesting to note that this is the same concentration range over which we
observe the discontinuous change in perylene reorientation dynamics.
The extraction of vibrational population relaxation information from stimulated emission
transients has been described in detail before.5,6,35,46,61,62 The chromophore behaves as a strongly
coupled three-level system. The three levels are the v=0 ground electronic state, S0(v=0), the
76

ground vibrational state of interest, S0(v=v), and the v=0 excited electronic state S1(v=0). The
difference between the pump laser frequency (vpump = v0-0) and the probe laser frequency (vprobe =
v0-0 – vvib) is set to 1375 cm-1. This Raman-like pumping scheme allows selective population of a
specific chromophore that exists in a system where there is a degenerate acceptor resonance in
the bath. The transient gain of the probe laser intensity (T/T) as a function of time after
excitation contains information on the vibrational population relaxation dynamics of the final
state (S0(v=1)). For the three-modulation detection scheme we use, the expected form of the signal
is
S (t )



T
k  exp(t /  elec )  1 exp( t / T1 ) 

elec



(3.6)

Where elec is the decay time constant associated with the depopulation of the excited electronic
state according to the spontaneous and stimulated emission, and T1 is the vibrational population
relaxation time constant. The time constant T1 is observed as a buildup in S(t) subsequent to
excitation. In Eq. 5, k is an arbitrary scaling factor, typically on the order of 10-5. As noted
above, for the acquisition of S(t), the polarization of the probe pulse is 54.7°with respect to that
of the pump pulse to eliminate molecular motion contributions to the signal. We show in Fig. 6
the instrument response function and the S(t) for 10-5 M perylene in the 99.0 % cyclohexane-1.0 %
ethanol binary solvent system. We have summarized the experimental T1 data as a function of
solvent system composition in Table 3.3 and Figure 3.7. The solvent-dependent change of
rotational diffusion rotor shape and the onset of the change in T1 both occur at a solution ethanol
77

concentration between 5.0% and 7.5%. For the solvent systems with less than 7.5% ethanol, we
measure T1 values in the range of 25 to 45 ps and for higher ethanol concentrations we observe
T1 values on the order of 5 to 13 ps. Since the cyclohexane molecule does not have a vibrational

1.0

T/T (normalized)

0.8
0.6
0.4
0.2
0.0
0

100

200

300

400

500

600

700

delay time (ps)
Figure 3.6. Vibrational population relaxation data (probe polarized at 54.7°with respect to
the pump) and instrument response function for perylene in 99.0 % cyclohexane-1.0 %
ethanol solution. The data and response function are normalized for presentation. These
data are typical of the T1 measurements reported here.

78

Table 3.3. Vibrational population relaxation times, T1, for perylene in cyclohexane-ethanol
mixed solutions.
solvent
cyclohexane
ethanol %
%
100
0
99.75
0.25

T1 – 1375 cm-1
35 ±5
31 ±10

99.50

0.5

36 ±6

99.25

0.75

26 ±10

99.00

1.00

27 ±7

97.50

2.50

44 ±10

95.00

5.00

32 ±6

92.50

7.50

5 ±1

90.00

10.00

5 ±1

75.00

25.00

8 ±2

50.00

50.00

11 ±2

25.00

75.00

13 ±1

0

100

9 ±1

79

mode in close energetic proximity to the 1375cm-1 perylene ring breathing mode, we assert that
direct vibrational energy transfer from perylene to cyclohexane is significantly less efficient than
it is for transfer between perylene and ethanol. The difference in T1 times for these two ethanol
concentrations reflects (at least) two different local environments that are experienced by
perylene chromophore, where for the ca. 5% and lower concentrations of ethanol, its terminal
methyl group is measurably more distant than it is for systems with higher ethanol concentrations.
The OR and T1 data point to there being a discontinuous change in the short range
organization of the binary solvent system for ethanol concentrations between 5.0% and 7.5%,
and that the change occurs in such a way as to cause the local environment sensed by perylene to
become more polar. That finding is not necessarily surprising, but the discontinuous nature of
the change in solute-solvent interactions is not consistent with a smooth change in the average
distance between perylene and ethanol. The similar nonlinear behavior of solvent-mediated
vibrational energy relaxation is seen in benzene and benzyl alcohol binary mixtures. 63 It is
useful to note that the range over which our measurements interrogate the local environment are
different. Chromophore orientational relaxation is mediated by intermolecular interactions that
are either dipole-induced dipole or induced dipole-induced dipole in nature. In either event the
interactions scale as r-6.

The vibrational relaxation T1 measurements are mediated by

quadrupole-dipole intermolecular interactions because a symmetric stretch of the (D2h) perylene
chromophore is the donor vibrational mode and the ethanol methyl group rocking mode is the
acceptor. Accordingly, the intermolecular interactions that mediate vibrational energy transfer
80

scale as r-8.34 The change in the local environment of perylene must be substantial to give rise to
the same ethanol concentration dependence for these two relaxation phenomena.
The body of data presented here suggests that at low ethanol concentrations (less than 5%)
the chromophore is solvated preferentially by cyclohexane, and that after a threshold
concentration of ethanol is achieved, there is a change in the local environment. If we were to
consider this result in the context of local organization, it is instructive to consider the length
scales of such an effect. In a homogeneous solution, a 5% ethanol solution would correspond to
an average spacing between the chromophore and an ethanol molecule of ca. 7.7 Å, and for a 7.5%
homogeneous solution, the average spacing would be ca. 6.8 Å. It is not plausible to suggest that
such a change in average distance could give rise to a factor of ca. 5 change in the measured T1
time. At such intermolecular distances and given the comparatively weak transition moments
that characterize molecular vibrations, the probability of a resonant energy transfer event would
be small in both cases and would be expected to change by a factor of two to three. There must
thus be a change in the organization of the solvent system that enriches the perylene local
environment in ethanol well beyond what its bulk concentration value would imply.

81

60
50

T1(ps)

40
30
20
10
0
1

10

100

ethanol concentration (% v/v)
Figure 3.7. Vibrational population relaxation times, T1, for the perylene 1375 cm-1
symmetric ring breathing mode as a function of solution ethanol content.

82

3.4 Conclusions
We have reported on the orientational relaxation and vibrational population relaxation
dynamics of perylene in the ethanol-cyclohexane binary solvent system. The two measurements
sense different but complementary aspects of the chromophore local environment, and for both
measurements we observe a discontinuous change in the environment of the chromophore for
ethanol concentrations between 5% and 7.5% (v/v). The reorientation data demonstrate that for
low concentrations of ethanol the chromophore is constrained to reorient primarily about the axis
normal to the plane of its π system. For higher ethanol concentrations, motion around an axis
that lies in the chromophore π plane dominates, suggesting an environment less confined by the
solvent. The T1 data show that at the same time that the chromophore is more able to reorient
about an in-plane axis, its proximity to ethanol terminal methyl groups is increased. These two
pieces of information suggest that for ethanol concentrations above ca. 5%, the perylene local
environment becomes dominated by ethanol, and the cyclohexane is displaced. This behavior is
consistent with the steady state absorbance and emission data, showing a change in the 0-0 band
frequency in this same range. Taken as a whole, the data point to the formation of regions of
relatively high ethanol concentration once enough ethanol is present to allow for their formation.
An unresolved issue is whether the chromophore plays a role in nucleating such local high
concentration regions of ethanol in cyclohexane, and this is an issue that is under investigation.

83

REFERENCES

84

REFERENCES

(1)
Flynn, G. W.; Parmenter, C. S.; Wodtke, A. M. The Journal of Physical Chemistry 1996,
100, 12817.
(2)

Rapp, D.; Kassal, T. Chemical Reviews 1969, 69, 61.

(3)
Chen, H.; Wen, X.; Guo, X.; Zheng, J. Physical Chemistry Chemical Physics 2014, 16,
13995.
(4)

Sander, W.; Huib, J. B. Nature 1999, 402, 507.

(5)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.

(6)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 6785.

(7)

Jiang, Y.; Blanchard, G. J. The Journal of Physical Chemistry 1994, 98, 6436.

(8)

Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 9411.

(9)
Dutt, G. B.; Doraiswamy, S.; Periasamy, N.; Venkataraman, B. The Journal of Chemical
Physics 1990, 93, 8498.
(10)
30.

Heitz, M. P.; Horne, J. C.; Blanchard, G. J.; Bright, F. V. Applied Spectroscopy 1997, 51,

(11)

Oxtoby, D. W. Annual Review of Physical Chemistry 1981, 32, 77.

(12)

Zhang, B.; Stratt, R. M. The Journal of Chemical Physics 2012, 137.

(13)

Benjamin, I. The Journal of Physical Chemistry B 2006, 110, 9375.
85

(14)

Iwaki, L. K.; Dlott, D. D. The Journal of Physical Chemistry A 2000, 104, 9101.

(15)

Harris, C. B.; Smith, D. E.; Russell, D. J. Chemical Reviews 1990, 90, 481.

(16) Namboodiri, M.; Kazemi, M. M.; Zeb Khan, T.; Materny, A.; Kiefer, J. Journal of the
American Chemical Society 2014, 136, 6136.
(17) Miyake, Y.; Hidemori, T.; Akai, N.; Kawai, A.; Shibuya, K.; Koguchi, S.; Kitazume, T.
Chemistry Letters 2009, 38, 124.
(18)

Turton, D. A.; Wynne, K. The Journal of Physical Chemistry B 2014, 118, 4600.

(19) Bulovic, V.; Shoustikov, A.; Baldo, M. A.; Bose, E.; Kozlov, V. G.; Thompson, M. E.;
Forrest, S. R. Chem. Phys. Lett. 1998, 287, 455.
(20) CRC Handbook of Chemistry and Physics, 71st Edition; Lide, D. R., Ed.; CRC Press:
Boca Raton, FL, 1990-1991.
(21)

Analytical and Bioanalytical Chemistry 2004, 378, 26.

(22) Stokes, R. H. Journal of the Chemical Society, Faraday Transactions 1: Physical
Chemistry in Condensed Phases 1977, 73, 1140.
(23)

Scatchard, G.; Satkiewicz, F. G. Journal of the American Chemical Society 1964, 86, 130.

(24) Yoshimori, A.; Day, T. J. F.; Patey, G. N. The Journal of Chemical Physics 1998, 109,
3222.
(25) Vaisman, I. I.; Berkowitz, M. L. Journal of the American Chemical Society 1992, 114,
7889.
(26) Tokuhiro, T.; Menafra, L.; Szmant, H. H. The Journal of Chemical Physics 1974, 61,
2275.
86

(27)

Dutt, G. B. The Journal of Chemical Physics 2000, 113, 11154.

(28)

Gardecki, J. A.; Maroncelli, M. Chemical Physics Letters 1999, 301, 571.

(29)

Alavi, D. S.; Waldeck, D. H. The Journal of Physical Chemistry 1991, 95, 4848.

(30)

Dutt, G. B.; Doraiswamy, S. The Journal of Chemical Physics 1992, 96, 2475.

(31)

Laria, D.; Skaf, M. S. The Journal of Chemical Physics 1999, 111, 300.

(32)

Pal, A.; Dass, G. Journal of Molecular Liquids 2000, 84, 327.

(33) Templeton, E. F. G.; Kenney-Wallace, G. A. The Journal of Physical Chemistry 1986, 90,
2896.
(34) Gray, C. G.; Gubbins, K. E. Theort of Molecular Fluids, Vol. 1. Fundamentals; Oxford
Science, 1984.
(35)

Blanchard, G. J. Rev. Sci. Instrum. 1996, 67, 4085.

(36) Kasajima, T.; Akimoto, S.; Sato, S.-i.; Yamazaki, I. The Journal of Physical Chemistry A
2004, 108, 3268.
(37) Kobayashi, I.; Nagao, S.; Terazima, M.; Kimura, Y. Journal of Physics: Conference
Series 2010, 215, 012089.
(38)

Karpovich, D. S.; Blanchard, G. J. The Journal of Physical Chemistry 1995, 99, 3951.

(39) Józefowicz, M. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy
2008, 71, 537.
(40)

Von Jena, A.; Lessing, H. E. Chemical Physics Letters 1981, 78, 187.
87

(41)

Sanders, M. J.; Wirth, M. J. Chemical Physics Letters 1983, 101, 361.

(42)

Blanchard, G. J.; Cihal, C. A. The Journal of Physical Chemistry 1988, 92, 5950.

(43)

Blanchard, G. J. Analytical Chemistry 1989, 61, 2394.

(44) Hartman, R. S.; Alavi, D. S.; Waldeck, D. H. The Journal of Physical Chemistry 1991, 95,
7872.
(45)

Brocklehurst, B.; Young, R. N. The Journal of Physical Chemistry 1995, 99, 40.

(46)

McCarthy, P. K.; Blanchard, G. J. Journal of Physical Chemistry 1996, 100, 5182.

(47)

Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 9328.

(48) Hay, C. E.; Marken, F.; Blanchard, G. J. The Journal of Physical Chemistry A 2010, 114,
12875.
(49) Qiu, C.; Blanchard, G. J. Spectrochimica Acta - Part A: Molecular and Biomolecular
Spectroscopy 2013, 110, 383.
(50)

Chuang, T. J.; Eisenthal, K. B. The Journal of Chemical Physics 1972, 57, 5094.

(51) Centeno, G.; Sánchez-Reyna, G.; Ancheyta, J.; Muñoz, J. A. D.; Cardona, N. Fuel 2011,
90, 3561.
(52) Jas, G. S.; Larson, E. J.; Johnson, C. K.; Kuczera, K. The Journal of Physical Chemistry
A 2000, 104, 9841.
(53) Bauer, D. R.; Brauman, J. I.; Pecora, R. Journal of the American Chemical Society 1974,
96, 6840.
(54)

Zwanzig, R.; Harrison, A. K. The Journal of Chemical Physics 1985, 83, 5861.
88

(55)

Perrin, F. J. Phys. Radium 1936, 7, 1.

(56)

Hu, C. M.; Zwanzig, R. The Journal of Chemical Physics 1974, 60, 4354.

(57)

Youngren, G. K.; Acrivos, A. The Journal of Chemical Physics 1975, 63, 3846.

(58)

Journal of the Society of Chemical Industry 1929, 48, 1036.

(59)

Edward, J. T. Journal of Chemical Education 1970, 47, 261.

(60)

Bado, P.; Wilson, S. B.; Wilson, K. R. Review of Scientific Instruments 1982, 53, 706.

(61)

Blanchard, G. J. Top. Fluoresc. Spectrosc. 1997, 5, 253.

(62)

Hambir, S. A.; Jiang, Y.; Blanchard, G. J. J. Chem. Phys. 1993, 98, 6075.

(63) Jones, B. H.; Huber, C. J.; Massari, A. M. The Journal of Physical Chemistry A 2013, 117,
6150.

89

CHAPTER 4
SPECTROSCOPIC STUDIES ON PREFERENTIAL SOLVATION IN THE
CYCLOHEXANE / N-BUTANOL BINARY SOLVENT SYSTEM†

4.1 Introduction
Molecular interactions in the liquid phase mediate numerous phenomena, including
chemical reaction(s), catalytic processes and chemical separations.1-4 Understanding the details
of these intermolecular interactions in the liquid phase has posed an ongoing challenge due to
their transient nature.5-7 Because of that, the experimental techniques that have provided the
greatest amount of information have been those that can interrogate local organization on short
time scales, and there is a large literature extant based on time-resolved spectroscopic studies of
solution phase dynamics and organization.
Studying binary solvent systems has proven to be a somewhat greater challenge than neat
liquids, with the fundamental issue of import being the homogeneity of the solution phase at the
molecular scale.8-10 There are many instances where local organization and/or heterogeneity in
liquid phase systems has given rise to dynamic behavior at odds with that expected for a
homogeneous medium. Studies using steady-state and time-resolved spectroscopic methods,
theoretical treatments and molecular dynamics simulations have been performed to understand
solute molecular motion in binary solvent systems.11-20 We have reported recently on a binary
solvent system that is anomalous by virtue of its constituents.10 The solvents ethanol and

†

The work in this chapter has been published entitled “Evidence for Preferential Solvation in the
Cyclohexane / n-Butanol Binary Solvent System” on The Journal of Physical Chemistry B, 2015,
119 (5), 1986.
90

cyclohexane are known to be miscible in all proportions, despite the fact that ethanol is a polar
molecule and cyclohexane is nonpolar.21,22 In that work we found that there was a discontinuous
change in the nature of the solution phase organization between 5% and 7.5% ethanol (v/v),
suggesting molecular scale heterogeneity in this system.

The discontinuous change in

organization was detected through both the vibrational population relaxation and rotational
diffusion dynamics of perylene dissolved in a series of ethanol/cyclohexane binary solvent
systems with varying component ratios.
Among the issues that were left unresolved by the previous work was the detailed role of
the alcohol in mediating local organization. It appeared, in particular, that the alcohol terminal
methyl groups were organized in a manner that allowed for closer proximity to the perylene
chromophore than would be possible based simply on their relative concentrations in solution.
In an effort to determine whether or not such local organization could be altered or minimized in
different cyclohexane-alcohol systems, we have examined the cyclohexane/n-butanol binary
solvent system. Because of the greater organic character of the alcohol, it is possible that
cyclohexane/n-butanol may be more homogeneous on short length scales. We report here on the
rotational diffusion and vibrational energy relaxation dynamics of perylene in a series of
cyclohexane/n-butanol binary mixtures. These data show that the cyclohexane/n-butanol binary
solvent system also exhibits a composition-dependent local organization, and that this molecular
organization is different than that seen for the cyclohexane/ethanol system, providing new
experimental evidence and insight into nanometer-scale compositional heterogeneity.

91

4.2 Experimental
Chemicals and Sample Handling. Perylene (99+ %) was purchased from Sigma-Aldrich
and used as received. n-Butanol (99+ %, ACS Spectrophotometric grade) and cyclohexane
(99+ %) were purchased from Jade Scientific in the highest purity grade available. All binary
solutions containing 10-5 M in perylene were made immediately prior to measurement and stored
in the dark. Pump-probe measurements were performed with the sample flowing through a
quartz flow cell connected to a magnetic gear pump (Micropump®, IDEX Health & Science,
LLC) to minimize thermal lensing contributions to the signal and also maintain temperature
control. All measurements were performed at 20°±1°C.
Steady-State Spectroscopy.

Excitation and emission spectra of perylene in

cyclohexane/n-butanol binary solvent systems of controlled composition were acquired using a
Jobin-Yvon Spex Fluorolog 3 spectrometer. For the excitation spectra, the excitation wavelength
was scanned from 300 nm to 450 nm with emission set to 460 nm. Emission spectra were
collected using the same instrument, exciting at 410 nm and collecting emission between 420 nm
and 600 nm. For all measurements, excitation and emission spectral resolution was 0.1 nm.
These data were used to determine the spectral origin and the pump and probe wavelengths for
perylene in each binary solvent system. The excitation and emission spectra for perylene in a
95/5% cyclohexane/n-butanol solution are presented in Figure 4.1. The spectra shown have been
normalized and are the average of at least three individual measurements.
Laser System. A pump-probe laser spectrometer was used for time-resolved anisotropy
decay and vibrational population relaxation measurements. This instrument and measuring
principle have been described in detail in chapter 2 and 3.10 For the measurements reported in
this chapter, the pump laser was set to specific wavelengths between 436.7 nm and 437.1 nm,

92

depending on the 0-0 transition energy of perylene in the binary solvent system under
examination (Figure 4.2). The probe laser wavelength was set to specific wavelengths between
464.6 nm and 465.1 nm, shifted 1375 cm-1 from the origin to access the chromophore symmetric
ring breathing mode. The same pump and probe laser wavelengths were used for both the
vibrational population relaxation and rotational diffusion measurements. For the reorientation
measurements, time-resolved polarized stimulated emission data were collected at polarizations
of 0° and 90° with respect to the polarization of the pump laser. For vibrational population
relaxation measurements the probe polarization was set to 54.7° with respect to the pump
polarization, removing molecular motion contributions to the signal. The time constants reported
in this work represent the average of at least six data acquisitions cycles, and each acquisition is
the average of at least three individual time-scans.

93

Linear response (a.u.)
300

350

400

450

500

550

600

wavelength (nm)
Figure 4.1. Excitation and emission spectra of perylene in 5.0 n-butanol / 95.0%
cyclohexane (v/v). Intensities have been normalized.

94

4.3 Results and Discussion
Previous studies of binary solvent systems have provided evidence for molecular-scale
heterogeneity,10 but to this point there is little in the way of a unifying understanding of liquid
phase dynamics and organization. This fact is the result of the complex, short persistence-time
interactions between solvent molecules, and an important means of understanding such
phenomena is to evaluate changes in solution phase behavior as a function of the chemical
identities and amounts of species present.25-28 The purpose of this work is to investigate the
cyclohexane/n-butanol binary solvent system and compare the results presented here to those
reported previously for the cyclohexane/ethanol system. We start with a discussion of our results
for cyclohexane/n-butanol.
Perylene was used here because of its well-understood spectroscopic properties, as
described in the previous chapters.23-25 This chromophore does not exhibit large spectral shifts in
response to the polarity of its local environment, primarily because of its non-polar nature. For
chromophores such as perylene, where no anomalous spectroscopic behavior is seen, the spectral
shifts of the origin tend to be subtle, but they are meaningful. We show in Fig. 1 the normalized
excitation and emission spectra of perylene in a 95% cyclohexane / 5% n-butanol (v/v) solution.
The quasi-mirror image excitation and emission spectra show that vibronic coupling does not
play a major role in the spectral response.26 These spectra are typical of those seen in all of the
solutions investigated here, but the origin of perylene is seen to shift with solvent system
composition (Figure 4.2a). This finding shows that the overall polarity of the chromophore local
environment changes with solvent system composition. The Stokes shift of perylene (Figure
4.2b) likewise changes slightly as a function of solvent system composition. Interestingly, the
monotonic but nonlinear trend seen in these data, i.e. a spectral blue-shift and increasing Stokes

95

shift with increasing n-butanol content, is different than that seen for cyclohexane/ethanol, where
a red-shift is seen for increasing ethanol content.10 These data suggest that the nature of alcohol
organization in the proximity of the chromophore is different for ethanol than for n-butanol,
which could possibly due to the organic character of the alcohol molecules. The increase in the
Stokes shift with butanol content (Figure 4.2b), combined with the shift being dominated by
changes in the emission band position(Figure 4.3) suggests that the excited state is stabilized to a
greater extent with increasing butanol concentration, consistent with the interaction of the
chromophore with the aliphatic component of n-butanol. The data shown in Figure4.2a are
necessary for performing the vibrational population relaxation measurements (vide infra), but
there is very little detailed information on local environment provided by the subtle shift alone.
The rotational diffusion behavior of perylene is known to be sensitive to local environment and
we consider this body of information next.
Rotational diffusion. Rotational diffusion studies provide useful information on solventsolute interactions and local organization.27-32 The orientational relaxation dynamics of perylene
have been examined extensively, both experimentally and theoretically.24,25,33-36

For time-

resolved measurements, the quantity of interest is R(t), the induced orientational anisotropy
decay. This function is the normalized difference in intensity of the polarized (stimulated)
emission transients,

R (t ) 

I || (t )  I  (t )
I || (t )  2 I  (t )

(4.1)

Where I||(t) and I(t) are the polarized emission transients acquired using the instrument
described above.

It is the functional form of R(t) that provides information on the local

environment of the chromophore. Chuang and Eisenthal have derived expressions relating the

96

a

0-0 transiton wavelength (nm)

437.1
437.0
436.9
436.8
436.7

0

20
40
60
80
n-butanol concentration (% v/v)

b

230

Stokes shift (cm-1)

100

220
210
200
190
180

0

20
40
60
80
n-butanol concentration (% v/v)

100

Figure 4.2. (a) Perylene 0-0 transition wavelength and (b) fluorescence Stokes shift as a
function of n-butanol concentration in n-butanol-cyclohexane binary solvent system.

97

440.0

band maximum (nm)

emission

439.5
439.0
438.5
435.5
absorption

435.0
434.5

0

20

40

60

80

100

n-butanol concentration (% v/v)

Figure 4.3. Absorption and emission band maxima as a function of binary solvent system
composition.

98

functional form of R(t) to the angle between the chromophore excited and emitting transition
moments and to the Cartesian components of its rotational diffusion constant.37

Because

perylene is of D2h point group symmetry, it is possible to define Cartesian axes that are
coincident with symmetry axes. For perylene, the long axis in the -system plane is designated
x, the short in-plane axis is y and the axis normal to the -plane is z. It is known that the S1 
S0 transition in perylene is x-axis polarized. With these assignments, the relationship between
the functional form of R(t) and the relative values of the Cartesian components of the rotational
diffusion constant, D = ⅓(Dx+Dy+Dz) can be evaluated. For a prolate rotor, where Dx ≠ Dy = Dz,

R(t )  0.4 exp(6Dz t )

(4.2)

And for an oblate rotor, where the unique axis is z, (Dz ≠ Dx = Dy),

R(t )  0.1exp((2Dx  4Dz )t )  0.3 exp(6Dxt )

(4.3)

We show in Figure 4.4a an example of the experimental quantities I||(t) and I(t) and in
Figure 4.4b the resulting R(t) function. The results of fits of either Eq. 4.2 or Eq. 4.3 to the
experimental R(t) data as a function of binary solvent system composition are given in Table 4.1.
For neat cyclohexane and neat n-butanol we observe two-component R(t) decays, indicating
reorientation as an oblate rotor (Eq. 4.3). For this rotor shape it is possible to extract both the x
(= y) and z components (Table 4.2). Perylene has been investigated previously in those solvents
and the data we report here are consistent with those earlier reports.24,38 Specifically, perylene is
confined to reorient faster about its z-axis than its x-axis in neat n-butanol and neat cyclohexane.
We attribute this confinement to solvent intermolecular forces, resulting in short range, transient
organization that mediates rotation about the perylene x-axis. For neat n-butanol the ratio of
Dz/Dx = 8.8 and for cyclohexane Dz/Dx = 28. Despite this apparent difference, the values of Dz
for both of these solvents are quite similar, 13.1 ± 0.4 GHz in cyclohexane, 12.2 ±0.5 GHz in n99

a

1.0

T/T (a.u.)

0.8
0.6
0.4
0.2
0.0
0

100

200

300 400
time (ps)

500

600

b

0.2

R(t)

700

0.1

0.0
0

100

200

300 400
time (ps)

500

600

700

Figure 4.4. (a) Experimental time scans I||(t) and I(t) for perylene in 75% n-butanol / 25%
cyclohexane, along with the instrumental laser cross-correlation function. The signal I||(t) is
blue, I(t) is red, and the instrument response function is black. (b) Anisotropy decay
function, R(t)=R(0)exp(-t/OR). The fitted curve is shown in red. The fitted results are
given in Table 4.1.

100

butanol, and it is differences in Dx for these two systems that is most prominent, with Dx = 0.5 ±
0.1 GHz in cyclohexane and 1.8 ± 0.2 GHz in n-butanol. We believe the greater confinement
imposed by cyclohexane is consistent with the shape of the solvent molecule, despite the
substantially higher bulk viscosity that characterizes n-butanol. In other words, viscosity of nbutanol is mediated to a significant extent by the presence of the –OH functionality, and our
results suggest that the perylene chromophore interacts much more strongly with the solvent
aliphatic chains. This assertion is consistent with the vibrational population relaxation data (vide
infra).
In contrast to the findings for the neat solvents, all of the cyclohexane/n-butanol binary
systems examined, ranging from 1% n-butanol to 75% n-butanol, exhibited single exponential
R(t) decay functionality (Tables 4.1 and 4.2). These data are consistent with Eq. 4.2 and, for
such decays, only Dz is able to be determined. Despite this limitation, the implication of these
findings is that the unique axis of rotation has changed from z to x, resulting from a change in the
local solvent molecular arrangement around the chromophore. For all binary solvent systems it
appears that the value of Dz is smaller than it is for the neat solvents by a factor of ~2. This
result indicates that the presence of n-butanol in cyclohexane disrupts the organization
achievable in the neat solvent, and that the local environment formed by this binary solvent
system appears to be independent of the bulk viscosity of the binary solvent system.
This latter finding is important not only because of the change of effective rotor shape but
also because it is not consistent with the predictions of the modified Debye-Stokes-Einstein
model (Eq. 4.4), which is known to be reasonably accurate for the calculation of solution phase
rotational diffusion time constants.39,40

 OR  6 D 1 

Vf
k BTS

101

(4.4)

In this model the rotational diffusion constant is related to the thermal energy in the system (kBT),
the shape (S) and volume (V) of the solute, the boundary condition for frictional interactions
between solvent and solute (f) and the bulk viscosity of the solvent system, (). For perylene, V
= 225 Å3, S = 0.69, T = 293 K, and it is the quantities f and  that may change with the
composition of the binary solvent system.

41,42,43,44

Using this model and the experimentally

determined values of D, we can obtain values for the local viscosity, , sensed by the reorienting
perylene molecule in the binary solvent systems.
The bulk viscosity of the cyclohexane/n-butanol binary system is estimated based on the
Refutas equation,45 with calculated results in good agreement with literature and measured
experimental values. The expected range of viscosities is between  ~ 1 cP and  ~ 3 cP.
Interpreting the experimental anisotropy decay data in the context of Eq. 4.4 reveals that
the local environment experienced by the chromophore is, in all cases, comparatively nonpolar,
and is dependent on the composition of the binary solvent system. We show in Figure 4.5a the
dependence of OR on solvent system bulk viscosity.

If the chromophore were in a polar

environment, the frictional contributions to intermolecular interactions would be described in the
“stick” limit, where f = 1. For reorientation in nonpolar environments, Hu and Zwanzig have
described so-called slip limit behavior, which accounts for the characteristically weaker
intermolecular interactions.43

Modeling slip limit reorientation requires knowledge of the

chromophore aspect ratio (S) and whether it reorients as a prolate or oblate rotor. As noted above,
for perylene S = 0.69, and in the slip limit, f = 0.085 for an oblate rotor and f = 0.081 for a prolate
rotor.

The data shown in Figure 4.5a indicate that the experimentally observed perylene

reorientation behavior lies between the stick and slip limits, and that the frictional interactions
between the chromophore and the solvent system change with solvent identity. We show in Fig.
102

4b the experimentally derived values for f as a function of n-butanol concentration. Two points
are clear from these data. The first is the frictional interaction term decreases with increasing
butanol concentration, and this finding is consistent with perylene interacting preferentially with
the n-butanol aliphatic chains, and not the hydroxyl functionality. The second feature of these
data is that there is a change in the slope of the f vs. butanol concentration dependence in the
vicinity of 5% to 7.5% n-butanol in cyclohexane. This finding is consistent with fluorescence
lifetime (Figure 4.6, Table 4.3) and vibrational population relaxation data (vide infra) and points
to there being a fundamental change in the nature of the interactions between perylene and the
solvent system.

Such a change is not consistent with a uniform distribution of solvent

components in solution. To gain further insight into this composition-dependent solvent local
environment we require a different means of interrogation, which we consider next.

103

Table 4.1. Reorientation times and zero-time anisotropies for perylene in cyclohexanebutanol mixed solutions

cyclohexane
%
100
99.00
97.50
95.00
92.50
90.00
75.00
50.00
25.00
0

solvent
n-butanol
%
0
1.00
2.50
5.00
7.50
10.00
25.00
50.00
75.00
100

(cP)a

R1(0)

1 (ps)

R2(0)

 2 (ps)

1.020
1.053
1.101
1.180
1.257
1.332
1.738
2.266
2.655
2.950

0.22±0.01
0.13±0.01
0.14±0.01
0.19±0.01
0.17±0.01
0.20±0.03
0.15±0.01
0.15±0.01
0.19±0.01
0.13±0.02

19 ±1
27 ±2
28 ±1
24 ±1
23 ±1
25 ±1
31 ±2
34 ±3
36 ±2
17 ±3

0.06±0.02

240 ±16

0.09±0.02

88 ±13

a

The viscosities of 100% cyclohexane and n-butanol are from CRC Handbook of Chemistry and
Physics, 71st ed. The viscosity of the binary solvent is calculated based on Refutas equation.45

104

Table 4.2. Cartesian components of the rotational diffusion constant, D, and reorientation
time, OR, for perylene in cyclohexane/n-butanol mixed solutions.

solvent
cyclohexane
n-butanol
%
%
100
0
99.00
1.00
97.50
2.50
195.00
5.00
92.50
7.50
90.00
10.00
75.00
25.00
50.00
50.00
25.00
75.00
0
100

Dz(GHz)

Dx(GHz)

Dz/Dx

6D-1 (ps)

13.1 ±0.4
6.2 ±0.3
6.0 ±0.2
7.1 ±0.3
7.3 ±0.3
6.9 ±0.4
5.2 ±0.4
4.9 ±0.4
4.6 ±0.4
12.2 ±0.5

0.5 ±0.1
1.8 ±0.2

28.1 ±1.1

36 ±1
26 ±1
28 ±1
23 ±1
23 ±1
25 ±1
31 ±2
34 ±3
36 ±2
32 ±2

105

8.8 ±0.4

a

250
200

DSE stick limit ( f = 1)

OR (ps)

150
100
experimental data

50
0

1.0

DSE slip limit
( fobl=0.081, fpro=0.085)

1.5
2.0
2.5
solution viscosity (cP)

3.0
b

0.45

experimental f

0.40
0.35
0.30
0.25
0.20
0.15
0

20
40
60
80
n-butanol concentration (% v/v)

100

Figure 4.5. (a) Experimental OR values () as a function of solvent system viscosity. The
calculated DSE reorientation times as a function of solvent system viscosity are shown as
lines for the stick limit (f = 1) and slip limit (fobl = 0.081, fpro = 0.085). (b) Frictional
interaction coefficients, f, determined from the experimental data as a function of solvent
system composition. The f values are all greater than the theoretical slip limit values.

106

10000

a

counts

1000

100

10

1
0

2000

6000

8000

10000

time (ps)

3750

fluorescence life time (ps)

4000

b

3500

3250

3000

2750

2500
0

20

40

60

80

100

n-butanol concentration (% v/v)
Figure 4.6. (a) Fluorescence lifetime data for perylene in 50/50 (%v/v) butanol/cyclohexane.
(b) Perylene fluorescence lifetime as a function of n-butanol concentration (%v/v).

107

Table 4.3. Dependence of perylene fluorescence lifetime on n-butanol concentration

solvent
cyclohexane %

Fluorescence lifetime of perylene in the binary

n-butanol

solvent (ps)

%

100

0

2690 ±55

99.00

1.00

2722 ±6

97.50

2.50

3008 ±8

95.00

5.00

3186 ±1

92.50

7.50

3218 ±6

90.00

10.00

3224 ±6

75.00

25.00

3280 ±34

50.00

50.00

3619 ±36

25.00

75.00

3511 ±9

0

100

3582 ±12

108

Vibrational population relaxation. In addition to orientational relaxation measurements,
the vibrational population relaxation dynamics of perylene in neat n-butanol and cyclohexane,
and in a series of cyclohexane/n-butanol binary solvent systems can provide insight into the
chromophore local environment. Information from these measurements is complementary to the
rotational diffusion measurements. However, rather than sensing the effective rotor shape of the
chromophore imposed by the local environment, vibrational population relaxation measurements
sense the (average) proximity and orientation of a solvent vibrational acceptor mode to a
chromophore vibrational donor mode. As such, vibrational population relaxation measurements
are more akin to excitation transport than to rotational diffusion measurements.
For the pump-probe measurements we report here, both vibrational population relaxation
and orientational relaxation information are present in the data. These two types of information
can be separated under the condition that the motional contributions to the experimental signal
are eliminated by acquiring data for the probe polarization angle of 54.7ºwith respect to that of
the pump polarization. Under this condition, the experimental signal contains information only
on the population evolution of the system. Specifically, the probe beam stimulates emission
from the S1v=0 state to a ground state vibration, S0v=1, determined by the frequency of the probe
molecule. Because this process can be viewed in the context of a coupled three-level system,
both stimulated emission and absorption occur, with the relative contribution from each being
determined by the population of the initial and final states of the transition. The time evolution
of the population in each state is determined by the loss channels available to each. The largest
loss channel is associated with vibrational relaxation of the ground state vibration, and the rate of
this relaxation process mediates the time required to achieve a quasi-steady-state population of

109

both the S0v=1 and S1v=0 states. Because of the manner in which the pump and probe pulse trains
are modulated, the signal appears as the difference between two exponential decays,


T
S (t )  k  exp( t /  S1 )  1 exp( t / T1 ) 
 elec



(4.5)

Where k is a scaling factor on the order of 10-5, S1 is the time constant for relaxation of the S1v=0
state, and T1 is the vibrational population relaxation time constant for the S0v=1 state.

A

representative example of these data is shown in Figure 4.7.
There are several important issues regarding this means of measuring T1. The first is that
the Raman-like excitation scheme for the selective population of the S0v=1 state of interest allows
for the excitation of chromophore vibrational modes that are degenerate with specific bath modes.
The measurement of degenerate vibrational energy transfer is not possible using a direct
excitation scheme. The second point is that the nature of the coupling between the donor mode
and the acceptor mode depends on the symmetry of the donor (chromophore) and the acceptor
(solvent molecules).
For perylene, which possesses a center of inversion, the vibrational mode we use as the
donor is the symmetric ring breathing mode, at 1375 cm-1. This vibrational mode is degenerate
with the terminal -CH3 rocking mode of n-butanol, however, there is no corresponding mode in
cyclohexane. Therefore, we are able to monitor the selective transfer of energy from perylene to
n-butanol.

Because the perylene ring breathing mode does not involve modulation of the

molecular dipole moment, it is IR-inactive and Raman active.46 The lowest order multi-pole
moment modulated by the ring breathing mode is the molecular quadrupole moment.

In

comparison, the n-butanol acceptor mode is IR-active, and the coupling between the perylene
donor mode and the n-butanol acceptor mode scales as r-8.47

Monitoring changes in the T1

relaxation time for this donor-acceptor pair as a function of the concentration of n-butanol should
110

provide information on whether or not the n-butanol aliphatic chain is involved in forming the
immediate environment of perylene. The experimental T1 times for the perylene 1375 cm-1 mode
are given in Table 4.4 and are plotted in Figure 4.8. The dependence of T1 on n-butanol
concentration is not consistent with the statistical distribution of n-butanol in cyclohexane. If the
n-butanol molecules were distributed uniformly in the cyclohexane, one would expect to see a
monotonic change in T1 as a function of butanol concentration, and this expectation can be
calculated (Table 4.4). Clearly there is a collisional (or possibly off-resonance) ceiling to the
vibrational relaxation time constant, and this is seen in Table 4.4 to be ca. 35 ps. The n-butanol
concentration-independence of T1 indicates that, even for low concentrations of n-butanol,
perylene experiences an environment that is dominated by n-butanol.

This becomes more

obvious when the n-butanol concentration rise up to 5% v/v. A result of this functional form
requires the n-butanol to be distributed non-uniformly in the cyclohexane.
The vibrational population relaxation and orientational relaxation data for perylene in
cyclohexane/n-butanol solutions provide consistent and complementary information. The most
noticeable similarity between the data sets is the compositional independence of the results, save
for the lowest concentration of n-butanol in cyclohexane. Taken collectively, these data point to
the chromophore residing in an environment dominated by n-butanol for all concentrations,
consistent with a heterogeneous distribution of n-butanol in cyclohexane. We observe behavior
that is approaching concentration independence at 1% v/v n-butanol, and is fully concentration
independent by 2.5% v/v n-butanol. For a perylene concentration of 10-5 M, these butanol
concentrations correspond to 11,000 and 27,000 n-butanol molecules per perylene, respectively.
It is thus not unexpected that the chromophore could reside in a predominantly n-butanol
environment for all systems studied. It is useful to note here that there is a very slight shift in the

111

perylene spectroscopic origin as a function of n-butanol concentration, so the environment
experienced by this chromophore is not determined exclusively n-butanol.
From the rotational diffusion data it appears that the chromophore resides in a nearly
constant environment over a wide range of n-butanol concentrations and that this environment is
characterized by a substantially slip-like boundary condition. The implication of this finding is
that it is the nonpolar aliphatic chain region of the n-butanol environment that dominates the
perylene local environment. While it is inviting to speculate that local organization in the
cyclohexane/n-butanol system may resemble (inverse) micellar structures, we do not have any
independent experimental verification for this speculation at this point. If such were the case, it
could be that any adventitious water in the solvent could support inverse micellar structures,
providing a nominally constant solvation environment for the chromophore. The DSE based
estimate for the local viscosity sensed by perylene appears to be a factor of two to three too low
to support such speculation, however.
It is useful to compare the data we report here with the analogous data for the
cyclohexane/ethanol binary solvent system we have reported on previously.10 In the context of
the rotational diffusion data, it is clear that the presence of butanol is more disruptive to local
organization around the chromophore than ethanol. In the cyclohexane/ethanol system, perylene
is found to reorient as an oblate rotor for ethanol concentrations of up to and including 5%,
whereas for the cyclohexane/n-butanol system the addition of 1% butanol causes the
chromophore to reorient as a prolate rotor. In terms of the T1 data, this trend is matched, with T1
for the perylene 1375 cm-1 mode being ca. 35 ps for ethanol concentrations up to 5%, and T1 for
this same mode diminishing to a ca. 10 ps floor with the addition of 2.5% n-butanol to
cyclohexane. From these data it is clear that the n-butanol terminal methyl group is in closer

112

proximity to the perylene chromophore than the ethanol terminal methyl group under all
circumstances.

While these results are not necessarily surprising, they do confirm the

importance of the amphiphilic nature of n-alcohols in the solvation of nonpolar molecules.

113

1.0

T/T (a.u.)

0.8
0.6
0.4
0.2
0.0
0

100

200

300

400

500

600

700

time (ps)

Figure 4.7. Vibrational population relaxation data (probe polarized at 54.7°with respect to
the pump) and instrument response function for perylene in 97.5% cyclohexane / 2.5% nbutanol solution. The data and response function are normalized for presentation. These
data are typical of the T1 measurements reported here.

114

40

T1 (ps)

30

20

10

0

0

20

40

60

80

100

n-butanol concentration (% v/v)

Figure 4.8. Vibrational population relaxation times, T1, for the perylene 1375 cm-1
symmetric ring breathing mode as a function of solution n-butanol content (%v/v).

115

Table 4.4. Vibrational population relaxation times, T1 in ps, for perylene in cyclohexanebutanol mixed solutions. Calculated T1 times are based on the r-8 dependence of T1
relaxation, assuming a statistical distribution of solvents.

solvent
T1 – 1375 cm-1 (ps)

T1 calculated (ps)

100
99.00

n-butanol
%
0
1.00

35 ±5
18 ±1

-1.94x109

97.50

2.50

12 ±2

4.94x107

95.00

5.00

8 ±1

3.07x106

92.50

7.50

7 ±1

59,600

90.00

10.00

9 ±3

18,300

75.00

25.00

10 ±3

4,380

50.00

50.00

11 ±2

261

25.00

75.00

10 ±2

46

0

100

12 ±2

12

cyclohexane %

116

4.4 Conclusions
We have investigated the cyclohexane/n-butanol binary solvent system by means of
measuring the orientational relaxation and vibrational population relaxation dynamics of
perylene. Our data demonstrate a substantial independence of both types of data on solution
composition, and this finding is consistent with the perylene experiencing a local environment
that is not described well as a homogeneous distribution of one solvent in another. Both sets of
data have shown evidence that, in this binary solvent mixture, the local environment of perylene
is dominated by the presence of n-butanol, resulting a molecular level heterogeneity in the
system.

The extent and details of molecular scale organization within the immediate

environment of the chromophore remains unresolved, and this is the subject of further
investigation.

117

REFERENCES

118

REFERENCES

(1)

Zewail, A. H. The Journal of Physical Chemistry A 2000, 104, 5660.

(2)

Zhang, B.; Stratt, R. M. The Journal of Chemical Physics 2012, 137.

(3)

Miller, D. W.; Adelman, S. A. International Reviews in Physical Chemistry 1994, 13, 359.

(4)
Fujisaki, H.; Straub, J. E. Proceedings of the National Academy of Sciences of the United
States of America 2005, 102, 6726.
(5)

Chen, H.; Wen, X.; Guo, X.; Zheng, J. Physical Chemistry Chemical Physics 2014.

(6)
Flynn, G. W.; Parmenter, C. S.; Wodtke, A. M. The Journal of Physical Chemistry 1996,
100, 12817.
(7)

Harris, C. B.; Smith, D. E.; Russell, D. J. Chemical Reviews 1990, 90, 481.

(8)

Mukherjee, A.; Bagchi, B. The Journal of Physical Chemistry B 2001, 105, 9581.

(9)

Stevenson, S. A.; Blanchard, G. J. The Journal of Physical Chemistry A 2006, 110, 3426.

(10)

Qiu, C.; Blanchard, G. J. The Journal of Physical Chemistry B 2014, 118, 10525.

(11) Yoshimori, A.; Day, T. J. F.; Patey, G. N. The Journal of Chemical Physics 1998, 109,
3222.
(12) Vaisman, I. I.; Berkowitz, M. L. Journal of the American Chemical Society 1992, 114,
7889.
(13) Tokuhiro, T.; Menafra, L.; Szmant, H. H. The Journal of Chemical Physics 1974, 61,
2275.
(14)

Dutt, G. B. The Journal of Chemical Physics 2000, 113, 11154.
119

(15)

Gardecki, J. A.; Maroncelli, M. Chemical Physics Letters 1999, 301, 571.

(16)

Alavi, D. S.; Waldeck, D. H. The Journal of Physical Chemistry 1991, 95, 4848.

(17)

Dutt, G. B.; Doraiswamy, S. The Journal of Chemical Physics 1992, 96, 2475.

(18)

Laria, D.; Skaf, M. S. The Journal of Chemical Physics 1999, 111, 300.

(19)

Pal, A.; Dass, G. Journal of Molecular Liquids 2000, 84, 327.

(20) Templeton, E. F. G.; Kenney-Wallace, G. A. The Journal of Physical Chemistry 1986, 90,
2896.
(21) Bulovic, V.; Shoustikov, A.; Baldo, M. A.; Bose, E.; Kozlov, V. G.; Thompson, M. E.;
Forrest, S. R. Chem. Phys. Lett. 1998, 287, 455.
(22) CRC Handbook of Chemistry and Physics, 71st Edition; Lide, D. R., Ed.; CRC Press:
Boca Raton, FL, 1990-1991.
(23)

Klein, U. K. A.; Haar, H. P. Chemical Physics Letters 1979, 63, 40.

(24)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.

(25)

Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 9411.

(26)

Karpovich, D. S.; Blanchard, G. J. The Journal of Physical Chemistry 1995, 99, 3951.

(27)

Sanders, M. J.; Wirth, M. J. Chemical Physics Letters 1983, 101, 361.

(28)

Blanchard, G. J. Analytical Chemistry 1989, 61, 2394.

(29) Hartman, R. S.; Alavi, D. S.; Waldeck, D. H. The Journal of Physical Chemistry 1991, 95,
7872.

120

(30) Qiu, C.; Blanchard, G. J. Spectrochimica Acta - Part A: Molecular and Biomolecular
Spectroscopy 2013, 110, 383.
(31) Hartman, R. S.; Konitsky, W. M.; Waldeck, D. H.; Chang, Y. J.; Castner, E. W. The
Journal of Chemical Physics 1997, 106, 7920.
(32) Schwartz, M.; Duan, D.; Berry, R. J. The Journal of Physical Chemistry A 2005, 109,
8637.
(33)

Brocklehurst, B.; Young, R. N. The Journal of Physical Chemistry 1995, 99, 40.

(34) Christensen, R. L.; Drake, R. C.; Phillips, D. The Journal of Physical Chemistry 1986, 90,
5960.
(35) Rosales, T.; Xu, J.; Wu, X.; Hodoscek, M.; Callis, P.; Brooks, B. R.; Knutson, J. R. The
Journal of Physical Chemistry A 2008, 112, 5593.
(36)

Xu, J.; Shen, X.; Knutson, J. R. The Journal of Physical Chemistry A 2003, 107, 8383.

(37)

Chuang, T. J.; Eisenthal, K. B. The Journal of Chemical Physics 1972, 57, 5094.

(38) Jas, G. S.; Larson, E. J.; Johnson, C. K.; Kuczera, K. The Journal of Physical Chemistry
A 2000, 104, 9841.
(39)

Debye, P. Polar Molecules; Chemical Catalog Co.: New York, 1929.

(40) Miyake, Y.; Hidemori, T.; Akai, N.; Kawai, A.; Shibuya, K.; Koguchi, S.; Kitazume, T.
Chem. Lett. 2009, 38, 124.
(41)

Edward, J. T. Journal of Chemical Education 1970, 47, 261.

(42)

Perrin, F. J. Phys. Radium 1936, 7, 1.

(43)

Hu, C. M.; Zwanzig, R. J. Chem. Phys. 1974, 60, 4354.

(44)

Youngren, G. K.; Acrivos, A. J. Chem. Phys. 1975, 63, 3846.
121

(45) Centeno, G.; Sánchez-Reyna, G.; Ancheyta, J.; Muñoz, J. A. D.; Cardona, N. Fuel 2011,
90, 3561.
(46)

Blanchard, G. J. Review of Scientific Instruments 1996, 67, 4085.

(47) Gray, C. G.; Gubbins, K. E. Theort of Molecular Fluids, Vol. 1. Fundamentals; Oxford
Science, 1984.

122

CHAPTER 5
PHOSPHOLIPID VESICLES STABILITY AND TEMPORAL VARIATIONS IN
ACYL CHAIN ORGANIZATION‡

5.1 Introduction
Vesicles are a structural motif used widely in the investigation of phospholipid organization
and dynamics in solution.

Synthetic vesicles formed using phospholipids, with or without

cholesterol, have proven to be reproducible with relatively facile control over vesicle dimensions
and composition.1,2

There are a number of practical reasons for the study of vesicles, including

their use as pharmaceutical delivery vehicles3-6 and as model systems.

For any application

using unilamellar phospholipid vesicles, their long-term dimensional and morphological stability
can be an important issue, and that is the focus of the work we present here.
The methods used most widely for forming phospholipid vesicles are extrusion and
sonication.7-11 With either method, the vesicle constituents are mixed prior to vesicle formation
to ensure homogeneity.

In an earlier report we had compared the rotational and translational

diffusion behavior of chromophores contained in vesicles formed by the two methods and we
found that, despite substantial differences in morphology, the molecular and mesoscopic length

‡

The work in this chapter has been published entitled “Phospholipid vesicle stability and
temporal variations in acyl chain organization” on Spectrochimica Acta - Part A: Molecular and
Biomolecular Spectroscopy 2013, 110, 383

123

scale behavior of these bilayer structures was the same.12

This finding was significant because

it meant that results reported in the literature using vesicles formed by both methods could be
compared to one another.

In this work we are interested in a related issue; the long-term

stability of unilamellar vesicles.
To this point, there have been several literature reports that suggest vesicles in aqueous
solution are not stable indefinitely, producing larger structures over time due to spontaneous
vesicle fusion.13,14

Indeed, if true, fresh vesicles must be made daily, and this is the

experimental procedure used most commonly.
The work we report here focuses on two aspects of vesicle stability.
dimensional stability.

The first is

We have used dynamic light scattering (DLS) to measure the average

size and size distribution of unilamellar vesicles as a function of time after formation by
extrusion.

These data show that vesicle fusion does not appear to play a significant role in

changing the average size or size distribution, even over extended periods of time. Rotational
diffusion measurements of perylene contained in the vesicle nonpolar acyl chain region
demonstrate that there are changes in the organization of that region as a function of time after
extrusion.

The lipid acyl chains are seen to become more ordered over time, with the

organization associated with annealing being reversed by re-extrusion of the vesicles. The
re-extrusion process alters the organization of the bilayer acyl chain region, with subsequent
recovery of order being either much slower than after the first extrusion process, or not
measurable.

These data demonstrate that, from a macroscopic perspective, phospholipid
124

vesicles do not appear to fuse when in the aqueous buffer solution we have used (pH ~8), but the
lipid acyl chain region does undergo annealing to form a more ordered local environment
following vesicle formation.

5.2 Experimental
Vesicle preparation.

The phospholipid 1,2-dimyristoyl-sn-phosphatidylcholine (DMPC) in

chloroform was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL) and used as received.
Perylene (99+ %) was purchased from Sigma-Aldrich (Milwaukee, WI) and used without further
purification.

To make a vesicle sample, 10 mg of lipid was mixed with chromophore in 400L

chloroform, and the solvent was evaporated under N2(g).

Aqueous Tris® buffer (Sigma-Aldrich)

was added to each sample to make the lipid concentration 1mg/mL.

The final perylene

concentration in each sample is 10-5 M (0.678 mol%). Tris® buffer (10 mM, pH 7.96) was
prepared using purified water from a Milli-Q Plus water purification system (Millipore, Bedford,
MA), and the solution was purged with Ar prior to use.

Following hydration, the

lipid-chromophore mixtures were processed through five freeze-thaw-vortex cycles to ensure
reproducible mixing of the constituents.

Each cycle consisted of freezing the mixture by

immersion in N2(l) for 5 min, then thawing in a 60℃ water bath for 5 min, and vortexing the
sample for 2 min.

After the completion of the freeze-thaw-vortex process, the

lipid-chromophore mixture was extruded using a mini-extruder apparatus (Avanti Polar Lipids,
Inc.), eleven times through a polycarbonate membrane with a nominal 100 nm pore diameter

125

(Whatman).

The 80% DMPC / 20% cholesterol and 70% DMPC / 30% cholesterol vesicles

were prepared using same protocol.

All extrusions were performed at room temperature.

Once the vesicle-containing solutions were extruded, they were kept in a sealed glass vial in the
dark at room temperature.

No further treatment was applied prior to use.

Dynamic light scattering (DLS) measurements.

DLS measurements were performed using

a Malvern Zetasizer instrument operating at 632.8 nm.
mg/mL for all measurements.

The concentration of DMPC is 1

Data were acquired at 20 ±1°C.

Time-resolved fluorescence measurements.

All fluorescence lifetime and anisotropy decay

data were acquired using a time-correlated single photon counting (TCSPC) instrument, as
shown in Figure 5.1.

This instrument has been described in detail before and we recap its

salient features here.15

The light source a CW mode-locked Nd:YVO4 laser (Spectra Physics

Vanguard) that produces 2.5 W average power at 355 nm and 2.5 W average power at 532 nm, at
80 MHz repetition rate with 13 ps pulses.

The outputs of the Nd:YVO4 laser are used to excite

cavity-dumped dye lasers (Coherent 702-2), producing 5 ps pulses.

The repetition rate of the

dye laser is adjustable between 80 MHz and 80 kHz using Gooch and Housego cavity dumping
electronics (4 MHz dumping rate was used in these experiments).

For the experiments reported

here, the dye laser was excited with 355 nm pulses and operated with Stilbene 420 dye (Exciton)
to produce ca. 5 ps pulses at 435 nm.

The pulse from the dye laser is divided into two parts,

with one portion directed to a reference photodiode (Becker & Hickl PHD-400-N), and the other
portion directed to the sample.

Emission is collected using a 40x reflecting microscope
126

objective (Ealing).

The collected emission is separated into components polarized parallel (0°)

and perpendicular (90°) to the vertically polarized excitation pulse using a polarizing cube beam
splitter.

The parallel and perpendicular polarized signal components are detected

simultaneously using two microchannel plate photomultipliers (MCP-PMT, Hamamatsu
R3809U-50), each equipped with a subtractive double monochromator (Spectral Products
CM-112) for wavelength selection.

The detection electronics (Becker & Hickl SPC-132)

produce a ca. 30 ps response function for each detection channel.

Data acquisition, detector

bias, and collection wavelength are all controlled using a LabVIEW® program (National
Instruments) written in-house on a PC.

Samples were measured at 20 ± 1 C after a ten minute

equilibration time.

127

Figure 5.1. Time-Correlated Single Photon Counting (TCSPC) instrument schematic.

128

5.3 Result and Discussion
The goal of the work reported here is to understand the long-term stability of phospholipid
vesicles, both at the molecular-scale and on longer length scales.

DMPC (Figure 5.2) was

chosen for this work because it is a well characterized and structurally simple phosphocholine.
For investigations of the vesicle non-polar region, we have used perylene (Figure 5.2) because it
is a well characterized non-polar chromophore that partitions essentially exclusively into the
vesicle acyl chain region.

The two bodies of data we present below contain complementary

information and we consider them separately before contemplating their collective significance.
It is also important to consider the organization of the lipid bilayers in these experiments.
As noted above, all of our measurements were performed at 20 ± 1°C, a temperature near phase
transitions in DMPC and DMPC/cholesterol systems.

It is known that DMPC exhibits a

gel-to-fluid phase transition at T = 24°C,16 and we have studied the dynamics of perylene above
and below this phase transition temperature.

17,18

The temperature of 20 ± 1°C is sufficiently

far from the phase transition to ensure the data are representative of the gel phase bilayer.

For

systems containing cholesterol, the phase behavior is modified by its presence,16,19 but for all of
the systems we report here, the phase of the system at 20 ± 1°C is not in close proximity to any
phase transitions.16,19

Information available from simulations suggests that the organization of

the cholesterol-containing systems is intermediate between that of the P’ and Lo phases.19
Liposome Mean Diameter and Size Distribution.
any work that utilizes this structural motif.

The stability of vesicles is a key issue in

This is a question that has been examined before,
129

with previous studies suggesting that phospholipid vesicles, especially small sonicated vesicles,
are thermodynamically unstable at temperatures below the lipid phase transition and tend to
aggregate or fuse to form larger unilamellar vesicles during long-term storage.20-23

The

aggregation of small vesicles alters their properties for reasons including curvature, and such
changes can affect the permeability of the vesicles and their bio-distribution in vivo.24

For

larger vesicles, there are indications that they can experience significant shape changes (i.e.
excursions from spherical shape).25

Our choice of 100 nm diameter vesicles is deliberate; larger

than vesicles formed by sonication and smaller than vesicles known to exhibit shape changes.
We also note that there have been efforts to stabilize vesicle structures through the use of
polymerization.26-28 We are concerned with the stability of phospholipid bilayers that have not
been structurally modified by cross-linking reactions.
Dynamic light scattering (DLS) measures the time correlation of the intensity of the
scattered light, and this field-field time correlation function decays exponentially at a rate that
depends on the size of the scattering particles.29
characterized using DLS.

Vesicle size and size distribution can be

We have monitored the diameter and distribution width of extruded

100 nm nominal diameter vesicles made of 100% DMPC, 80% DMPC / 20% cholesterol and 70%
DMPC / 30% cholesterol using DLS.

We have chosen to include cholesterol in some vesicles

because of its reported effect of making phosphocholine bilayers less fluid.30-32 Cholesterol is
implicated in numerous structural and functional capacities, playing an important role in
maintaining vesicle fusion/docking structures.2
130

14:0 DMPC
14:0
DMPC

perylene
perylen

cholesterol
cholestero

e

l

Figure 5.2. Structures of 14:0 DMPC, perylene and cholesterol.

131

For the vesicles we have formed, 1440 hours after initial extrusion there is no discernible
change in their measured mean diameter, regardless of whether or not cholesterol was present
(Figure 5.3 inset).

While it may be tempting to infer a slight change in diameter for 70%

DMPC / 30% cholesterol vesicles as a function of time (Figure 5.3a), we are not willing to
speculate on any trend in these data outside of the experimental uncertainty. The DLS data
shown in Table 1 represent three to six individual determinations for each value.

These data

indicate that vesicle aggregation or fusion does not occur to a measurable extent for our
experimental conditions.
While the size of the vesicles is time invariant over 1440 hours, we note that there is a
composition-dependent trend in their measured diameter (Figure 5.3a), as well as the distribution
width, (Figure 5.3b) increasing with the amount of cholesterol present.

The measured

diameters range from 114 nm for 100% DMPC vesicles to 128 nm for 80% DMPC / 20%
cholesterol vesicles and 138 nm for 70% DMPC / 30% cholesterol vesicles. With values that
are this close, it is fair to consider the uncertainty in the experimental DLS data, and to assess
whether or not these differences can be seen in the experimental distributions (Figure 5.4a) and
correlation functions (Figure 5.4b).
There is indeed a reproducible difference in the experimental data for vesicles of different
compositions, and there are two possible explanations for these data.

The first is that the

observed change in mean diameter is due to the cholesterol in the vesicle mediating the curvature
of the bilayer and thus influencing the diameter of the vesicles. The other explanation is that
132

the presence of cholesterol in the vesicle alters its effective refractive index in such a way as to
produce a different calculated diameter, even though the actual size of the vesicle is constant.
The refractive index of cholesterol has been reported as 1.52233 and the refractive index of
DMPC is taken to be ~1.424.34

Using these values and a Lorentz-Lorenz treatment to estimate

the refractive index of our vesicles,35
n122  1
n12  1
n22  1




1 2
2 2
n122  2
n1  2
n2  2

(5.1)

Where we take component 1 to be DMPC and component 2 to be cholesterol.

We have used

different refractive indices in calculating the DLS average diameter to address this issue (Table
5.2). The refractive index values we used are 1.424 for 100% DMPC, 1.443 for 80% DMPC /
20% cholesterol, and 1.453 for 70% DMPC / 30% cholesterol.

As can be seen from the data in

Table 5.2, the vesicle sizes calculated using different refractive index values cannot account for
the measured composition-dependent variation in vesicle size.

The average diameter of the

vesicles does, in fact, depend on bilayer composition, as is shown in Figure 5.3a andFigure 5.3b.
Molecular Reorientation.
bilayer.

Perylene was chosen as a probe of the acyl chain region of the lipid

Due to its nonpolar, planar structure, it incorporates selectively into the hydrophobic

acyl chain region of DMPC bilayer leaflets, with its location within the phospholipid bilayer
depending on the diameter (curvature) of the vesicle.18,36 Recent work indicated that perylene
moves within the bilayer to be in closest proximity to the acyl chain terminal methyl groups for
vesicles of ca. 1 µm diameter and larger.

For vesicles less than 800 nm in diameter, perylene

133

20

1 hour
24 hours
72 hours
168 hours
1440 hours

intensity

15

a

5

140
average size (nm)

10

0
30

40

50 60 70 8090
100

200

300

400

vesicle diameter (nm)

130
120
110
0

20

40

60

80 100 120 140 160 180

60

peak width (nm)

55

b

50
45
40
35
30
0

20

40

60

80 100 120 140 160 180

time after vesicle formation (hrs)

Qiu and Blanchard Figure 2

Figure 5.3. (a) Measured mean size of vesicles from DLS with time. Inset: DLS size
distribution data for DMPC vesicles as a function of time after formation. (b) Measured
distribution width from DLS of vesicles with time. For both plots, solid circles represent
100% DMPC vesicles, open circles represent 80% DMPC / 20% cholesterol vesicles, and
solid triangles represent 70% DMPC / 30% cholesterol vesicles.
134

Table 5.1. Dynamic light scattering (DLS) data for extruded DMPC and DMPC/cholesterol
vesicles: Measured mean vesicle diameter (d) and distribution width (w) (nm).

Time after
extrusion
(hrs)

100% DMPC

80% DMPC / 20%
cholesterol

70% DMPC / 30%
cholesterol

d (nm)

w (nm)

d (nm)

w (nm)

d (nm)

w (nm)

1

114 ±2

36 ±2

128 ±1

38 ±1

138 ±2

49 ±8

16

114 ±1

36 ±3

128 ±1

38 ±2

140 ±6

46 ±8

24

114 ±1

35 ±1

128 ±2

40 ±3

139 ±5

50 ±4

48

114±1

35 ±2

127 ±1

36 ±3

136 ±3

50 ±3

72

114 ±2

37 ±1

128 ±1

37 ±5

133 ±2

46 ±5

144

113 ±1

34 ±1

127 ±1

36 ±2

132 ±2

37 ±3

168

114 ±1

35 ±2

126 ±1

36 ±2

133 ±1

39 ±6

1200

111 ±1

38 ±3

130 ±8

37 ±1

135 ±2

50 ±2

1440

110 ±1

36 ±2

126 ±2

36 ±2

134 ±1

48 ±2

135

a

intensity

20
15
10
5
0

correlation coefficient

100
1000
particle diameter (nm)

10000

b

1.0
0.8
0.6
0.4
0.2
0.0
0.1

1

10
100
time (s)

1000

10000

Figure 5.4. (a) Experimental size distributions for nominal 100 nm diameter vesicles
comprised of DMPC (solid line), 80% DMPC / 20% cholesterol (dashed line) and 70%
DMPC / 30% cholesterol (dotted line), all acquired within one hour after extrusion to form
the vesicles. (b) Experimental correlation functions associated with the same three
samples.
136

Table 5.2. Dependence of calculated vesicle diameter on refractive index. Uncertainties in
the diameters are ±1 nm.

Refractive index

d (nm)
100% DMPC

1.424
1.443
1.453

113.4
---

d (nm)
80% DMPC / 20%
cholesterol
127.3
126.5
--

137

d (nm)
70% DMPC / 30%
cholesterol
132.2
-131.9

resides preferentially within the acyl chains of one of the bilayer leaflets. 18 That is amongst the
reasons we have chosen to use 100 nm diameter vesicles in this work.17,18
To characterize the reorientation dynamics of perylene, we combined the experimental
polarized fluorescence transients acquired using the TCSPC instrument to generate the induced
orientational anisotropy function, R(t).
R(t ) 

I || (t )  I  (t )
I || (t )  2 I  (t )

(5.2)

Where the quantities I||(t) and I(t) are the time-resolved emission intensities polarized parallel
and perpendicular to the vertically polarized excitation pulse.

We infer information on the

motional dynamics of perylene from the functional form of R(t).

This information is reflective

of the organization of the DMPC bilayer acyl chain region.

The theory for the interpretation of

R(t) is well established,37-43 with the number of exponential decays contained in R(t) and their
time constant(s) being related to the Cartesian components of the chromophore rotational
diffusion constant and the relative orientations of the absorbing and emitting transition dipole
moments.
The anisotropy decay of perylene can be described either as a single component exponential
decay or a two component exponential decay, depending on its immediate environment.
Relating the functional form of the anisotropy decay to the Cartesian components of the
rotational diffusion constant requires assignment of the Cartesian axes and the assumption of a
rotor shape.37

The excited and emitting transition dipole moments of perylene (S1↔S0) lie

138

along the chromophore long in-plane axis, which we assign as the x-axis. The short in-plane
axis is designated y, with the z-axis perpendicular to the chromophore -system plane.

With

these assignments, a prolate rotor (Dx>Dy=Dz) produces a single-exponential R(t) decay
functionality,

R(t )  0.4 exp(6Dz t )

(5.3)

and an oblate rotor (Dz>Dy=Dx) gives rise to a two exponential component anisotropy decay,

R(t )  0.1exp((2Dx  4Dz )t )  0.3 exp(6Dxt )

(5.4)

In this work, we observed two-component decays for all anisotropy decay measurements (Fig. 4),
consistent with perylene reorienting as an oblate rotor in 100 nm diameter DMPC vesicles.18
We extract the Cartesian components of the rotational diffusion constant, D, from the
experimental anisotropy decay data.

We present the values of Dz and Dx(=Dy) derived from the

experimental data as well as the ratio Dz/Dx as a function of time after formation and vesicle
composition in Figure 5.6Figure 5.7Figure 5.8.
The data in Figure 5.6 show measurable changes in Dz and Dx within 624 hours after vesicle
preparation for 100% DMPC vesicles, and the data in Figure 5.7 and Figure 5.8 provide
analogous information for cholesterol-containing vesicles over a time span of 1320 hours.
While the changes in Dx and Dz may appear subtle over the time span of the experiment, these
changes are actually significant and informative.

As can be seen more clearly from the Dz/Dx

ratio (Figure 5.6b), following formation by extrusion, the aspect ratio of the (ellipsoidal) volume
swept out by perylene in 100% DMPC vesicles is on the order of 6.7:1, rising to ~9:1 after
139

several hours of annealing. Re-extrusion at 370 hours reduces the Dz/Dx ratio to ~6.5:1, where
it remains in subsequent measurements.

This finding is interesting for several reasons.

The

first is that the change in aspect ratio following initial extrusion indicates an annealing process.
As acyl chains become more organized, the ability of the chromophore to execute out-of plane
rotation about its x and/or y axes is reduced relative to its ability to rotate about z. This change
could result from either an increase in Dz or a decrease in Dx. The data in Figure 5.6a show that
the change in aspect ratio is associated primarily with a decrease in Dx, while Dz remains
approximately constant.

Such a change in dynamics is consistent with the conversion of

gauche to trans conformers in the acyl chains.

It is thought that, for 100 nm diameter vesicles,

perylene resides in the bilayer, amongst the acyl chains, and the chromophore might be sensing
the onset of annealing in this region.

The fact that there is no discernible change in Dz suggests

that this quasi-lamellar structure created within the bilayer acyl chain region is relatively stable
over time.

Interestingly, following re-extrusion of the vesicles, no change is seen in their

average diameter, but the Dz/Dx ratio is reduced to ~6.5:1, similar to that seen after initial vesicle
formation.

As elapsed time increases following re-extrusion, however, the Dz/Dx ratio does not

appear to increase as it did after initial vesicle formation.

Examination of the D z and Dx data

shows that this change in Dz/Dx is due principally to an increase in Dx, with approximately
constant Dz. The fact that Dx does not decrease upon re-extrusion suggests that one of two
possible explanations is operative. The first is that the system does not anneal to the same
conformational distribution that it had assumed prior to re-extrusion.
140

The direct implication of

0.15

R(t)

0.10

0.05

0.00

-0.05

0

2000

4000

6000

8000

time (ps)

Figure 5.5. Anisotropy decay data for perylene in 100-nm diameter DMPC vesicles at 293 K,
showing decay that is best fit by two exponential components. The fitted line is indicated
as the solid line and the residuals of the fit are shown centered around the zero line. The
fitted function is R(t) = R1(0)exp(-t/1) + R2(0)exp(-t/2), where R1(0) = 0.08 ±0.01, 1 = 421 ±
45 ps, R2(0) = 0.20 ±0.01, 2 = 2358 ±139 ps.

141

Dx, Dz (MHz)

900
800
700
600
500
400
300

a

Dz
re-extrusion

Dx

100
50

0

100

200

300

400

500

600

time (hrs)
12

re-extrusion

b

11
10
Dz/Dx

9
8
7
6
5
4

0

100

200

300

400

500

600

time (hrs)
Figure 5.6. (a) Dependence of perylene Dx and Dz on time elapsed after formation for
nominal 100 nm diameter DMPC vesicles at 293 K. (b) Dependence of Dz/Dx ratio for
these vesicles. Vesicles were re-extruded through 100 nm diameter pore filter at 370
hours.
142

such a finding is that either the annealed vesicle or the re-extruded vesicle represents a
metastable state rather than a thermodynamic minimum for the system.

The other possibility is

that re-extrusion causes a relocation of the chromophore within the vesicles, from primarily
within the acyl chains to primarily within the inter-leaflet gallery.

It is not clear why

re-extrusion of vesicles would lead to a different distribution of chromophores within the acyl
chain region, however.
Another interesting finding from these data is that the addition of cholesterol to the vesicles
alters the reorientation dynamics of perylene and also the time- and extrusion-dependent
variation in Dz and Dx.

For vesicles containing 80% DMPC and 20% cholesterol, we find that

annealing after the initial extrusion is not seen, but when the vesicles are re-extruded (Figure 5.7)
there is a marked increase in Dz, with only a slight increase in Dx. The ratio Dz/Dx (Figure 5.7b)
increases immediately following re-extrusion and returns at long times to a value of ca. 8.
There are several possible explanations for these findings.

Assuming phase separation of

cholesterol and DMPC in the vesicles, it is reasonable to expect that the DMPC-rich acyl chain
regions will experience an annealing process that is qualitatively similar to that seen for 100%
DMPC vesicles. While the time required for any acyl chain annealing to occur will almost
certainly be different for the two vesicles, our data appear to indicate the absence of this process
over an extended time period.

It is possible that the perylene chromophore resides in the

cholesterol-rich regions of the vesicle and is thus insensitive to changes in the phospholipid acyl
chain region.

If this is the case, and re-extrusion disrupts the local organization of the
143

cholesterol region, perylene would experience increased rotational freedom due to the less
confining nature of its environment. As reorganization of the cholesterol-rich regions proceeds
following re-extrusion, the perylene local environment returns to what it was prior to extrusion,
and that is what we observe experimentally. Dz, Dx and their ratio is seen to return to the same
range of values seen prior to re-extrusion of the vesicles.
Data on the 70% DMPC / 30% cholesterol vesicles (Figure 5.8) appear to behave similarly
to the 80% DMPC / 20% cholesterol system, but the effect(s) of re-extrusion are even less
pronounced. There is no clear trend in these data, attributable perhaps to the presumably larger
cholesterol domains experiencing less disruption during the re-extrusion process.

It is also

possible that due to differences in the phase-separated morphology of the two vesicle
compositions that the re-extrusion process is less disruptive to the local organization of the 70/30
vesicles.

Regardless of the explanation, our measurements are either insensitive to any

structural perturbation imposed on the 70/30 vesicles by re-extrusion or the structural
perturbation associated with re-extrusion is minimal.
It is possible to gauge the frictional intermolecular interactions experienced by perylene in
the bilayer structures.

We can calculate the rotational diffusion constant, D, using the Cartesian

components Dx (=Dy) and Dz, based on37
D

Dx  D y  Dz
3

(5.5)

The rotational diffusion constant, D, is related to the viscosity, η, of the surrounding environment

144

according to the modified Debye-Stokes-Einstein equation,38,40,41

 OR 

1
Vf

6 D k BTS

(5.6)

Where V is the hydrodynamic volume of the chromophore,44

f is a frictional term to account

for solvent-solute interactions,40 kB is the Boltzmann constant, T is the temperature (K), and S is a
shape factor determined from Perrin’s equations to account for the nonspherical shape of the
chromophore.41
225 Å3.44

For perylene, the shape factor, S is 0.70,45 and the hydrodynamic volume, V is

The viscosity, , calculated in this manner is shown in Tables 5.3-5.5.

We find

there is no resolvable cholesterol concentration dependence of the acyl chain microviscosity
sensed by perylene, suggesting the chromophore resides in closest contact with DMPC even in
vesicles containing cholesterol.

It is also possible, although less likely, that the reorientation

dynamics of perylene are the same to within the experimental uncertainty in both phospholipid
and cholesterol domains. Regardless of the explanation, the aspect ratio of the ellipsoid of
rotation of the chromophore is a more sensitive gauge of changes in the organization of the lipid
bilayer than the extraction of an effective micro-viscosity term from the data. This is not a
surprising result given the approximations that are inherent in arriving at values for the local
viscosity.

145

a

Dx, Dz (MHz)

1400
1200
1000
800
600
400

Dz

re-extrusion

100
50
0

Dx

0

200

400

600

800

1000 1200 1400

time (hrs)
b

16
14
Dz/Dx

12
10
re-extrusion

8
6
0

200

400

600

800

1000 1200 1400

time (hrs)
Figure 5.7. (a) Dependence of perylene Dx and Dz on time elapsed after formation for
nominal 100 nm diameter 80% DMPC / 20% cholesterol vesicles at 293 K. (b)
Dependence of Dz/Dx ratio for these vesicles. Vesicles were re-extruded through 100 nm
diameter pore filter at 250 hours.
146

a

Dx, Dz (MHz)

1400
1200
1000
800
600
400

Dz
re-extrusion

100
50
0

Dx

0

200

400

600

800

1000 1200 1400

Dz/Dx

time (hrs)
24
22
20
18
16
14
12
10
8
6

b

re-extrusion

0

200

400

600

800

1000 1200 1400

time (hrs)
Figure 5.8. (a) Dependence of perylene Dx and Dz on time elapsed after formation for
nominal 100 nm diameter 70% DMPC / 30% cholesterol vesicles at 293 K. (b)
Dependence of Dz/Dx ratio for these vesicles. Vesicles were re-extruded through 100 nm
diameter pore filter at 250 hours.

147

Table 5.3. Reorientation time constant and Cartesian components of the rotational
diffusion constant and aspect ratio Dz/Dx as a function of time for DMPC vesicles.

Time after
extrusion
(hrs)

1 (ps)

2 (ps)

Dx (MHz)

Dz (MHz)

Dz/Dx

η (cp)

0

353 ±83

1707 ±255

98 ±15

659 ±167

6.7 ±1.7

7.3 ±1.4

24

341 ±55

2214 ±246

75 ±8

696 ±118

9.2 ±1.6

7.3 ±1.0

48

326 ±48

2021 ±216

82 ±9

725 ±112

8.8 ±1.4

7.0 ±0.9

216

337 ±69

2151 ±282

77 ±10

703 ±153

9.1 ±2.0

7.2 ±1.3

360

374 ±60

2434 ±297

68 ±8

635 ±108

9.3 ±1.6

8.0 ±1.1

370*

317 ±102

1529 ±161

109 ±11

734 ±153

6.7 ±1.3

6.5 ±1.7

456

459 ±162

1769 ±246

94 ±13

498 ±122

5.3 ±1.0

9.0 ±1.6

528

385 ±88

1959 ±208

85 ±9

606 ±148

7.1 ±1.7

8.0 ±1.5

624

479 ±154

2186 ±424

76 ±15

484 ±168

6.3 ±2.2

9.7 ±2.6

*The sample was re-extruded at 370 hours after initial extrusion.

148

Table 5.4. Reorientation time constant and Cartesian components of the rotational
diffusion constant and aspect ratio Dz/Dx as a function of time for 80% DMPC / 20%
cholesterol vesicles.

Time
(hr)

1 (ps)

2 (ps)

Dx (MHz)

Dz (MHz)

Dz/Dx

η (cp)

0

273 ±23

1569 ±307

106 ±21

864 ±77

8.1 ±0.7

5.8 ±0.5

48

409 ±79

2545 ±270

65 ±7

578 ±118

8.8 ±1.8

8.7 ±1.5

168

459 ±36

2852 ±274

58 ±6

516 ±43

8.8 ±0.7

9.8 ±0.7

240

343 ±71

2885 ±398

58 ±8

700 ±151

12.1 ±2.6

7.6 ±1.4

250*

193 ±38

1885 ±289

88 ±14

1253 ±253

14.1 ±2.9

4.3 ±0.8

1200

335 ±40

2085 ±128

80 ±5

706 ±88

8.8 ±1.1

7.2 ±0.7

1320

470 ±55

2647 ±181

63 ±4

500 ±62

7.9 ±1.0

9.9 ±1.0

*The sample was re-extruded at 250 hours after initial extrusion.

149

Table 5.5. Reorientation time constant and Cartesian components of the rotational
diffusion constant and aspect ratio Dz/Dx as a function of time for 70% DMPC / 30%
cholesterol vesicles.

Time
(hr)

1 (ps)

2 (ps)

Dx (MHz)

Dz (MHz)

Dz/Dx

η (cp)

0

304 ±42

2012 ±268

83 ±11

780 ±115

9.4 ±1.4

6.6 ±0.8

48

240 ±32

2466 ±459

68 ±13

1007 ±139

14.9 ±2.1

5.4 ±0.7

168

242 ±40

2350 ±226

71 ±7

1000 ±172

14.1 ±2.4

5.4 ±0.8

240

217 ±53

3197 ±449

52 ±6

1125 ±279

21.6 ±5.4

5.0 ±1.1

250*

217 ±21

2848 ±194

59 ±4

1122 ±112

19.2 ±1.9

5.0 ±0.4

1200

344 ±38

2484 ±137

67 ±4

693 ±80

10.3 ±1.2

7.5 ±0.7

1320

515 ±49

3873 ±306

43 ±3

464 ±46

10.8 ±1.1

11.3 ±1.0

*The sample was re-extruded at 250 hours after initial extrusion.

150

5.4 Conclusions
We have investigated the size distribution and anisotropy decay dynamics of nominal 100
nm diameter DMPC and DMPC-cholesterol vesicles in aqueous solution.

The primary purpose

of this work is to determine whether or not structural changes occur in phospholipid vesicles
following extrusion.

We observed no significant change in vesicle size and size distribution,

indicating vesicle aggregation does not occur to a significant extent for our experimental
conditions. Rotational diffusion data on perylene sequestered within the bilayer acyl chain
region indicate an annealing process that proceeds over hours after vesicle extrusion.

After

re-extrusion, the original anisotropy decay behavior of perylene in 100 nm diameter vesicles is
recovered, indicating that extrusion is accompanied by a loss of organization within the lipid acyl
chain region.

These findings, taken collectively, demonstrate that time-dependent changes in

phospholipid vesicle properties are confined more to the molecular-scale and less to macroscopic
changes (e.g. vesicle fusion) than had been thought previously.

151

REFERENCES

152

REFERENCES

(1)
Mollee, H. M.; Steenvoorden, D. P. T.; De Vringer, T.; Crommelin, D. J. A. J. Pharm. Sci.
2001, 90, 588.
(2)

Zhang, J.; Xue, R. H.; Ong, W. Y.; Chen, P. Biophys. J. 2009, 97, 1371.

(3)
Puri, A.; Kramer-Marek, G.; Campbell-Massa, R.; Yavlovich, A.; Tele, S. C.; Lee, S. B.;
Clogston, J. D.; Patri, A. K.; Blumenthal, R.; Capala, J. Journal of Liposome Research 2008, 18,
293.
(4)
Agarwal, A.; Mackey, M. A.; El-Sayed, M. A.; Bellamkonda, R. V. ACS Nano 2011, 5,
4919.
(5)
Smith, B.; Lyakhov, I.; Loomis, K.; Needle, D.; Baxa, U.; Yavlovich, A.; Capala, J.;
Blumenthal, R.; Puri, A. Journal of Controlled Release 2011, 153, 187.
(6) Hossann, M.; Syunyaeva, Z.; Schmidt, R.; Zengerle, A.; Eibl, H.; Issels, R. D.; Lindner, L.
H. Journal of Controlled Release 2012, 162, 400.
(7)
Driessen, A. J. M.; Vandenhooven, H. W.; Kuiper, W.; Vandekamp, M.; Sahl, H. G.;
Konings, R. N. H.; Konings, W. N. Biochemistry 1995, 34, 1606.
(8)

Hunter, D. G.; Frisken, B. J. Biophys. J. 1998, 74, 2996.

(9)
Macdonald, R. C.; Macdonald, R. I.; Menco, B. P. M.; Takeshita, K.; Subbarao, N. K.; Hu,
L. R. Biochim. Biophys. Acta 1991, 1061, 297.
(10)

Mayer, L. D.; Hope, M. J.; Cullis, P. R. Biochim. Biophys. Acta 1986, 858, 161.

(11) Maulucci, G.; De Spirito, M.; Arcovito, G.; Boffi, F.; Castellano, A. C.; Briganti, G.
Biophys. J. 2005, 88, 3545.
153

(12) Lapinski, M. M.; Castro-Forero, A.; Greiner, A. J.; Ofoli, R. Y.; Blanchard, G. J.
Langmuir 2007, 23, 11677.
(13)

Bonifacino, J. S.; Glick, B. S. Cell 2004, 116, 153.

(14)

Nir, S.; Bentz, J.; Wilschut, J.; Duzgunes, N. Prog. Surf. Sci. 1983, 13, 1.

(15)

Hay, C. E.; Marken, F.; Blanchard, G. J. J. Phys. Chem. A 2010, 114, 4957.

(16)
264.

Halstenberg, S.; Heimburg, T.; Hianik, T.; Kaatze, U.; Krivanek, R. Biophys. J. 1998, 75,

(17)

Koan, M. M.; Blanchard, G. J. J Phys Chem B 2006, 110, 16584.

(18)

Lapinski, M. M.; Blanchard, G. J. Chem. Phys. Lipids 2008, 153, 130.

(19) de Meyer, F. J.-M.; Benjamini, A.; Rodgers, J. M.; Misteli, Y.; Smit, B. Journal of
Physical Chemistry B 2010, 114, 10451.
(20)

Scheetz, M. P.; Chan, S. I. Biochemistry 1972, 11, 4573.

(21) Yohannes, G.; Pystynen, K.-H.; Riekkola, M.-L.; Wiedmer, S. K. Anal. Chim. Acta 2006,
560, 50.
(22)
609.

Yue, B.; Huang, C.-Y.; Nieh, M.-P.; Glinka, C. J.; Katsaras, J. J Phys Chem B 2005, 109,

(23)
109.

Juliano, R. L.; Hsu, M. J.; Regen, S. L.; Singh, M. Biochim. Biophys. Acta 1984, 770,

(24)

Kao, Y.; Juliano, R. L. Biochim. Biophys. Acta 1981, 677, 453.

(25)

Kas, J.; Sackmann, E. Biophys. J. 1991, 60, 825.
154

(26)

Seki, K.; Tirrell, D. A. Macromol. 1984, 17, 1692.

(27)

Discher, D. E.; Eisenberg, A. Science 2002, 297, 967.

(28) Lee, J. C. M.; Bermudez, H.; Discher, B. M.; Sheehan, M. A.; Won, Y. Y.; Bates, F. S.;
Discher, D. E. Biotech. Bioeng. 2001, 73, 135.
(29)

Ruf, H.; Georgalis, Y.; Grell, E. Method Enzymol. 1989, 172, 364.

(30) Trouard, T. P.; Nevzorov, A. A.; Alam, T. M.; Job, C.; Zajicek, J.; Brown, M. F. J. Chem.
Phys. 1999, 110, 8802.
(31) Gracia, R. S.; Bezlyepkina, N.; Knorr, R. L.; Lipowsky, R.; Dimova, R. Soft Matter 2010,
6, 1472.
(32)

Kucerka, N.; Pencer, J.; Nieh, M.-P.; Katsaras, J. Euro. J. Phys. E 2007, 23, 247.

(33) Khaykovich, B.; Kozlova, N.; Choi, W.; Lomakin, A.; Hossain, C.; Sung, Y.; Dasari, R.
R.; Feld, M. S.; Benedek, G. B. Proc. Nat. Acad. Sci. 2009, 106, 15663.
(34)

Howland, M. C.; Szmodis, A. W.; Sanii, B.; Parikh, A. Biophys. J. 2007, 92, 1306.

(35)

Mehra, R. Proc. Indian Acad. Sci. (Chem. Sci.) 2003, 115, 147.

(36)

Targowski, P.; Atzeni, S. H.; Davenport, L. Biophys. J. 1993, 64, A72.

(37)

Chuang, T. J.; Eisenthal, K. B. J. Chem. Phys. 1972, 57, 5094.

(38)

Debye, P. Polar Molecules Chemical Catalog Co.: New York 1929.

(39)

Favro, D. L. Phys. Rev. 1960, 119, 53.

(40)

Hu, C. M.; Zwanzig, R. The Journal of Chemical Physics 1974, 60, 4354.
155

(41)

Perrin, F. J. Phys. Radium 1934 5, 497.

(42)

Szabo, A. Journal of Chemical Physics 1984, 81, 150.

(43)

Tao, T. Biopolymers 1969, 8, 609.

(44)

Edward, J. T. J. Chem. Ed. 1970, 47, 261.

(45)

Jiang, Y.; Blanchard, G. J. J. Phys. Chem. 1994, 98, 6436.

156

CHAPTER 6
SUMMARY AND FUTURE WORK

6.1 Summary of the Dissertation Work
The ultimate goal of this research is to understand the molecular interactions and energy
transfer properties in different fluid systems, and further provide the insight on predicting and
tailoring energy flow in complex fluid systems. Compared with neat solvents, the relaxation
dynamics is far more complex in binary systems, due to the complexity of solute-solvent and
solvent-solvent interactions, as well as the transient nature of all relevant processes.1-4
Preferential solvation and molecular scale transient organization are known to exist in liquid
phase multi-component systems.5-9
In this dissertation, we have examined intermolecular interaction and the vibrational
population relaxation dynamics between the solute and solvent molecules in binary liquid phase
systems over a wide composition range at room temperature. Binary liquid systems have been
chosen because they serve as a useful starting point for the examination of multi-component
solvent systems. The bulk properties, such as viscosity, thermal conductivity and dielectric
constant of many binary solvent systems have been established.10-13 However, the ability to
interrogate and monitor molecular scale non-uniformity limits the investigation of binary solvent
systems. In our studies, an organic polycyclic aromatic hydrocarbon, perylene, has been chosen
as the molecular probe to evaluate whether or not local organization and compositional
heterogeneity plays a role in binary solvent systems behaviors. In both ethanol/cyclohexane and
n-butanol/cyclohexane binary systems, nanoscale molecular heterogeneity is seen.

157

In chapter 2 of this dissertation, a time-resolved pump-probe spectroscopic instrument
has been described, which allows simultaneous measurement of rotational diffusion and
vibrational energy relaxation dynamics on a variety of liquid based systems. This instrument
combines picosecond dye lasers and a three-frequency modulation scheme, and provides ca. 5 ps
time resolution for both types of measurements. The complementary time-resolved data
generated from this instrument is related to the local environment of probe chromophore.
In chapter 3, we reported on the molecular motion and vibrational population relaxation
of perylene as a function of constituent concentration for ethanol/cyclohexane binary solvent
system. This body of work provided experimental evidence of a discontinuous change in the
reorganization and amount of ethanol present in the immediate proximity of perylene. Both
anisotropy decay and vibrational population relaxation results show the discontinuous change in
the perylene local molecular environment occurs at the same ethanol concentration, between 7.5%
(v/v) and 5% (v/v) ethanol. The anisotropy decay data show that perylene is constrained to
reorient primarily about the axis normal to its -plane with the ethanol concentration lower than
5% (v/v) and this molecular motion changes to a reorientation about an axis that lies in the plane as the ethanol concentration increases to 7.5% (v/v) or above, suggesting a less confined
molecular environment. The vibrational relaxation time constant (T1) for the perylene 1375 cm-1
mode shows a clear ethanol concentration-dependence as well. The T1 values correspond to the
accessibility and orientation of ethanol terminal methyl group to the perylene ring breathing
mode as there is no vibrational mode in close energy proximity to the perylene 1375 cm-1 mode
cyclohexane possesses. As a result, changes in T1 reflect the changes in (at least) two aspects of
the perylene molecular environment in ethanol/cyclohexane systems as the solvent composition
changes. Combining both sets of data, we conclude that the perylene is solvated preferentially

158

by cyclohexane at low ethanol concentration (<5%), and above a ~7.5% threshold concentration,
the local environment is dominated by ethanol.
With

this

understanding

of

the

molecular

relaxations

of

perylene

in

the

ethanol/cyclohexane binary solvent systems, we investigated and reported in chapter 4 the nbutanol/cyclohexane binary solvent system. This system is similar to ethanol/cyclohexane, but
with longer chain alcohol that could in principle, alter the chromophore local environment. The
objective of this work is to gain further insight to the role of alcohol in mediating the local
organization in the binary mixtures. The steady-state spectra, fluorescence lifetime, anisotropy
decay and vibrational population relaxation dynamics data show collectively that this binary
solvent system also exhibits an alcohol concentration dependent nanometer-scale organizational
heterogeneity.

The local viscosities experienced by perylene have been determined in the

context of the Debye-Stokes-Einstein model based on the anisotropy decay data for the binary
solvent systems. Viscosities that remarkably lower than the bulk solvent viscosities are found,
suggesting that the chromophore resides in a non-polar environment. In addition, between 5%
and 7.5% (v/v) n-butanol, the drop in local apparent viscosity demonstrates the dramatic
organizational change associated with the amount of n-butanol in these systems. The vibrational
population relaxation time (T1) results also show a discontinuous change with increasing nbutanol concentration, which points to an environment dominated by n-butanol, once it is present
in sufficient quantity. Based on these studies, it appears that perylene resides in a nearly constant
environment over a wide range of n-butanol concentration, which might be due to the formation
of structures in solvation alike to inverse-micelles by n-butanol. We are aware that, to date, there
is no corroboration for such structure, however.

159

As mentioned in the chapter 1, in addition to multi-component mixed solvents, another
system of interest in the context of molecular organization and dynamics studies is phospholipid
bilayers, for which synthetic unilamellar phospholipid vesicles serve as a easily accessible
structure with wide application. The issue we resolved in chapter 5 was the long-term structural
stability, both dimensional and morphological, of vesicles compressed of the phospholipid 1,2dimyristoyl-sn-phosphatidylcholine (DMPC), with or without cholesterol, in aqueous
environment.

The average size and size distribution of unilamellar DMPC vesicles was

examined as a function of time after extrusion by dynamic light scattering (DLS), which shows
that vesicle-vesicle fusion does not play a significant role under our experimental conditions. By
incorporating the nonpolar chromophore perylene into the acyl chain region of DMPC vesicles,
the molecular scale organization of these vesicles was able to be studied by anisotropy decay
measurements using a time-correlated single photon counting (TCSPC) instrument. The
rotational diffusion results indicate an annealing process that proceeds over hours after vesicle
extrusion, and also establish the initial relatively disorganized structure can be recovered by the
re-extrusion. The key finding of this work is that the time-dependent structural change in the
phospholipid vesicle acyl region, after hundreds of hours after extrusion is confined to the
molecular level, without macroscopic changes, such as vesicle fusion. These results provide new
insight into the long-term stability of phospholipid vesicles in aqueous medium, information that
is important for further studies of phospholipids.

160

6.2 Future Directions
The work we reported in this dissertation shows promise in gaining molecular scale
information on heterogeneous fluids and other complex systems. Future studies can be carried
out in different directions to complement the knowledge we have at this point.
For the alcohol/cyclohexane binary solvent systems, one remaining question is the details
of the role of alcohol in mediating preferential solvation.

Specifically, the composition-

dependent heterogeneity changes as a function of the alcohol structure and amount. Using
longer chain alcohols and comparing the anisotropy decay and vibrational population relaxation
data from those measurements, can provide a clearer picture of the local molecular environment
formed by the solvent molecules around the probe. The isomers of n-butanol could be studied,
for example, to provide insight into the effect of steric hindrance on the accessibility of the
alcohol aliphatic functionality to the chromophore.
Other fluorescent probes can also be used in the measurements. 1-Methyperylene can be
used instead of perylene, for example. 1-Methylperylene can be synthesized by alkylation of
perylene with methyllithium.14 Both perylene and 1-Methylperylene contain a ring breathing
mode at ~1375cm-1, which degenerate with the IR-active terminal methyl group of compounds
containing aliphatic functionality. The basic structural difference between the two chromophores
lies in their symmetry. Perylene is a centrosymmetric (D2h) molecule with the Raman-active ring
breathing mode, which modulates the molecular quadrupole moment and gives rise to a
quadrupole-dipole coupling which scales as r-8.15

1-Methyperylene contains no center of

inversion (Cs symmetry), and its ring breathing mode modulates the molecular dipole moment,
giving rise to dipole-dipole coupling with r-6 scales.15 These two probe molecules can be used to

161

probe vibrational energy relaxation processes in different length scales,16,17 and allows us for a
more detailed molecular level understanding of nanometer scale heterogeneity in fluid systems.
Another direction for this research is to understand molecular motion and energy flow in
single and/or multi-component phospholipid vesicles.

This can be done by incorporating

perylene or 1-Methyperylene into DMPC vesicles, and monitoring anisotropy decay and
vibrational population relaxation dynamics as a function of vesicle composition, and the identity
of the medium in which the vesicle reside. We can also vary the vesicle formation (e.g.
extrusion vs. sonication),18 temperature, or size/curvature of vesicles to gain a comprehensive
understanding of the composition-structure relationship for these systems, which will provide
insight into other lipid bilayers and related systems, such as biosensor applications and
pharmaceutical delivery researches.19-22

162

REFERENCES

163

REFERENCES

(1)

Chandler, D. W.; Ewing, G. E. The Journal of Physical Chemistry 1981, 85, 1994.

(2)

Mukherjee, A.; Srinivas, G.; Bhattacharyya, S.; Bagchi, B. J Chem Sci 2001, 113, 393.

(3)

Fujisaki, H.; Stock, G. The Journal of Chemical Physics 2008, 129.

(4)

Gazi, H. A. R.; Biswas, R. The Journal of Physical Chemistry A 2011, 115, 2447.

(5)

Shimizu, S.; Matubayasi, N. The Journal of Physical Chemistry B 2014, 118, 3922.

(6)
Banerjee, D.; Laha, A. K.; Bagchi, S. Journal of the Chemical Society, Faraday
Transactions 1995, 91, 631.
(7)
Sala, S.; Danten, Y.; Ventosa, N.; Tassaing, T.; Besnard, M.; Veciana, J. The Journal of
Supercritical Fluids 2006, 38, 295.
(8)
Nunes, N.; Ventura, C.; Martins, F.; Leitão, R. E. The Journal of Physical Chemistry B
2009, 113, 3071.
(9)
Farajtabar, A.; Jaberi, F.; Gharib, F. Spectrochimica Acta Part A: Molecular and
Biomolecular Spectroscopy 2011, 83, 213.
(10) Flynn, G. W.; Parmenter, C. S.; Wodtke, A. M. The Journal of Physical Chemistry 1996,
100, 12817.
(11)

Rapp, D.; Kassal, T. Chemical Reviews 1969, 69, 61.

(12) Chen, H.; Wen, X.; Guo, X.; Zheng, J. Physical Chemistry Chemical Physics 2014, 16,
13995.
(13)

Sander, W.; Huib, J. B. Nature 1999, 402, 507.

(14)

E. Zieger, H.; M. Laski, E. Tetrahedron Letters 1966, 7, 3801.
164

(15)

Calais, J.-L. International Journal of Quantum Chemistry 1990, 38, 497.

(16)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.

(17)

Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 6785.

(18) Lapinski, M. M.; Castro-Forero, A.; Greiner, A. J.; Ofoli, R. Y.; Blanchard, G. J.
Langmuir 2007, 23, 11677.
(19) Puri, A.; Kramer-Marek, G.; Campbell-Massa, R.; Yavlovich, A.; Tele, S. C.; Lee, S. B.;
Clogston, J. D.; Patri, A. K.; Blumenthal, R.; Capala, J. Journal of Liposome Research 2008, 18,
293.
(20) Agarwal, A.; Mackey, M. A.; El-Sayed, M. A.; Bellamkonda, R. V. ACS Nano 2011, 5,
4919.
(21) Smith, B.; Lyakhov, I.; Loomis, K.; Needle, D.; Baxa, U.; Yavlovich, A.; Capala, J.;
Blumenthal, R.; Puri, A. Journal of Controlled Release 2011, 153, 187.
(22) Hossann, M.; Syunyaeva, Z.; Schmidt, R.; Zengerle, A.; Eibl, H.; Issels, R. D.; Lindner,
L. H. Journal of Controlled Release 2012, 162, 400.

165