CHARACTERIZATION OF MOLECULAR INTERACTIONS AND ORGANIZATION IN
SOLUTION AND INTERFACIAL SYSTEMS
By
Christine Ella Hay

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Chemistry – Doctor of Philosophy
2014

ABSTRACT
CHARACTERIZATION OF MOLECULAR INTERACTIONS AND ORGANIZATION IN
SOLUTION AND INTERFACIAL SYSTEMS
By
Christine Ella Hay
Liquid solutions and liquid-liquid interfaces are the media for numerous chemical
reactions, where the local environments are conducive to interactions and exchanges. As a
means of broadening our understanding of the molecular organization in liquid solutions and in
proximity to phase boundaries, we report on the rotational diffusion dynamics of probe
molecules in increasingly complex systems. Neat liquids represent the simplest systems
interrogated. In our first study, we measure the induced orientational anisotropy function, R(t),
using time-correlated single photon counting of two chromophores, resorufin and 6-(N-(7nitrobenz-2-oxa-1,3-diazol-4-yl)amino)hexanoic acid (NBDHA) in water and in N-octyl-2pyrrolidone (NOP). Our data show distinctly different anisotropy decays in aqueous solution
compared to NOP solutions, and taken in the context of the Chuang and Eisenthal model, this
indicates a change in effective rotor shape swept out by the solute’s rotation. The similarity in
the behavior of the two different chromophores in these solvent systems points to solvent-solvent
interactions and local organization as the dominant factors in mediating motional dynamics.
By adding varying concentrations of electrolyte to aqueous and NOP systems, the
rotational diffusion dynamics of the anionic chromophore, resorufin, are altered as a result of
interaction with electrolyte cations. In aqueous solutions containing lithium perchlorate
(LiClO4), resorufin exhibits a single exponential anisotropy decay as was the case in pure water,
however, reorientation times are dependent on electrolyte concentration. In contrast to the
observed behavior of resorufin in pure NOP, where bi-exponential decay occurs, we observe a

single exponential anisotropy decay for resorufin in NOP with the addition of non-aqueous
electrolyte. For resorufin in NOP, the reorientation time constant increases with increasing
electrolyte concentration, consistent with formation of a complex between the resorufin anion
and the electrolyte cation.
Elucidating the local organization near phase boundaries compared to bulk solutions
represents a more complex problem to interrogate. We use a TCSPC confocal imaging
instrument to obtain depth-resolved fluorescence lifetime and anisotropy decay data for
lissamine rhodamine B sulfonyl chloride (LRSC) in NOP saturated with water and in ethylene
glycol (EG) supported on a glass surface. The fluorescence anisotropy data of LRSC in NOP
and EG as a function of distance from NOP|glass and EG|glass interfaces reveal a gradient
spanning tens of micrometers from the NOP|glass interface into the NOP phase, and no
corresponding gradient in EG. From these data, we assert there is a compositional heterogeneity
in the form of water nano-droplets in the NOP phase, where water molecules solvate LRSC.
NOP is miscible with small amounts of water allowing for the presence of water inclusions in
solution, while phase separation occurs for sufficiently high water concentrations.
By forming a liquid-liquid interface composed of water and NOP supported on glass, we
observe both compositional and dielectric gradients. We report on spatially resolved
fluorescence lifetime and anisotropy decay time constant data of LRSC in the NOP phase near
the liquid-liquid (NOP|water) interface and the liquid-solid (NOP|glass) interface. These data
reveal a micron-scale position-dependent molecular environment characterized by an anisotropy
decay gradient as a function of distance from the NOP|glass interface. In addition, the presence
of a fluorescence lifetime gradient normal to the NOP|water interface indicates the existence of a
gradient in the concentration of water nano-droplets.

ACKNOWLEDGMENTS

This dissertation is the culmination of work that could not have been completed without
the contributions of others. I want to take this opportunity to thank those involved in my
graduate career. First and foremost, I would like to thank my research advisor, Dr. Gary
Blanchard, for his guidance, support, generosity, and optimism. I would also like to thank our
collaborator, Dr. Frank Marken, of the University of Bath, UK for his insight and suggestions. I
want to express gratitude to Dr. Dana Spence, Dr. Greg Swain, and Dr. David Weliky for serving
on my dissertation advisory committee.
My fellow Blanchard group members have been an essential part of my graduate studies
and research. I would like to thank the following former and current Blanchard research group
members for their advice, support, and friendship: Dr. Benjamin Oberts, Dr. Monika (Domińska)
Ngowe, Dr. Heather Pillman, Dr. Margaretta (Dimos) Koster, Dr. Douglas Gornowich, Dr. Iwan
Setiawan, Chen Qiu, Fredy Pratama, Stephen Baumler, Xiaoran Zhang, and Hannah Mize.
Finally, I would like to express thanks to my friends in the MSU Chemistry Department
and the MSU Forensic Chemistry Program, as well as my Master’s research advisor, Dr. Ruth
Waddell Smith for their support during my time at MSU. I would like to give a special thanks to
my family for their encouragement, love, and support.

iv

TABLE OF CONTENTS

LIST OF TABLES ...................................................................................................................................... vii
LIST OF FIGURES ................................................................................................................................... viii

Chapter 1 ......................................................................................................................................... 1
Introduction ..................................................................................................................................... 1
1.1 Significance and Research Objective.................................................................................... 1
1.2 Literature Review.................................................................................................................. 3
REFERENCES ......................................................................................................................... 15
Chapter 2 ....................................................................................................................................... 21
Solvent-Dependent Changes in Molecular Reorientation Dynamics: The Role of Solvent-Solvent
Interactions .................................................................................................................................... 21
2.1 Introduction ......................................................................................................................... 21
2.2 Experimental Methods ........................................................................................................ 24
2.3 Results and Discussion ....................................................................................................... 27
2.4 Conclusions ......................................................................................................................... 43
REFERENCES ......................................................................................................................... 44
Chapter 3 ....................................................................................................................................... 48
The Effects of Electrolyte Concentration on the Rotational Dynamics of Resorufin................... 48
3.1 Introduction ......................................................................................................................... 48
3.2 Experimental Methods ........................................................................................................ 53
3.3 Results and Discussion ....................................................................................................... 54
3.4 Conclusions ......................................................................................................................... 67
REFERENCES ......................................................................................................................... 69
Chapter 4 ....................................................................................................................................... 72
Detection and Characterization of a Liquid|Solid Interfacial Compositional Gradient of Water
Nano-Droplets in Wet N-Octyl-2-Pyrrolidone ............................................................................. 72
4.1 Introduction ......................................................................................................................... 72
4.2 Experimental Methods ........................................................................................................ 74
4.3 Results and Discussion ....................................................................................................... 79
4.4 Conclusions ......................................................................................................................... 99
REFERENCES ....................................................................................................................... 100
Chapter 5 ..................................................................................................................................... 103
Characterization of Liquid|Liquid|Solid Multi-Interfacial Compositional Gradients................. 103
5.1 Introduction ....................................................................................................................... 103
5.2 Experimental Methods ...................................................................................................... 105
5.3 Results and Discussion ..................................................................................................... 108
5.4 Conclusions ....................................................................................................................... 118
REFERENCES ....................................................................................................................... 120
v

Chapter 6 ..................................................................................................................................... 123
Conclusions ................................................................................................................................. 123
6.1 Conclusions ....................................................................................................................... 123
6.2 Future Directions .............................................................................................................. 126
REFERENCES ....................................................................................................................... 130

vi

LIST OF TABLES

Table 2.1 Calculated changes in permanent dipole moment, Δμ, upon excitation for the
chromophores resorufin and NBDHA in solvents water and NOP. ........................... 31
Table 2.2 Comparison of reorientation dynamics of resorufin48 and NBDHA45 in selected
solvents. ...................................................................................................................... 38
Table 2.3 Anisotropy decay time constants, τOR, and Cartesian components of the average
rotational diffusion constants, D, for resorufin and NBDHA in water and in NOP. .. 38
Table 3.1 Reorientation times, τOR, of resorufin in neat NOP and NOP with various
concentrations of TBAB or TOAB electrolytes.......................................................... 60
Table 3.2 Reorientation times, τOR, of resorufin in water and various concentrations of LiClO4
(aq). ............................................................................................................................. 65
Table 4.1 Estimated number of water molecules surrounding LRSC (model prediction for
number of waters, n).a ................................................................................................. 91
Table 5.1 Fluorescence lifetime, τfl, (ps) of LRSC as a function of depth into the NOP phase of a
NOP|water system and distance from the NOP|water interface................................ 109
Table 5.2 Fluorescence anisotropy decay time constants, τOR, (ps) of LRSC as a function of
depth into the NOP phase of a NOP|water system and distance from the NOP|water
interface..................................................................................................................... 110

vii

LIST OF FIGURES

Figure 1.1 Structure of N-octyl-2-pyrrolidone (NOP). ................................................................ 10
Figure 1.2 Time-resolved fluorescence lifetime decay for resorufin in aqueous solution. The
best fit line (red) is a one-component exponential decay function. ............................ 12
Figure 1.3 (a) Parallel (black) and perpendicular (red) emission transients and (b) orientational
anisotropy function, R(t), for resorufin in aqueous solution. The best fit line (red) is a
one-component exponential decay function. .............................................................. 13
Figure 2.1 Structures of (a) resorufin sodium salt and (b) NBD hexanoic acid (NBDHA) shown
with Cartesian x and y axes for clarity. ....................................................................... 23
Figure 2.2 Time-correlated single photon counting (TCSPC) instrument................................... 26
Figure 2.3 Normalized absorption and emission spectra of (a) NBDHA in water and (b)
NBDHA in NOP. ........................................................................................................ 29
Figure 2.4 Normalized absorption and emission spectra of (a) resorufin in water and (b)
resorufin in NOP. ........................................................................................................ 30
Figure 2.5 (a) Orientational anisotropy function, R(t), plotted as a function of time for resorufin
in water. (b) R(t) as a function of time for NBDHA in water. The data are fitted to
single exponential decays. .......................................................................................... 34
Figure 2.6 (a) Orientational anisotropy function, R(t), plotted as a function of time for resorufin
in NOP. (b) R(t) as a function of time for NBDHA in NOP. The data are fitted to
two-component exponential decays. ........................................................................... 35
Figure 2.7 Three-dimensional shape representing the volume swept out when a long-axis
polarized chromophore (a) rotates about the x-axis as a prolate rotor and (b) rotates
about the z-axis as an oblate rotor. .............................................................................. 37
Figure 3.1 Structures of non-aqueous electrolytes, (a) tetrabutylammonium bromide (TBAB)
and (b) tetraoctylammonium bromide (TOAB), with hydrodynamic volumes of 299
Å3 and 571 Å3, respectively. ....................................................................................... 52
Figure 3.2 Illustration of the resorufin anion reorienting as an oblate rotor with diffusion
primarily about the z-axis in neat NOP and sweeping out a volume representative of a
prolate rotor after formation of a complex with a tetrabutyl- or tetraoctylammonium
cation. .......................................................................................................................... 56
Figure 3.3 Reorientation time of the reorienting moiety as a function of electrolyte concentration
(log10 scale) for resorufin in TBAB/NOP (filled circles) and TOAB/NOP (open
circles). Error bars represent ± 1 standard deviation. ................................................. 57
viii

Figure 3.4 Normalized (a) absorption and (b) emission spectra of resorufin in neat NOP (black),
10-3 M TOAB (red), 10-2 M TOAB (blue), and 10-1 M TOAB (green). ...................... 62
Figure 3.5 Reorientation time as a function of electrolyte concentration (log10 scale) for
resorufin in aqueous LiClO4. Errors bars represent ± 1 standard deviation. .............. 64
Figure 3.6 Normalized (a) absorption and (b) emission spectra of resorufin in water (black), 5 ×
10-3 M LiClO4 (red), 10-2 M LiClO4 (blue), 5 × 10-2 M LiClO4 (green), and 10-1 M
LiClO4 (orange). ......................................................................................................... 66
Figure 4.1 Structures of (a) lissamine rhodamine B sulfonyl chloride (LRSC) and (b) N-octyl-2pyrrolidone (NOP). ..................................................................................................... 75
Figure 4.2 Time-resolved fluorescence lifetime and anisotropy imaging instrument with timecorrelated single photon counting (TCSPC) detection. .............................................. 77
Figure 4.3 Schematic of sample and optical excitation/collection configuration for NOP|glass
and EG|glass interfaces. (NOP|glass interface shown.) ............................................. 78
Figure 4.4 Dependence of LRSC in NOP fluorescence lifetime (●) and anisotropy decay time
constant (○) on distance from the glass interface. The line drawn through the
fluorescence lifetime data is intended as a guide to the eye. The glass support was
cleaned with piranha solution and rinsed with water prior to use. Refer to Fig. 4.3 for
the experimental configuration and the x, y, z coordinate designation. ...................... 81
Figure 4.5 Dependence of LRSC in NOP fluorescence lifetime (●) and anisotropy decay time
constant (○) on distance from the glass interface. The line drawn through the
fluorescence lifetime data is intended as a guide to the eye. The glass support was
cleaned using UV-generated ozone and rinsed with ethanol prior to use. Refer to Fig.
4.3 for the experimental configuration and the x, y, z coordinate designation. .......... 83
Figure 4.6 Dependence of LRSC in EG fluorescence lifetime (●) and anisotropy decay time
constant (○) on distance from the glass interface. The line drawn through the
fluorescence lifetime data (solid) and the anisotropy decay data (dashed) are intended
as guides to the eye. The glass support was cleaned using piranha solution and rinsed
with water prior to use. Refer to Fig. 4.3 for the experimental configuration and the x,
y, z coordinate designation. ......................................................................................... 86
Figure 4.7 (a) Calculated number of water molecules in the immediate proximity of LRSC as a
function of distance from the NOP|glass interface, for the interface cleaned with
piranha solution and water (●) and cleaned with ethanol and UV-generated ozone (○).
(b) Calculated diameters of the reorienting entities containing the chromophore and
the number of water molecules indicated in (a), as a function of distance from the
NOP|glass interface. .................................................................................................... 92
Figure 4.8 Normailzed absorption spectra of LRSC in water (black), LRSC in dry NOP (red),
LRSC in NOP after equilibration with water (blue), and LRSC in NOP after
extraction from water (green). .................................................................................... 94
ix

Figure 4.9 Dynamic light scattering (DLS) data showing the size distribution of water nanodroplets present in wet NOP after exposure to water (●) and dry NOP (○), showing
the absence of species in the same size range as the water nano-droplets. For all DLS
measurements no chromophore was present............................................................... 96
Figure 5.1 Schematic of sample and optical excitation/collection configuration for
NOP|water|glass interface. ........................................................................................ 106
Figure 5.2 Time-resolved images of partitioning of LRSC initially deposited into the aqueous
phase (bottom) and migrating into the NOP phase (top). In each image, the cuvette
on left contains LRSC and the cuvette on the right is NOP|water with no
chromophore to demonstrate the existence of the phase separation. ........................ 107
Figure 5.3 (a) Reorientation time and (b) calculated number of water molecules as a function of
distance from the NOP|glass interface. Refer to Fig. 5.1 for the experimental
configuration and the x, y, z coordinate designation. ................................................ 111
Figure 5.4 (a) Reorientation time and (b) calculated number of water molecules as a function of
distance from the NOP|water interface. Refer to Fig. 5.1 for the experimental
configuration and the x, y, z coordinate designation. ................................................ 112
Figure 5.5 (a) Fluorescence lifetime as a function of distance from the NOP|glass interface for
three different distances from the water interface. (b) Fluorescence lifetime as a
function of distance from the NOP|water interface for six different distances from the
NOP|glass interface. Experimental error bars representing the standard deviation are
shown for each data point. Refer to Fig. 5.1 for the experimental configuration and
the x, y, z coordinate designation. ............................................................................. 115
Figure 5.6 (a) Dielectric constant as a function of distance from the NOP|glass interface for three
different distances from the water interface. (b) Dielectric constant as a function of
distance from the NOP|water interface for six different distances from the NOP|glass
interface. Refer to Fig. 5.1 for the experimental configuration and the x, y, z
coordinate designation. ............................................................................................. 116

x

Chapter 1
Introduction
1.1 Significance and Research Objective
The formation of liquid interfaces dates back to the origins of life on Earth and continues
to impact many aspects of modern life. Early liquid-liquid interfaces were the site of
photochemical reactions allowing light harvesting to take place in primitive cells1 and are still
incorporated into energy conversion devices today, while liquid-solid interfaces, which are also
ubiquitous in nature, are utilized in the making sensors and microscale devices.2 Biological cell
membranes, which have fluid properties similar to a liquid-liquid interface and surfaces that
behave more akin to a liquid-solid interface, represent a well-studied interfacial system and
function as the site of many chemical reactions necessary for cell function including a majority
of energy conversion processes in living organisms.3 Liquid-liquid interfaces can take on many
structural forms depending on the physical properties at work. For example, partially miscible
and immiscible liquids can form emulsions or colloidal suspensions of structured liquids that are
utilized in industrial products, such as detergents, paints, and cosmetics. Liquid interfaces
including liquid-air, liquid-liquid, and liquid-solid are present in both simple and complex
systems and play a vital role in many chemical processes, such as chemical separations, phase
transfer catalysis, drug delivery, and charge transfer at electrode surfaces.
From a fundamental perspective, a liquid interface can be used to study the unique
properties of interfaces that allow them to function as reaction centers.4-7 Central to the function
of an interface is the structure and dynamics occurring in the interfacial region. Macroscopic
properties of interfaces, such as surface tension at liquid-solid interfaces, have been widely
studied, however, the microscopic properties of interfaces have been more difficult to ascertain.8

1

In the study of an interfacial system, knowledge about both sides of the system is important. For
a liquid interface, this knowledge begins with understanding interactions and dynamics within
the liquid phase.
Fluorescence spectroscopy is sensitive to microenvironments in liquid solutions and,
therefore, is particularly useful for the study of solvation phenomena. Rotational diffusion
measurements made by time-resolved fluorescence spectroscopic techniques provide a means of
measuring the dynamic motion of a probe molecule and elucidating properties of the local
organization in the solution. Through the use of optical spectroscopic techniques, a molecular
scale understanding of solution phase, solute-solvent and solvent-solvent, interactions is sought
to increase our knowledge of the liquid phase as a means to have greater control over chemical
processes, such as heterogeneous catalysis and ion transfer reactions.
The objective of the research presented here is to gain a better fundamental understanding
of the molecular organization in solution phase media and in interfacial systems. The goals of
this research project are to use time-resolved fluorescence spectroscopy to study the local
environment of probe molecules in neat solutions, in solutions containing additives in the form
of electrolytes, and finally in proximity to liquid-solid and liquid-liquid interfaces. The liquidliquid and liquid-solid phase boundaries, as well as the three phase boundary formed between the
two interfaces are of particular interest for photochemical and photoelectrochemical reactions
involving ion transfer.
The solvent systems used throughout this research are water and N-octyl-2-pyrrolidone
(NOP). Both water and NOP exhibit unique solvation properties and form a liquid-liquid
interface at room temperature. Three fluorescent probe molecules are used as solutes in these
studies; resorufin sodium salt, 6-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino] hexanoic acid

2

(NBDHA), and lissamine rhodamine B sulfonyl chloride (LRSC). More detailed descriptions of
the probe molecules are presented in the subsequent chapters, while the remainder of this chapter
focuses on relevant research highlighting studies on solvation and liquid interfaces.
1.2 Literature Review
Physical properties of liquids have been studied for centuries and advancements in
measurement techniques have allowed researchers to delve in progressively greater depth into
the complexity of liquid solutions. Both experimental and theoretical techniques have enhanced
the field by showing how nano- and microscale interactions affect macroscale properties. This
review focuses on optical spectroscopic techniques used for the study of intermolecular
interactions as a means to understand solvation phenomena in liquid solutions and in interfacial
systems. Prior to a review of significant research, a brief introduction to the liquid phase is
necessary.
The liquid phase is governed by interactions between its components. In a neat liquid,
the interactions are homogeneous in nature, whereas in solutions composed of a solute and a
solvent, where the solvent is the major component, interactions occur between dissimilar
molecules. In both cases intermolecular interactions between solvent-solvent and/or solutesolvent play a deterministic role in the physical properties of the liquid.9 The theory describing
the behavior of liquids lies between kinetic gas theory and solid state physics and is not as well
understood as either of these phases.10
Macroscopic descriptions of solvents are based on physical properties including boiling
point, density, and surface tension among others, however, the individual, interacting solvent
molecules do not behave as a macroscopic continuum. Molecular scale solvent properties, such
as dipole moment and electronic polarizability provide a better description of the range of

3

processes operating at the microscopic level.10 Microscale solvent properties along with the
extent of solvent-solvent interactions lead to the distinctive organization found in solvents like
water. Liquid water forms dynamic three-dimensional hydrogen bonded networks, whereas
nonpolar solvents, such as hydrocarbons, are less structured due to their relatively weak solventsolvent interactions.11 The nature of the solvent-solvent and solute-solvent interactions is
important for a molecular scale understanding of the solvation of solutes. The molecular
interactions occurring in solutions can be divided into two broad classes; specific interactions,
where interactions occur between specific sites of interacting molecules, such as hydrogen
bonding, and non-specific interactions, which include van der Waals, as well as dipole-dipole
interactions.9 The interactions between dissimilar molecules, also referred to as cohesion forces,
originate from Coulombic forces between the nuclei and electrons of adjacent atoms and are
either attractive or repulsive in nature.12 It is these interactions that create the framework that
controls a solute’s local environment in solution, while also affecting adsorption and/or other
structural arrangement at phase boundaries.
Molecular rotation in liquid solutions provides a means for determining the extent of
intermolecular interaction. In order to gauge the dependence of rotational motion on
intermolecular interaction, time-resolved fluorescence spectroscopy is used to measure the
rotational diffusion dynamics of fluorescent probe molecules. The time-scale of fluorescence is
on the order of nanoseconds (10-9 s), which is sufficient time for interaction with other molecules
in solution.13 Molecular interaction during the excited state lifetime of a fluorescing species
leads to such phenomena as quenching14-17 and solvent relaxation,18-20 while the relatively long
duration of the excited state lifetime of the probe molecule allows for measurement of rotational
diffusion via time-resolved measurements.21

4

Our understanding of rotational diffusion in liquid solutions is based on hydrodynamic
theory and the foundation of this theory is the Debye-Stokes-Einstein (DSE) model.22 The DSE
model (Eq. 1.1) relates the orientational relaxation time, τOR, of spherical rotors in continuum
solvents to the viscosity of the medium (η), the thermal energy of the system (kBT), and
hydrodynamic volume of a rotating probe (V).22

τ OR =

ηV
k BT

(1.1)

Researchers have used this theory to describe rotational diffusion in systems ranging from neat
liquids to biological cell mimics. There are limitations to the DSE model and additional
modifications have been included to account for frictional interactions and the ellipsoidal shape
of non-spherical rotors.23,24 Deviations from the predictions of the DSE model can be exploited
as indicators of strong solute-solvent interactions in the form of electrostatic interactions and/or
hydrogen bonding.25-27 Frictional interactions between the solvent and the solute described in
terms of “slip” and “stick” boundary conditions define the limits within the DSE model and are
related to the strength of interaction or drag experienced by the rotating solute. The slip limit,
proposed Hu and Zwanzig,23 implies a solute rotates without frictional interactions with solvent
molecules and the torque experienced by the solute is only that required to displace the solvent
molecules within the volume swept out by the rotating solute.28 Alternately, the stick limit
implies greater frictional interaction between the solute and the solvent molecules making up the
solvation shell, which causes drag on the solute, slowing the rotational motion.28 Most studies
cite good agreement with hydrodynamic theory when reorientation times fall between the slip
and stick boundary conditions.
Two classes of probe molecules are utilized for rotational diffusion studies depending on
the system of study and information sought: polar (including charged) molecules and nonpolar
5

molecules. Nonpolar probes are often used to test hydrodynamic theories based on the
relationship between solute rotation and physical properties of the system.27,29-36 Interactions
between nonpolar solutes and nonpolar solvents are typically non-specific in nature, and solutes
reorient without large contributions from frictional interactions typically associated with specific
intermolecular interactions. Systems interrogated by polar probe molecules, where specific
interactions are common, often fail to show good agreement with hydrodynamic theory. The
rotational diffusion of polar probes has been assessed in various media, where ionic association,
hydrogen bonding, as well as structural features of probe molecules were reported to play a role
in dynamic solvation.37-41
Rotational diffusion dynamics are known to depend on the size of the solute
(hydrodynamic volume) relative to the size of the solvent, the shape of the solute, and the
viscosity of the solvent.42 Solute size effects become apparent as solute size approaches the size
of the solvent at which point, macroscopic diffusion theories break down, and a molecular scale
hydrodynamic theory is necessary to describe solute diffusion.30 Jiang et al., studied the effect of
solvent length on the rotational diffusion of a nonpolar probe, perylene, in a series of normal
alkanes.31 Semi-quantitative agreement with the modified DSE model was achieved and a
solvent chain length dependence was observed for the rotational diffusion of perylene. Solvent
length was shown to affect the boundary condition, where reorientation times in solvent
significantly smaller than perylene (n-alkanes: C5 – C8) was near the stick limit conditions, and
reorientation times in solvents significantly larger than perylene (n-alkanes: C9, C10, C12, C16)
approached the slip limit. In this study, an increase in the solvent chain length to the
approximate length of the solute molecule resulted in a change in boundary condition, which was
attributed to a fundamental change in the solute-solvent interactions for longer chain length

6

alkanes compared to shorter chain length alkanes. Size effects of solutes relative to solvent will
be addressed in chapter 2 in regards to the solvent, NOP.
Rotational diffusion dynamics of polar solutes in polar solvents have been shown to
exhibit changes in boundary conditions as well as changes in the friction experienced by the
solute due to the nature of the solvent. Dutt et al., studied four structurally similar polar probe
molecules, cresyl violet, nile blue, nile red, and oxazine 720, in various solvents including
amides, aprotic solvents, alcohols, polyalcohols, and water.28 In amides and the aprotic solvent
dimethyl sulfoxide (DMSO), reorientation of the probe molecules exhibits good agreement with
the stick model of hydrodynamic theory. In n-alcohols, additional friction, denoted dielectric
friction,25,26,43-45 alters the reorientation as a result of the dielectric properties of the solvent,
whereas, in polyalcohols, reorientation times lie between slip and stick boundary conditions.
Reorientation times of the polar chromophores, cresyl violet, nile blue, and oxazine 720 in water
showed good agreement with hydrodynamic theory and were reported to be affected the least by
dielectric friction. The greater friction between the polar probe molecules and alcohols
demonstrated increased strength of the solute-solvent interactions.
In forming the solvation structure, solvent-solvent interactions can also play a role in
mediating local organization. This phenomenon is apparent in systems forming secondary
structures, such as micelles, vesicles, and lipid bilayers. Studying local environments via
rotational diffusion dynamics necessitates the incorporation of fluorescent probes into more
complex systems to interrogate a specified region. Biological mimics exhibiting secondary
structures are composed of amphiphilic molecules allowing for the interrogation of the nonpolar
and polar regions separately. With the use of nonpolar probe molecules, studies have shown the
internal structures of micelles and vesicles differ based on size, composition, and fluidity of the

7

system.46,47 Additionally nonpolar probes have been used to interrogate changes to the
microviscosity of the aliphatic region of lipid vesicles upon perturbation.48,49 Probing polar
regions has allowed for the determination of intermolecular interactions as a function of
headgroup charge and interrogation of headgroup organization with different polar headgroup
structure.50,51 Setiawan et al. interrogated the polar headgroup region of lipid bilayers dosed
with varying concentrations of alcohol in order to assess the rotational dynamics and
morphology of the bilayer.52,53 This study on lipid bilayers was accomplished using a confocal
fluorescence lifetime and anisotropy imaging instrument, which allowed for scanning of surface
structure of the lipid bilayer. In chapter 4, this instrument is utilized for measuring depth profiles
adjacent to liquid phase boundaries.
The concept of solvent organization also relates to phase boundaries, where the ordering
of molecules in the liquid occurs in a way similar to the formation of a lipid bilayer. This
interfacial molecular ordering only persists for a few Angstroms into the bulk liquid.54
Molecular dynamic simulations have been used to show the interfacial structure at liquid-liquid
interfaces composed of immiscible and partially miscible liquids is molecularly sharp, with the
partially miscible system of water and 1-hexanol forming a bilayer structure extending a few
molecular lengths into the bulk.55 An interfacial width on the order of 1 nm56,57 requires
specialized measurement techniques in order to elucidate the solvation properties at liquid-liquid
and liquid-solid interfaces. In addition to the spatial restrictions, there are inherent difficulties in
differentiating interfacial from bulk phenomena at liquid-liquid interfaces. Molecular properties
at the interface differ from those measured in the bulk liquid due to an asymmetry of surface
forces that affects the structure and dynamics at the interface, however, many spectroscopic
techniques cannot separate the signal from adsorbed species at the interface and the species

8

present in the bulk liquid.58 Second order non-linear optical spectroscopic methods have been
used previously to probe liquid-liquid interfaces including second harmonic generation
(SHG),7,59-61 sum frequency generation (SFG),62-64 and total internal reflectance fluorescence
(TIRF) spectroscopy.65-69 These techniques are utilized due to their surface specificity to within
a few nanometers on either side of the interface.58
Although interfacial regions are reported to only extend on the order of nm, there are
phenomena that can be studied in the bulk adjacent to the interface. The Marken group utilized
electrochemistry and photoelectrochemistry to monitor ion transfer in a two-phase flow system
and at the liquid-liquid-solid interface, or three phase boundary, of a microdroplet.70-74 Three
phase boundaries are unique in that coupled processes such as electron transfer from the organic
phase to solid electrode and the anion transfer from the aqueous phase to the organic phase can
be assessed using electrochemical techniques.75 MacDonald et al., formed a biphasic liquidliquid flow system using NOP and aqueous electrolyte and were able to monitor redox reaction
currents in the reaction zone at the three phase, organic|aqueous electrolyte|electrode,
boundary.73 This study took advantage of the phase separation of water and NOP at room
temperature.
The solvent systems of water and NOP used for the research presented in subsequent
chapters were selected based on the success of the photoelectrochemical studies at three phase
boundaries. NOP exhibits unique solvation properties and has been incorporated into a number
of industrial formulations. Some important properties of NOP and are summarized here. NOP is
a nonionic surfactant and is one of a series of N-alkyl-2-pyrrolidones which exhibit high surface
activity in aqueous solutions.76 The structure of NOP (Fig. 1.1) contains a hydrophilic lactam
functionality that is capable of hydrogen bonding and an alkyl chain contributing to the

9

hydrophobic character. N-alkyl-2-pyrrolidones, including NOP, are used in a variety of
industrial processes and household products as non-toxic solvents (N-methyl) and wetting agents
(N-hexyl thru N-dodecyl).77 Binary temperature-composition phase diagrams show the water
and NOP form two phases at room temperature and NOP has a lower consolute temperature
(LCT) below 0 °C.78 The solubility of NOP in water is 0.12% (v/v) (20 °C),77 while no mutual
solubility is reported.
O

N

Figure 1.1 Structure of N-octyl-2-pyrrolidone (NOP).
Mixed micelles form when NOP is in the presence of ionic surfactants, such as sodium
dodecyl sulfate (SDS), however, due to the LCT, two phases form between NOP and aqueous
solution prior to reaching a critical micelle concentration (CMC).77 It is reported, pure NOP
micelles do not form because the length of the alkyl chains in NOP is insufficient to drive
formation of micelles, which requires aggregation of the amphiphile with the hydrophobic tails
pointing inward to form a hydrophobic core.79 The interaction between pyrrolidone headgroups
typically dominates and the formation of dimers and/or inverse micelles is more likely to occur
than traditional micelle formation. These unique properties make NOP a very interesting
medium to study in order to uncover the intermolecular interactions of this solvent through
incorporation of probe molecules and measurement of rotational diffusion dynamics.
In order to elucidate information on the organization and dynamics of fluorescent solutes
in unique environments, we collect fluorescence lifetime and fluorescence anisotropy decay data
using a time-resolved spectroscopic technique called time-correlated single photon counting

10

(TCSPC). Time-resolved measurements are made by using short pulses of light from a laser
source to excite the fluorescent probe and then recording the emission intensity as a function of
time. After excitation, the fluorescent probe emits photons, where the fluorescence lifetime is
the average amount of time required to transition from the first excited singlet state (S1) to the
ground singlet state (S0). The fluorescence intensity recorded after each excitation has a
probability to emit within a given period of time.21 After many excitation and emission events a
histogram builds up, which typically takes the form of an exponential decay (Fig 1.2). The
principles of single-photon counting are based on the assumption that the probability distribution
from detection of single photon emission built up after many excitation events is a representative
sampling of the actual emission intensity versus time distribution of all photons.80 Fluorescence
lifetime is determined from the total emission intensity, whereas fluorescence anisotropy decay
is calculated from the difference in intensity of two polarizations (parallel and perpendicular,
with respect to the vertically polarized excitation polarization) divided by the total emission
intensity. The lifetime decay in Fig. 1.2 is fit to a single exponential function defined by Eq. 1.2,

=
I (t ) I 0 exp(−t / τ fl )

(1.2)

where I(t) is the intensity as a function of time, I0 is the intensity at time zero, and τfl is the time
constant of the fluorescence intensity decay. Fluorescence lifetime is recorded using magic
angle conditions, where the sample is excited with vertically polarized excitation light and the
emission polarizer set to the magic angle, 54.7° with respect to excitation polarization.
Alternatively, the signal from emission transients collected at parallel and perpendicular
polarizations with respect to the excitation light are combined (  +2 ⊥ ) to obtain the total
fluorescence intensity.21 These two methods achieve equivalent results.

11

Figure 1.2 Time-resolved fluorescence lifetime decay for resorufin in aqueous solution.
The best fit line (red) is a one-component exponential decay function.
An example of parallel and perpendicular emission transients is shown in Fig. 1.3a, where
differences in the parallel and perpendicular counts are apparent at early times and the plots
converge over time. This behavior is representative of the depolarization of the initial nonrandom population created by exciting with vertically polarized light into a random population
with equal parallel and perpendicular counts at long times.80 The orientational anisotropy
function, R(t), is obtained from Eq. 1.3,

R(t ) =

I (t ) − I ⊥ (t )

I (t ) + 2 I ⊥ (t )

(1.3)

where the intensity of fluorescence polarized parallel with respect to the vertically polarized
excitation light is given by I (t ) and the fluorescence polarized perpendicular is given by I ⊥ (t ) .
12

Figure 1.3 (a) Parallel (black) and perpendicular (red) emission transients and (b)
orientational anisotropy function, R(t), for resorufin in aqueous solution. The best fit line
(red) is a one-component exponential decay function.

13

The orientational anisotropy decay in Fig. 1.3b is fit to a single exponential function defined by
Eq. 1.4,

=
R(t ) R0 exp(−t / τ OR )

(1.4)

where R(t) is the orientational anisotropy function, R0 is the initial anisotropy, and τOR is the
orientational relaxation time constant.
The work described in the subsequent chapters utilizes time-resolved fluorescence
spectroscopy to collect fluorescence anisotropy decay and fluorescence lifetime data as a means
of assessing the rotational diffusion dynamics of fluorescent probe molecules in various media.
We use the molecular rotation of these probes as a tool to achieve the objective of understanding
the structure and organization in solution phase media and in interfacial systems. In chapter 2,
we elucidate solvent-solvent and solute-solvent interactions in aqueous and organic bulk media
by measuring the rotational diffusion of structurally different probe molecules. In chapter 3, we
study the effect of charged additives and the role of ion pairing on the rotational motion of probe
molecules in aqueous electrolyte and NOP with added electrolyte. In chapter 4, we create a
depth profile of fluorescence lifetime and reorientation time constants in a thin film as a function
of distance from a liquid-solid interface and examine the composition of a colloidal suspension
in the thin film. In chapter 5, we study the effect of a liquid-liquid (aqueous-NOP) interface
formed normal to the liquid-solid interface on the droplet size and distribution in the colloidal
suspension.

14

REFERENCES

15

REFERENCES

1. Liquid-Liquid Interfaces: Theory and Methods; Volkov, A. G.; Deamer, D. W., Eds.; CRC
Press: Boca Raton, FL, 1996.
2. Nakaoka, S.; Surblys, D.; Yamaguchi, Y.; Kuroda, K.; Nakajima, T.; Fujimura, H. Chemical
Physics Letters 2014, 591, 306.
3. Volkov, A. G.; Deamer, D. W.; Tanelian, D. L.; Markin, V. S. Liquid Interfaces in Chemistry
and Biology; John Wiley & Sons, Inc.: New York, 1998.
4. Benjamin, I. Chemical Reviews 1996, 96, 1449.
5. Benjamin, I. Chemical Physics Letters 2009, 469, 229.
6. Perera, J. M.; Stevens, G. W. Analytical and Bioanalytical Chemistry 2009, 395, 1019.
7. Fedoseeva, M.; Richert, S.; Vauthey, E. Langmuir 2012, 28, 11291.
8. Interfacial Nanochemistry: Molecular Science and Engineering at Liquid-Liquid Interfaces;
Watarai, H.; Teramae, N.; Sawada, T., Eds.; Kluwer Academic/Plenum Publishers: New York,
2005.
9. Dutt, G. B. ChemPhysChem 2005, 6, 413.
10. Reichardt, C.; Welton, T. Solute-Solvent Interactions. In Solvents and Solvent Effects in
Organic Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA, 2010; pp 7.
11. Reichardt, C.; Welton, T. Classification of Solvents. In Solvents and Solvent Effects in
Organic Chemistry; Wiley-VCH Verlag GmbH & Co. KGaA, 2010; pp 65.
12. Intermolecular Forces: An Introduction to Modern Methods and Results; Huyskens, P. L.;
Luck, W. A. P.; Zeegers-Huyskens, T., Eds.; Springer Berlin Heidelberg: Berlin, Heidelberg,
1991.
13. Ingle, J. D., Crouch, S.R. Spectrochemical Analysis; Prentice Hall: New Jersey, 1988.
14. Biwersi, J.; Tulk, B.; Verkman, A. S. Analytical Biochemistry 1994, 219, 139.
15. Morandeira, A.; Furstenberg, A.; Gumy, J. C.; Vauthey, E. Journal of Physical Chemistry A
2003, 107, 5375.
16. Morandeira, A.; Furstenberg, A.; Vauthey, E. Journal of Physical Chemistry A 2004, 108,
8190.
17. Montejano, H. A.; Gervaldo, M.; Bertolotti, S. G. Dyes and Pigments 2005, 64, 117.
16

18. Weber, G.; Lakowicz, J. R. Chemical Physics Letters 1973, 22, 419.
19. Vincent, M.; Gallay, J.; Demchenko, A. P. Journal of Physical Chemistry 1995, 99, 14931.
20. Toptygin, D.; Brand, L. Chemical Physics Letters 2000, 322, 496.
21. Lakowicz, J. R. Principles of fluorescence spectroscopy, 3rd ed.; Springer: New York, 2006.
22. Debye, P. Polar Molecules Chemical Catalog Co.: New York 1929.
23. Hu, C. M.; Zwanzig, R. Journal of Chemical Physics 1974, 60, 4354.
24. Perrin, F. Journal of Physics and Radium 1934, 5, 497.
25. Wiemers, K.; Kauffman, J. F. Journal of Physical Chemistry A 1999, 104, 451.
26. Dutt, G. B.; Srivatsavoy, V. J. P.; Sapre, A. V. Journal of Chemical Physics 1999, 110,
9623.
27. Dutt, G. B. Journal of Chemical Physics 2000, 113, 11154.
28. Dutt, G. B.; Doraiswamy, S.; Periasamy, N.; Venkataraman, B. Journal of Chemical Physics
1990, 93, 8498.
29. Ben-Amotz, D.; Scott, T. W. Journal of Chemical Physics 1987, 87, 3739.
30. Ben-Amotz, D.; Drake, J. M. Journal of Chemical Physics 1988, 89, 1019.
31. Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 6436.
32. Jas, G. S.; Wang, Y.; Pauls, S. W.; Johnson, C. K.; Kuczera, K. Journal of Chemical Physics
1997, 107, 8800.
33. Dutt, G. B.; Krishna, G. R. Journal of Chemical Physics 2000, 112, 4676.
34. Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 6785.
35. Inamdar, S. R.; Mannekutla, J. R.; Mulimani, B. G.; Savadatti, M. I. Chemical Physics
Letters 2006, 429, 141.
36. Mize, H. E.; Blanchard, G. J. Journal of Physical Chemistry B 2013, 117, 16260.
37. Blanchard, G. J. Journal of Physical Chemistry 1991, 95, 5293.
38. Yu, J. W.; Berg, M. Chemical Physics Letters 1993, 208, 315.
39. Rasimas, J. P.; Blanchard, G. J. Journal of Physical Chemistry 1996, 100, 11526.
40. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2002, 106, 10718.
17

41. Esquivelzeta-Rabell, M.; Peon, J.; Cuevas, G. Journal of Physical Chemistry B 2009, 113,
8599.
42. Dutt, G. B.; Raman, S. Journal of Chemical Physics 2001, 114, 6702.
43. Madden, P.; Kivelson, D. Journal of Physical Chemistry 1982, 86, 4244.
44. Simon, J. D.; Thompson, P. A. Journal of Chemical Physics 1990, 92, 2891.
45. Balabai, N.; Sukharevsky, A.; Read, I.; Strazisar, B.; Kurnikova, M.; Hartman, R. S.;
Coalson, R. D.; Waldeck, D. H. Journal of Molecular Liquids 1998, 77, 37.
46. Stevenson, S. A.; Blanchard, G. J. Journal of Physical Chemistry B 2006, 110, 13005.
47. Lapinski, M. M.; Blanchard, G. J. Chemistry and Physics of Lipids 2008, 153, 130.
48. Pillman, H. A.; Blanchard, G. J. Journal of Physical Chemistry B 2010, 114, 3840.
49. Pillman, H. A.; Blanchard, G. J. Journal of Physical Chemistry B 2011, 115, 3819.
50. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry B 2003, 107, 7102.
51. Greenough, K. P.; Blanchard, G. J. Spectrochimica Acta Part A-Molecular and
Biomolecular Spectroscopy 2009, 71, 2050.
52. Setiawan, I.; Blanchard, G. J. Journal of Physical Chemistry B 2014, 118, 537.
53. Setiawan, I.; Blanchard, G. J. Journal of Physical Chemistry B 2014, 118, 3085.
54. Ding, F.; Hu, Z.; Zhong, Q.; Manfred, K.; Gattass, R. R.; Brindza, M. R.; Fourkas, J. T.;
Walker, R. A.; Weeks, J. D. Journal of Physical Chemistry C 2010, 114, 17651.
55. Wang, H. B.; Carlson, E.; Henderson, D.; Rowley, R. L. Molecular Simulation 2003, 29,
777.
56. Carpenter, I. L.; Hehre, W. J. Journal of Physical Chemistry 1990, 94, 531.
57. Steel, W. H.; Walker, R. A. Nature 2003, 424, 296.
58. Brevet, P. F.; Girault, H. H. Second Harmonic Generation at Liquid/Liquid Interfaces. In
Liquid-Liquid Interfaces: Theory and Methods; Volkov, A. G., Deamer, D. W., Eds.; CRC Press:
Boca Raton, FL, 1996.
59. Grubb, S. G.; Kim, M. W.; Rasing, T.; Shen, Y. R. Langmuir 1988, 4, 452.
60. Uchida, T.; Yamaguchi, A.; Ina, T.; Teramae, N. Journal of Physical Chemistry B 2000,
104, 12091.

18

61. Yamaguchi, A.; Uchidai, T.; Inai, T.; Nochi, K.; Teramae, N. Bunseki Kagaku 2006, 55,
457.
62. Messmer, M. C.; Conboy, J. C.; Richmond, G. L. Journal of the American Chemical Society
1995, 117, 8039.
63. Brown, M. G.; Walker, D. S.; Raymond, E. A.; Richmond, G. L. Journal of Physical
Chemistry B 2003, 107, 237.
64. Walker, D. S.; Richmond, G. L. Journal of Physical Chemistry C 2008, 112, 201.
65. Wirth, M. J.; Burbage, J. D. Journal of Physical Chemistry 1992, 96, 9022.
66. Sassaman, J. L.; Wirth, M. J. Colloids and Surfaces A-Physicochemical and Engineering
Aspects 1994, 93, 49.
67. Ishizaka, S.; Habuchi, S.; Kim, H. B.; Kitamura, N. Analytical Chemistry 1999, 71, 3382.
68. Ishizaka, S.; Nakatani, K.; Habuchi, S.; Kitamura, N. Analytical Chemistry 1999, 71, 419.
69. Ishizaka, S.; Kim, H. B.; Kitamura, N. Analytical Chemistry 2001, 73, 2421.
70. Collins, A. M.; Zhang, X. H.; Scragg, J. J.; Blanchard, G. J.; Marken, F. ChemPhysChem
2010, 11, 2862.
71. Collins, A. M.; Blanchard, G. J.; Hawkett, J.; Collison, D.; Marken, F. Langmuir 2011, 27,
6471.
72. Collins, A. M.; Blanchard, G. J.; Marken, F. Electroanalysis 2012, 24, 246.
73. MacDonald, S. M.; Watkins, J. D.; Gu, Y.; Yunus, K.; Fisher, A. C.; Shul, G.; Opallo, M.;
Marken, F. Electrochemistry Communications 2007, 9, 2105.
74. MacDonald, S. M.; Watkins, J. D.; Bull, S. D.; Davies, I. R.; Gu, Y.; Yunus, K.; Fisher, A.
C.; Page, P. C. B.; Chan, Y.; Elliott, C.; Marken, F. Journal of Physical Organic Chemistry
2009, 22, 52.
75. Marken, F.; Watkins, J. D.; Collins, A. M. Physical Chemistry Chemical Physics 2011, 13,
10036.
76. Rosen, M. J.; Zhen, H. Z.; Gu, B.; Murphy, D. S. Langmuir 1988, 4, 1273.
77. Login, R. B. Journal of the American Oil Chemists Society 1995, 72, 759.
78. Lou, A. J.; Pethica, B. A.; Somasundaran, P.; Yu, X. Journal of Colloid and Interface
Science 2002, 256, 190.
79. Login, R. B.; Chaudhuri, R. K.; Haldar, R. K.; Hashem, M. M.; Helioff, M. W.; Tracy, D. J.
Surface active lactams US Patent No. 5093031, 1992.
19

80. O'Connor, D. V.; Phillips, D. Time-Correlated Single Photon Counting; Academic Press,
Inc.: Orlando, FL, 1984.

20

Chapter 2
Solvent-Dependent Changes in Molecular Reorientation Dynamics: The Role
of Solvent-Solvent Interactions
2.1 Introduction
The interaction between dissimilar molecules is central to a wide range of chemical
phenomena. In the solution phase, such interactions are termed ‘solvation’ phenomena, which
mediate processes, such as chemical separations, heterogeneous and homogeneous catalytic
reactions, and organic syntheses. Despite the centrally important role of solvation to a variety of
chemical phenomena, there remains to be developed a comprehensive and predictive model of
solute-solvent interactions that takes into account the molecular structure and functionality of
both the solute and the solvent. To this point, there have been many studies of solvation, with a
host of inertial and dielectric phenomena being shown to contribute to solvent accommodation of
solute(s).1-12 Longer timescale interactions, typically evaluated using solute rotational diffusion
measurements have shown that van der Waals, dipole-dipole, and hydrogen-bonding interactions
can all contribute to the observed dynamics. In most cases, specific interactions between solute
and solvent are found to play important roles in mediating solute dynamics, but garnering soluteindependent information on the way solvents interact with like molecules has proven to be
elusive.
The interactions between solute and solvent are thought to depend on solvent molecular
structure and organization, and the rotational diffusion dynamics of the solute have been used as
a sensitive means of evaluating such interactions.5,13-33 For many systems, reorientation
dynamics can be treated in the context of the modified Debye-Stokes-Einstein (DSE) equation,3436

where solute motion, characterized by a reorientation time, is related to its hydrodynamic

volume, the bulk viscosity of the solvent, and the temperature of the system. The initial form of
21

this model is predicated on the assumption of a spherical solute rotating in a continuum solvent,
and the predicted form of the anisotropy decay for such a system is a single exponential decay.34
This model has been modified to include solvent shape and the frictional contribution to solutesolvent interactions in an effort to include ellipsoidal solutes and to provide better agreement
between experiment and theory.35,36 This model, with modifications, first discussed in chapter 1
(Eq. 1.1) has worked well to provide a qualitative picture of solute-solvent interactions for a
variety of systems, but there are notable instances where the anisotropy decay is characterized by
more than one exponential decay,17,20,21 and under such circumstances a more comprehensive
model of molecular reorientation is required. The theoretical treatment derived by Chuang and
Eisenthal allows for the characterization of the anisotropy decay function by up to five
exponentials, and these decays are related to the Cartesian components of the rotational diffusion
constant, D.37 Relating variations in D and, in particular, the relative values of Dx, Dy, and Dz,
provides insight into the shape of the volume swept out by the reorienting solute. The shape of
this volume is related to both the shape of the solute molecule and any constraints imposed by
the surrounding solvent system.
In this work, we report on the orientational relaxation behavior of two molecules,
resorufin and 6-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino] hexanoic acid (NBDHA) (Fig.
2.1), in two solvents, water and N-octyl-2-pyrrolidone (NOP). NOP has an amphiphilic structure
(Fig. 1.1) with a polar lactam functionality capable of hydrogen-bonding and a hydrocarbon
chain. Both solvents have potential for specific solute-solvent interactions, however, water and
NOP differ significantly in size and organization. We selected water and NOP for their ability to
form a liquid-liquid interface at room temperature, which is the focus of the study in chapter 5.

22

a

N

–
O Na+

O

O

y
x
OH

b

HN
O

N
O
N
NO2

Figure 2.1 Structures of (a) resorufin sodium salt and (b) NBD hexanoic acid (NBDHA)
shown with Cartesian x and y axes for clarity.
The two chromophores selected are similar in the sense that they are both polar, with several
heteroatom sites present in their conjugated ring structures. The overall shape and aspect ratio of
the two molecules differ significantly, and this difference has, in most previous works, led to
anisotropy decay dynamics that either cannot be compared directly or modeled using the same
solute-solvent boundary condition. The two chromophores we have examined here produce
decay dynamics that can be understood in the context of the Chuang and Eisenthal model,37 and
despite their dissimilar shapes, these molecules exhibit commensurate reorientation dynamics in
each solvent system. Specifically, we observe a single exponential anisotropy decay for both
23

chromophores in water, and a two-component exponential decay for both chromophores in NOP.
The observed changes in rotor shape for resorufin and NBDHA in the solvents, water and NOP,
demonstrate that solvent-solvent interactions and local organization can play a deterministic role
in mediating solute-solvent interactions. These findings are consistent with the steady-state
excitation and emission data for the solutes, and suggest that it is possible, with further
investigation, to achieve an understanding of solute-solvent interactions that relies to a
significant extent on the organization and properties of the solvent system.
2.2 Experimental Methods
The chromophore, NBDHA, was obtained from Invitrogen and used as received.
Resorufin sodium salt was purchased from Sigma Aldrich and used as received. NOP was
purchased from Aldrich at 98% purity and used without further purification. The water used in
these experiments was purified in-house with a Milli-Q filtration system (Millipore). For steadystate absorption measurements in both aqueous and NOP solvents, chromophore concentrations
between 10-4 and 10-5 M were used. These concentrations of NBDHA and resorufin in both
solvents were also used for time-resolved fluorescence measurements.
Steady-state absorbance measurements were made with a Cary 300 Bio, double-beam
UV-visible absorbance spectrophotometer over the range of 400 nm to 700 nm, at a spectral
resolution of 1 nm for all measurements. Steady-state spontaneous emission spectra were
recorded using a Jobin-Yvon Fluorolog III spectrometer over the range of 450 nm to 700 nm.
The excitation wavelength was set to 400 nm for both resorufin and NBDHA emission spectral
acquisition. The spectral resolution for emission experiments was 3 nm in all cases.
Time-resolved measurements, fluorescence lifetime and fluorescence anisotropy decay,
were collected using the spectrometer described below. A schematic of the time-correlated

24

single photon counting (TCSPC) set-up is shown in Figure 2.2. The light source for the system
is a CW mode-locked Neodymium:Yttrium Vanadate (Nd:YVO4) laser with 13 ps 1064 nm
pulses and 80 MHz repetition rate (Spectra Physics Vanguard). The Nd:YVO4 laser pumps a
cavity-dumped dye laser (Coherent 702-2) characterized by 5 ps pulses at wavelengths between
430 nm and 850 nm. The repetition rate of the laser is adjustable between 80 kHz and 80 MHz
and is determined by the cavity dumping electronics (Gooch & Housego). A typical repetition
rate is 4 MHz. The dye used to excite the NBDHA solutions was stilbene 420 (Exciton),
providing excitation wavelengths between 445 nm and 465 nm, while pyrromethene 567
(Exciton) with an excitation range of 550 nm to 595 nm was used to excite resorufin solutions.
The fundamental excitation pulse from the dye laser is split, with one portion directed to a
reference photodiode the other portion directed to the sample. The sample excitation pulse is
polarized vertically. Emission from the sample is collected using a 40X reflecting microscope
objective. The collected emission is separated into polarization components parallel and
perpendicular to that of the excitation pulse using a polarizing cube beamsplitter. The parallel
and perpendicular polarized signal components are detected simultaneously using Spectral
Products CM-112 subtractive double monochromators equipped with Hamamatsu R3809U-50
microchannel plate-photomultiplier tube (MCP-PMT) detectors. The detection electronics
(Becker & Hickl SPC-132) resolve the parallel and perpendicular transients separately, yielding
ca. 30 ps response functions for each detection channel. The detection electronics include a
time-to-amplitude converter (TAC) and constant fraction discriminator (CFD) that temporally
resolve the fluorescence signal for each polarization component. In Figure 2.2, a delay is used in
the reference channel to ensure proper timing of the reference and emission signals. Data are
collected using multichannel analyzers (MCA) that are integral components of the SPC-132

25

electronics. Data acquisition, detector bias, and collection wavelength are controlled using a
program written in-house using National Instruments LabVIEW® software on a PC.
13 ps pulses, 12 ns inter-pulse spacing
355 nm (2.5 W), 532 nm (2.5 W)

Nd:YVO4 laser
autocorrelator
Cavity-dumped
dye laser
polarization
focusing control
lens
sample

5 ps pulses, variable inter-pulse spacing
Tunable 430-850 nm (ca. 100 mW)

reference diode

∥

MCP-PMT

⊥

monochromator
polarizing
microscope
objective beamsplitter

REF

delay

CFD

TAC

CFD

TAC

MCA

Integrated detection
electronics

Figure 2.2 Time-correlated single photon counting (TCSPC) instrument.

26

2.3 Results and Discussion
In this work, we report on the steady-state spectroscopy and dynamics of the
chromophores resorufin and NBDHA in the solvents water and NOP. Both of these
spectroscopic measurements provide insight into the local environments of these probe
molecules, and we consider the steady-state optical response first. The absorption and emission
spectra of these two chromophores reveal a measurable dependence on solvent, and the
information obtained from the spectral characteristics can be used to infer information about the
chromophore local environment. For NBDHA in water (ε0 ~ 80, n = 1.33)38 (Figs. 2.3a), we
observe that both the absorption and emission spectra are red-shifted relative to the same data
taken in NOP solvent (Fig2.3b). NBDHA absorption maxima in water and NOP were 484 nm
and 469 nm, respectively, and the emission maxima were 555 nm and 529 nm in water and NOP,
respectively. While there is scant data in the literature on the dielectric response of NOP, we
infer from the dielectric constant for N-methyl pyrrolidone (ε0 ~ 32, n = 1.49) 38 that NOP has a
dielectric constant well below that of water. The red-shifted absorption and emission spectra
suggest that either the ground state of NBDHA is stabilized in NOP to a greater extent than it is
in water, or the excited state of NBDHA is stabilized more in water than in NOP. Given the
Stokes shift observed for NBDHA in both solvent systems, it appears that the excited state of
NBDHA is more polar than its ground state and is therefore stabilized by the more polar solvent,
which is predicted by spectroscopic theory.39 In comparison, the absorption and emission
spectra of resorufin in water (Fig. 2.4a) are slightly blue-shifted compared to the same data taken
in NOP (Fig. 2.4b), and the spectral shape and Stokes shift appears to be solvent-dependent for
resorufin to a greater extent than for NBDHA.

27

Absorption maxima for resorufin were 571 nm and 594 nm in water and NOP,
respectively, while the emission maxima were 585 nm and 616 nm in water and NOP,
respectively. For resorufin, either the excited state is stabilized to a greater extent in NOP than
in water or the ground state is stabilized more efficiently than the excited state in water. While
either of these is a possibility, it is instructive to consider that resorufin can exist either as a
contact ion pair with its sodium (Na+) counter ion, or as a dissociated ionic species, depending on
the ability of the solvent system to accommodate charged molecules.40

28

Figure 2.3 Normalized absorption and emission spectra of (a) NBDHA in water and (b)
NBDHA in NOP.

29

Figure 2.4 Normalized absorption and emission spectra of (a) resorufin in water and (b)
resorufin in NOP.

30

We consider the information available from the Stokes shifts for these chromophores in
the different solvents. There are several models in the literature that attempt to relate
fluorescence Stokes shift to molecular properties of the solute, such as its size and change in
dipole moment upon excitation, and the dielectric response of the solvent system.41-43 The
difference between these models lies in the manner in which the dielectric response of the
solvent is treated. We use the Bakhshiev formulation in this work,42

(
(

)
)

2
2∆µ 2  ε 0 − 1 n 2 − 1  2n + 1


∆ν = ν a −ν f =
−
a03hc  ε 0 + 2 n 2 + 2  n 2 + 2

(2.1)

where Δ v is the fluorescence Stokes shift, Δμ is the change in solute permanent dipole moment
upon excitation, a0 is the solute cavity radius, h is Planck’s constant, c is the speed of light, ε0 is
the zero-frequency dielectric constant of the solvent, and n is the solvent refractive index. The
quantity Δ v is measured experimentally and we use this value to determine the quantity of
interest, which is the change in dipole moment on excitation, Δμ. Using Eq. 2.1 we calculate Δμ
for the two chromophores in water and NOP (Table 2.1).
Table 2.1 Calculated changes in permanent dipole moment, Δμ, upon excitation for the
chromophores resorufin and NBDHA in solvents water and NOP.
Solvent

Resorufin
Δμ (D)

NBDHA
Δμ (D)

Water

1.02

1.63

NOP

0.58

1.67

D = Debye (3.335641 × 10–30 Coulomb × meter)44

31

These calculations, based on the experimental Stokes shift data (Figs. 2.3 and 2.4), show
that resorufin experiences a substantial change in its permanent dipole moment on excitation,
with the change being almost a factor of two greater for water (Δμ = 1.02 D) than for NOP (Δμ =
0.58 D). This is not a surprising result based on the fact that resorufin is a sodium salt and the
extent of dissociation is mediated by the solvent. Thus, in NOP we expect little or no
dissociation and we observe a spectral red shift and line narrowing (Fig. 2.3b) relative to that
observed for resorufin in water (Fig. 2.3a), where dissociation is expected. The chromophore of
NBDHA does not possess a dissociable proton, so we expect a different result. We find that for
NBDHA in both water and NOP, we recover a change in dipole moment of Δμ ~ 1.65, with no
difference between the different solvents. The spectral shifts and changes in band shape for both
chromophores in the two solvents suggest that the solvation environment formed by these two
solvents differs significantly. While these data are interesting and provide some insight into
solute-solvent interactions, we require a means of probing such local environments in more
detail. For this purpose we turn to time-resolved fluorescence spectroscopy.
As noted above, the two chromophores exhibit similar reorientation dynamics in the two
solvent systems, producing single exponential anisotropy decays in water and two-component
anisotropy decays in NOP (Figs. 2.5 and 2.6). The similarity between the reorientation behavior
of these two outwardly different chromophores bears closer examination because of its
implications on the role of solvent-solvent interactions in mediating solute-solvent interactions.
To compare these data, we need to understand in detail the information content of the
reorientation data. The starting point in the interpretation of our data is to generate the induced
orientational anisotropy function, R(t), from the combination of the polarized fluorescence
transients according to Eq. 2.2,

32

R(t ) =

I (t ) − I ⊥ (t )

I (t ) + 2 I ⊥ (t )

(2.2)

where I (t) is the fluorescence transient polarized parallel to the vertically polarized excitation
pulse and I ⊥ (t) is the fluorescence transient polarized perpendicular to the excitation pulse. It is
the functional form and the time constant(s) associated with the decay of R(t) that are related to
the local environment of the reorienting molecule. Chuang and Eisenthal developed a model for
fluorescence anisotropy decay that provides a means of interpreting anisotropy decay functions
containing up to five exponential components.5,29,45-47 Despite this potential for complexity, one
or two decays are the functional forms observed most commonly. For the excitation and
emission transition moments oriented parallel to one another, along the long axis in the
chromophore π-system plane, reorientation of the chromophore as a prolate rotor (Dx > Dy = Dz)
gives rise to a single exponential decay of R(t) (Eq. 2.3) and reorientation as an oblate rotor (Dz
> Dx = Dy) produces a two-component R(t) function, as described by Eq. 2.4.

=
R(t ) 4 exp(−6 Dz t )
10

R(t )= 1 exp(−(2 Dx + 4 Dz )t ) + 3 exp(6 Dxt )
10
10

(2.3)
(2.4)

With these relationships in mind, we consider the experimental data for resorufin and NBDHA in
water and NOP, with an eye toward understanding how the solvent local environment influences
the average diffusion constant, D.

33

Figure 2.5 (a) Orientational anisotropy function, R(t), plotted as a function of time for
resorufin in water. (b) R(t) as a function of time for NBDHA in water. The data are fitted
to single exponential decays.
34

Figure 2.6 (a) Orientational anisotropy function, R(t), plotted as a function of time for
resorufin in NOP. (b) R(t) as a function of time for NBDHA in NOP. The data are fitted to
two-component exponential decays.
35

The rotational diffusion behavior of resorufin and NBDHA in several solvents has been
studied previously,18,22,24,45,48 and there is a frame of reference for the interpretation of the data
we present here. Resorufin, a molecule capable of dissociation to produce an anionic
chromophore and Na+, is characterized by a transition dipole moment that lies along its long (x),
π-plane axis as shown in Figure 2.1a.48 Resorufin reorients as a prolate rotor in small polar
solvents such as methanol and acetonitrile, and has been shown to exhibit varying degrees of Na+
dissociation in the butanols.22,40 The NBD chromophore is used widely as a probe in biological
studies and has been attached to phospholipids or cholesterol for the interrogation of lipid bilayer
structure and dynamics.45-47 NBDHA (Fig. 2.1b) has a hexanoic acid pendant functionality and
reorients as a prolate rotor in solvents such as n-alcohols, acetonitrile, DMF and DMSO.45
Figure 2.7 depicts the different volumes swept out by prolate and oblate rotors that are polarized
along the long (x) axis. There is a limited range of solvents in which both chromophores have
been studied, and those data are compiled in Table 2.2. These data suggest that the reorientation
dynamics of resorufin and NBDHA bear a very limited resemblance to one another, and this
result is typical for head-to-head comparisons of the reorientation dynamics of most
chromophores. This chromophore-specific behavior may be expected due to the charge state of
the resorufin and NBDHA chromophores, and their significantly different shapes.

36

Figure 2.7 Three-dimensional shape representing the volume swept out when a long-axis
polarized chromophore (a) rotates about the x-axis as a prolate rotor and (b) rotates about
the z-axis as an oblate rotor.

37

Table 2.2 Comparison of reorientation dynamics of resorufin48 and NBDHA45 in selected
solvents.
η (cP)a

Resorufin48
τOR (ps)

NBDHA45
τOR (ps)

Methanol

0.544

78 ± 14

131 ± 24

Ethanol

1.074

216 ± 45

174 ± 6

1-propanol

1.945

361 ± 47

302 ± 16

1-butanol

2.544

434 ± 69

570 ± 26

Acetonitrile

0.369

34 ± 8

50 ± 3

DMF

0.794

54 ± 14

162 ± 11

DMSO

1.987

56 ± 10

273 ± 10

Solvent

a

Viscosity (25 °C) taken from CRC Handbook44
For the data we report here on NBDHA and resorufin reorientation in water and in NOP,

we find that the results are qualitatively consistent with the hydrodynamic volume of each
chromophore (165 Å3 for resorufin, 242 Å3 for NBDHA)49 in the sense that the anisotropy
decays for NBDHA are somewhat slower than for resorufin (Table 2.3). For the reorientation of
both chromophores in water, a single exponential anisotropy decay is observed.
Table 2.3 Anisotropy decay time constants, τOR, and Cartesian components of the average
rotational diffusion constants, D, for resorufin and NBDHA in water and in NOP.
Resorufin
τOR (ps)

Solvent
fast

NBDHA
τOR (ps)
slow

fast

slow

Water

133 ± 11

213 ± 63

NOP

347 ± 19

1432 ± 172

657 ± 97

1718 ± 395

Solvent

Dz /108 (Hz)

Dx(=Dy) /108 (Hz)

Dz /108 (Hz)

Dx(=Dy) /108 (Hz)

Water

12.6 ± 1.1

NOP

6.6 ± 0.4

8.4 ± 2.5
1.2 ± 0.2

3.4 ± 0.4

38

1.0 ± 0.2

Resorufin and NBDHA likely reorient as non-spherical rotors in water, but based on the limited
information content available from a single exponential decay (Eq. 2.3) compared to a twocomponent decay (Eq. 2.4), we are left to interpret our data in the context of the simplest model
that relates molecular reorientation to the properties of the solvent, the DSE model.34 In the most
basic form of this model, the recovered reorientation time constant, τOR, is related to the solvent
bulk viscosity, η, the solute hydrodynamic volume, V, and the thermal energy in the system, kBT
by the relation in Equation 2.5.

τ OR =

ηV
k BT

(2.5)

This formulation assumes a spherical rotor in a “continuum” solvent, and it is known that the
DSE model is a qualitative predictor of experimental data. For many systems, either
polar/charged solutes in polar solvents or nonpolar solvents, there exists a discrepancy between
model and experiment. This difference is frequently explained in the context of the solutesolvent boundary condition, i.e. the strength of frictional interactions between the solvent and
solute. Due to the structural difference between resorufin and NBDHA and the possible
difference in the charge state of the molecules, it would be reasonable to expect that there is a
difference in the solute-solvent boundary conditions for the two systems, where the boundary
condition denoted “stick” or “slip” is a measure of the frictional interaction between the solute
and the solvent.
We can evaluate whether or not this is the case by comparing the quantity τOR/V for the
two solutes. This comparison is valid because the solvent and the temperature are the same for
both reorientation measurements. For resorufin, τOR/V = 0.80 ± 0.07 ps/Å3 and for NBDHA,
τOR/V = 0.82 ± 0.27 ps/Å3. The similarity in these ratios indicates that the frictional

39

contribution(s) to solute-solvent interactions for these two solutes is fundamentally the same,
despite the structural differences between these two solutes.
The reorientation dynamics for the two chromophores in water may not be as revealing as
they could be because of the comparatively low viscosity and fast reorientation time constants
measured, however, these reorientation time constants we report for resorufin and NBDHA in
water are substantially longer than the instrumental response function for our TCSPC system (ca.
30 ps). As a result, time resolution is not a factor for our measurements and the similarity in
chromophore behavior in NOP is significant. Resorufin and NBDHA both produce a twocomponent exponential anisotropy decay in NOP. This finding indicates that in NOP, both
chromophores reorient as oblate rotors, and in water both chromophores reorient as prolate
rotors. From Eqs. 2.3 and 2.4 it is possible to extract the Cartesian components of the rotational
diffusion constant, D. The ratio of D, calculated from the average of the Cartesian components
for resorufin and NBDHA (Table 2.3), was found to be 1.67. This ratio is in reasonably close
agreement with the ratio of chromophore hydrodynamic volumes when taking into account the
DSE model where D is inversely proportional the hydrodynamic volume. There are two
important issues we consider based on these results. The first is the comparability of the results
for the two chromophores, which stands in contrast to many other cross-chromophore
comparisons, and the second point is that the change in rotor shape observed for both
chromophores in the two solvents implies that it is solvent local organization in at least one of
these solvents that is responsible for our findings.
It is instructive to examine the dynamics of these chromophores in the solvent NOP
because these data provide additional insight into the solvation behavior of both solutes. For
both chromophores, we observe a two-component anisotropy decay in NOP, consistent with the

40

functional form presented in Eq. 2.4. The two-component anisotropy decay for both
chromophores implies that they are reorienting as oblate rotors, which is a substantial change
from the behavior of the chromophores in water. The recovery of a two-component anisotropy
decay allows for the evaluation of the Cartesian components of D (Table 2.3). The ratio of Dz/Dx
for these chromophores is related to the ellipticity of the volume swept out during reorientation.
For resorufin, Dz/Dx = 5.5 ± 1.3 and for NBDHA, Dz/Dx = 3.4 ± 1.1. These values may appear
to be slightly different, but when uncertainties in the data are taken into account, they are not
distinguishable from one another. Given the different structures of these two chromophores, the
similarity in their reorientation dynamics in NOP suggests that the local environments in which
they reside bear some similarity.
As a result of the similar behavior of the two chromophores in both water and NOP, it is
useful to consider why both chromophores exhibit different effective rotor shapes in the two
solvents and also why such a difference in the observed dynamics may occur. To address the
first issue, we look to the relative sizes of the solute and solvent molecules. As noted above,
resorufin has a hydrodynamic volume of 165 Å3 and for NBDHA V = 242 Å3 calculated using
the method of atomic increments.49 The DSE model, which cannot be applied directly to our
NOP data for reasons of its functional form, has as a core assumption that the solvent behaves as
a dielectric continuum. This is approximated by the condition that individual solvent molecules
are much smaller than the solute. This condition is met for water but not for NOP. The
hydrodynamic volume of water is 21 Å3 and that of NOP is 219 Å3.49 Aside from a key
assumption of the DSE model being violated, the more significant point is that the volumes of
the chromophores are close to the same as that of NOP, and under that condition, solvent-solvent
interactions could become competitive with solute-solvent interactions in determining the local

41

organization of the solution. The fact that both chromophores reorient as oblate rotors (Fig.
2.7b) in NOP suggests that the local environment formed by the solvent has a propensity to
preferentially restrict out-of-plane rotation relative to in-plane rotation. The amphiphilic
properties of NOP point to a tendency to form organized clusters, but it is not appropriate,
necessarily, to view these data as indicating the formation of lamellar structure(s) in neat NOP.
Although it is apparent that the confinement imposed on the chromophores by NOP provides
more restriction for motions involving significant out-of-plane components.
The observation of a solvent-dependent change of rotor shape has been observed
previously. Cresyl violet, a cationic oxazine dye, was shown to reorient as a prolate rotor in
ethylene glycol and as an oblate rotor in n-dodecanol.17,50 In that work, it was shown that the
existence of a mesophase in n-dodecanol between 24 °C and 30 °C gave rise to solvent
organization that mediated the rotational diffusion dynamics of the chromophore. This effect
was observed due to the hydrodynamic volumes of n-dodecanol and the chromophore, cresyl
violet. Analogous behavior has been reported for nonpolar systems as well. The reorientation
dynamics of perylene and 1-methyl perylene were shown to change as the size of the n-alkane
solvent increased relative to the size of the chromophores.20,21 Thus, the existence of changes in
local solvation environment is known to occur in both polar and nonpolar systems. The unique
aspect of the work we present here is that the change in dynamics for two substantially different
chromophores as they are examined in two solvent systems can be understood in the context of a
single explanation; the solvent-solvent interactions for NOP mediate the local environment(s) in
which solutes can reside. In previous work, the results were only known to be chromophore
specific, and the relative importance of solute-solvent and solvent-solvent interactions could only
be speculated upon. In the present work, it is clear that solvent-solvent interactions determine

42

the local solvation environments that are possible in a given solvent. We were able to resolve
this issue in the present case because of the specific solvent system we used, but the result could
be more general with the caveat that the dominant role of solvent-solvent interactions will be
manifested when the solute is similar in size to the solvent.
2.4 Conclusions
We have examined the steady-state and time-resolved spectroscopic behavior of two
chromophores, resorufin and NBDHA, in the solvents water and NOP. The steady-state
spectroscopic data point to significantly different solute-solvent interactions for the two
chromophores in the different solvents, with resorufin exhibiting a different change in dipole
moment upon excitation for the two solvents and NBDHA not exhibiting the analogous change.
The time-resolved fluorescence anisotropy decay data show that both chromophores reorient as
prolate rotors in water, with the same frictional interactions. For reorientation in NOP, a
different rotor shape is observed, but both chromophores behave in the same manner, exhibiting
an oblate rotor shape. The change in rotor shape is correlated with the increase in the
hydrodynamic volume of the solvent relative to the solutes. Because of the similarity of the
dynamics for the two chromophores, solvent-solvent interactions are implicated as the mediating
factor in determining the local solvation environments that are allowed to form in this solvent.
In the next chapter, the reorientation dynamics of resorufin are evaluated in the presence of ionic
species.

43

REFERENCES

44

REFERENCES

1. Dote, J. L.; Kivelson, D.; Schwartz, R. N. Journal of Physical Chemistry 1981, 85, 2169.
2. Dote, J. L.; Kivelson, D. Journal of Physical Chemistry 1983, 87, 3889.
3. Madden, P.; Kivelson, D. Journal of Physical Chemistry 1982, 86, 4244.
4. Kivelson, D.; Spears, K. G. Journal of Physical Chemistry 1985, 89, 1999.
5. Dutt, G. B.; Doraiswamy, S.; Periasamy, N.; Venkataraman, B. Journal of Chemical Physics
1990, 93, 8498.
6. Alavi, D. S.; Hartman, R. S.; Waldeck, D. H. Journal of Chemical Physics 1991, 94, 4509.
7. Hartman, R. S.; Alavi, D. S.; Waldeck, D. H. Journal of Physical Chemistry 1991, 95, 7872.
8. Dutt, G. B.; Raman, S. Journal of Chemical Physics 2001, 114, 6702.
9. Kawski, A.; Bojarski, P. Asian Journal of Spectroscopy 2003, 7, 97.
10. Gayathri, B. R.; Mannekutla, J. R.; Inamdar, S. R. Journal of Fluorescence 2008, 18, 943.
11. Mannekutla, J. R.; Inamdar, S. R.; Mulimani, B. G.; Savadatti, M. I. Journal of
Fluorescence 2010, 20, 797.
12. Kawski, A.; Bojarski, P. Chemical Physics 2010, 375, 52.
13. Maroncelli, M.; Fleming, G. R. Journal of Chemical Physics 1987, 86, 6221.
14. Castner, E. W.; Maroncelli, M.; Fleming, G. R. Journal of Chemical Physics 1987, 86, 1090.
15. Nagarajan, V.; Brearley, A. M.; Kang, T. J.; Barbara, P. F. Journal of Chemical Physics
1987, 86, 3183.
16. Ben-Amotz, D.; Scott, T. W. Journal of Chemical Physics 1987, 87, 3739.
17. Blanchard, G. J. Journal of Chemical Physics 1987, 87, 6802.
18. Blanchard, G. J.; Cihal, C. A. Journal of Physical Chemistry 1988, 92, 5950.
19. Blanchard, G. J. Journal of Physical Chemistry 1989, 93, 4315.
20. Jiang, Y.; Blanchard, G. J. Journal of Physical Chemistry 1994, 98, 6436.
21. Goldie, S. N.; Blanchard, G. J. Journal of Physical Chemistry A 1999, 103, 999.
45

22. Spears, K. G.; Steinmetz, K. M. Journal of Physical Chemistry 1985, 89, 3623.
23. Alavi, D. S.; Waldeck, D. H. Journal of Physical Chemistry 1991, 95, 4848.
24. Kurnikova, M. G.; Balabai, N.; Waldeck, D. H.; Coalson, R. D. Journal of the American
Chemical Society 1998, 120, 6121.
25. Dutt, G. B.; Srivatsavoy, V. J. P.; Sapre, A. V. Journal of Chemical Physics 1999, 110,
9623.
26. Dutt, G. B.; Ghanty, T. K. Journal of Physical Chemistry A 2004, 108, 6090.
27. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2002, 106, 10718.
28. Kubinyi, M.; Grofcsik, A.; Papai, I.; Jones, W. J. Chemical Physics 2003, 286, 81.
29. Indirapriyadharshini, V. K.; Ramamurthy, P. Journal of Chemical Sciences 2007, 119, 161.
30. Zhou, P. W.; Song, P.; Liu, J. Y.; Shi, Y.; Han, K. L.; Het, G. Z. Journal of Physical
Chemistry A 2008, 112, 3646.
31. Mali, K. S.; Dutt, G. B.; Mukherjee, T. Journal of Chemical Physics 2008, 128.
32. Esquivelzeta-Rabell, M.; Peon, J.; Cuevas, G. Journal of Physical Chemistry B 2009, 113,
8599.
33. Zhou, P.; Song, P.; Liu, J.; Han, K.; He, G. Physical Chemistry Chemical Physics 2009, 11,
9440.
34. Debye, P. Polar Molecules Chemical Catalog Co.: New York 1929.
35. Hu, C. M.; Zwanzig, R. Journal of Chemical Physics 1974, 60, 4354.
36. Perrin, F. Journal of Physics and Radium 1934, 5, 497.
37. Chuang, T. J.; Eisenthal, K. B. Journal of Chemical Physics 1972, 57, 5094.
38. CRC Handbook of Chemistry and Physics, 71st Edition; Lide, D. R., Ed.; CRC Press: Boca
Raton, FL, 1990-1991.
39. Lakowicz, J. R. Principles of fluorescence spectroscopy, 3rd ed.; Springer: New York, 2006.
40. Rasimas, J. P.; Blanchard, G. J. Journal of Physical Chemistry 1996, 100, 11526.
41. Lippert, E. Zeitschrift für Elektrochemie 1957, 61, 962.
42. Bakhshiev, N. G. Optika I Spektroskopiya 1964, 16, 821.

46

43. Chamma, A.; Vaillet, P. Comptes Rendus de l'Académie des Sciences - Series IIC Chemistry 1970, 270, 1901.
44. CRC Handbook of Chemistry and Physics, 94th Edition; Haynes, W. M., Ed.; CRC
Press/Taylor and Francis: Boca Raton, FL, 2013-2014.
45. Greenough, K. P.; Blanchard, G. J. Journal of Physical Chemistry B 2006, 110, 6351.
46. Greenough, K. P.; Blanchard, G. J. Journal of Physical Chemistry A 2007, 111, 558.
47. Greenough, K. P.; Blanchard, G. J. Spectrochimica Acta Part A-Molecular and
Biomolecular Spectroscopy 2009, 71, 2050.
48. Gudgin Templeton, E. F.; Quitevis, E. L.; Kenney-Wallace, G. A. Journal of Physical
Chemistry 1985, 89, 3238.
49. Edward, J. T. Journal of Chemical Education 1970, 47, 261.
50. Blanchard, G. J.; Wirth, M. J. Journal of Physical Chemistry 1986, 90, 2521.

47

Chapter 3
The Effects of Electrolyte Concentration on the Rotational Dynamics of
Resorufin
3.1 Introduction
The study of fluorescent probe rotational diffusion dynamics in solution is an important
tool in an ongoing effort to understand solvation phenomena. Rotational diffusion of a probe
molecule is sensitive to both solute-solvent and solvent-solvent interactions, and it is primarily
the interactions between solute and solvent that mediate processes such as chemical separations
and catalytic reaction efficiency. The interactions between a solute and its local environment
include van der Waals interactions, dipole-dipole, and hydrogen-bonding interactions, among
others, and the molecular structures of the solute and solvent determine which of these forces
play significant roles in the observed solute behavior.1-7 The introduction of ionic species into a
solution increases the number of interactions that can potentially occur, especially when the
solute is also a charged species. A phenomenon known as dielectric friction is often cited as
contributing to the rotational diffusion dynamics of cationic and anionic solutes.3-5,8-19 The
frictional forces experienced by a polar or charged solute in solution are reported to be a
combination of both hydrodynamic friction due to in part to the size of the solute and dielectric
friction caused by electrostatic interactions, although the contribution of the latter phenomenon is
typically modest. Rotational diffusion studies of solutions with added electrolyte have reported
changes in chromophore dynamics citing alteration of the solvent local organization in the
vicinity of the chromophore and in some cases the formation of contact ion pairs.20-26
The most widely used model to treat rotational diffusion is the modified Debye-StokesEinstein (DSE) model first introduced in chapter 1 and shown in Eq. 3.1,27-29

48

τ OR
=

1
ηVf
=
6 D k BTS

(3.1)

which relates the rotational diffusion time constant, τOR, to the solute hydrodynamic volume, V,
the solvent bulk viscosity, η, Boltzmann constant, kB, and the system temperature, T.
Additionally, the modified DSE model, which forms the basis of hydrodynamic theory of
rotational diffusion, takes into account, a frictional term, f, reflecting the solute-solvent boundary
condition,28 and a shape factor, S, to account for the non-spherical shape of the reorienting
moiety.29 The reorientation time, τOR, which is an experimentally measured quantity, is also
inversely proportional to the average rotational diffusion constant, D. The original form of the
DSE equation (Eq. 2.5) assumes spherical rotors, whereas the modified DSE equation (Eq. 3.1)
accounts for an ellipsoidal rotor executing rotational motion in a continuum solvent. Despite the
obvious oversimplification of the solute-solvent system, the modified DSE model has been
shown to provide remarkably close agreement with experimental data. Among the issues that are
most limiting to the DSE model is the assumption of a solute-solvent boundary condition, and
the approximation that the chromophore sweeps out a spherical volume as it reorients in solution.
The terms f and S that are included in the modified form of the DSE equation account for the
solute-solvent frictional boundary condition28 and the shape of the volume swept out by the
solute.29 The coefficient f ranges from 0 and 1, where 0 is the “slip” boundary with minimal
solute-solvent interactions and 1 the “stick” limit, characterized by significant frictional solutesolvent interactions.28 The shape factor, S, derived from Perrin’s equations, accounts for the
non-spherical shape of both the solute and solvent.29 Incorporation of frictional and shape
factors into the DSE model provides reasonable agreement between experiment and theory, but
cannot account quantitatively for rotational diffusion data because the solvent is treated as a
continuum fluid. The DSE model also is not capable of accounting directly for two-component
49

anisotropy decays, and the incidence of this functional form requires consideration of factors that
are not included in the assumptions of the model. For example, in cases where the
hydrodynamic volume of the solvent equals or exceeds that of the solute, the approximation of a
continuum solvent is violated, and solvent organization can exert a measurable influence on the
shape of the volume swept out by the rotating solute. The solute-solvent system of resorufin
(Fig. 2.1a) in N-octyl-2-pyrrolidone (NOP) (Fig. 1.1) exhibited a two-component exponential
decay, which was interpreted in the context of the solute sweeping out an oblate ellipsoidal
volume because the solvent-solvent interactions of NOP hindering out-of-plane chromophore
rotation.30 The rotational diffusion of resorufin in NOP contrasts with a single exponential
anisotropy decay observed for resorufin in water, where the approximation of a continuum
solvent is violated less seriously. Since solvent organization plays a significant role in mediating
resorufin rotational motion, it is important to understand the potential role of solution phase ionic
species in mediating chromophore rotational motion.
The work presented here addresses the role of electrolyte (ionic species) in mediating the
rotational diffusion behavior of resorufin in two different solvent systems, water and NOP. As
was shown in chapter 2, the local environment of resorufin in NOP is characterized by
organization determined substantially by solvent-solvent interactions, while the second system,
resorufin in water, manifests a chromophore local environment where solute-solvent interactions
play an important role. In the presence of ionic species, resorufin reorients as a prolate rotor,
exhibiting a single exponential anisotropy decay in both water and NOP. This behavior
represents a change in the effective rotor shape of resorufin in NOP upon the addition of
electrolyte. Three electrolytes were selected to study the effect of complexation between
electrolyte cations and the sodium salt of resorufin. The role of ion pairing between resorufin

50

anion, which is present in aqueous solution is probed by addition of lithium perchlorate (LiClO4),
an electrolyte utilized in aqueous electrochemical studies. Two quaternary ammonium bromide
non-aqueous electrolytes, tetrabutylammonium bromide (TBAB) and tetraoctylammonium
bromide (TOAB) (Fig. 3.1), which contain bulky cations are added to NOP solutions in an
attempt to disrupt the solvation shell created by NOP. In water, the lithium cation (Li+) from
LiClO4 can replace the sodium counter-ion of resorufin without significant perturbation,
however, the size and shape of the quaternary ammonium cations used in NOP solutions gives
rise to a pronounced change in resorufin reorientation dynamics. Since the hydrodynamic
volumes of TBAB and TOAB exceed that of NOP, the solvent-solvent interactions reported to
mediate the reorientation of resorufin are disrupted when the cationic ammonium species are
associated with the anionic chromophore. We use the Chuang and Eisenthal model31 to interpret
the decay dynamics of resorufin in water and NOP in the presence of electrolyte, and the
modified DSE equation to estimate the equilibrium constant for complexation. Our data point to
measurable complexation between the anionic chromophore and the cationic ammonium species.

51

a
–

Br
+
N

b
–

Br
+
N

Figure 3.1 Non-aqueous electrolytes, (a) tetrabutylammonium bromide (TBAB) and (b)
tetraoctylammonium bromide (TOAB), with hydrodynamic volumes of 299 Å3 and 571 Å3,
respectively.

52

3.2 Experimental Methods
The sodium salt of resorufin was purchased from Sigma Aldrich and used as received.
N-octyl-2-pyrrolidone (NOP) was purchased from Aldrich at 98% purity and used without
further purification. The water used in these experiments was purified in-house with a Milli-Q
filtration system (Millipore). Lithium perchlorate (LiClO4) was purchased from Aldrich at
99.99% purity and used as received. The two non-aqueous electrolytes, tetrabutylammonium
bromide (TBAB) and tetraoctylammonium bromide (TOAB) (Aldrich), were used without
further purification. Aqueous solutions of LiClO4 were prepared with electrolyte concentrations
between 5 × 10-3 M and 10-1 M. TBAB and TOAB were dissolved in NOP with electrolyte
concentrations between 5 × 10-4 M and 10-1 M. For steady-state absorption measurements in
both aqueous and NOP solvents, chromophore concentrations of 10-5 M were used. The
resorufin solution (10-5 M) was mixed with the electrolyte solution at 50% (v/v) for both steadystate and time-resolved fluorescence measurements.
Steady-state absorbance measurements were performed using a Hitachi U-4001, doublebeam UV-visible absorbance spectrophotometer over the range of 400 nm to 700 nm, at a
spectral resolution of 1 nm for all measurements. Steady-state spontaneous emission spectra
were recorded using a Hitachi F-4500 fluorescence spectrophotometer over the range of 500 nm
to 700 nm for sample in water and 530 nm to 700 nm for samples in NOP. The excitation
wavelengths were set to 490 nm and 520 nm for the emission spectral acquisition of resorufin in
water and resorufin in NOP, respectively.
Time-resolved measurements, fluorescence lifetime and fluorescence anisotropy decays,
were collected using the time-correlated single photon counting (TCSPC) system described in
detail in section 2.2. The dye used to excite resorufin solutions was pyrromethene 567 (Exciton)

53

dissolved in 1-phenoxy-2-propanol (Exciton), providing excitation wavelengths in the range of
550 nm to 595 nm.
3.3 Results and Discussion
We report here on the effects of electrolyte concentration on the reorientation dynamics
of resorufin in water and in NOP. Due to the differences in these bodies of data, we consider the
NOP and aqueous solutions separately. Both systems are characterized by measurable
interactions between the anionic chromophore and the cationic species present in solution. It is
the difference in the sizes of the cations that are primarily responsible for the different
reorientation dynamics observed in these two systems.
NOP Solutions. Resorufin is known to exhibit single exponential anisotropy decay
dynamics in water and a two-component anisotropy decay in NOP with reorientation time
constants in neat solvent.30 In the work presented in chapter 2, we found that solvent-solvent
interactions between NOP molecules dominated the environment formed around the resorufin
chromophore, mediating its reorientation dynamics by restricting rotation to primarily about the
axis perpendicular to the chromophore π-system plane (z-axis, Fig. 2.1). The addition of
electrolytes, TBAB and TOAB, with hydrodynamic volumes larger than that of the solvent
affects solvent organization and also gives rise to ionic interactions between the electrolyte and
the chromophore, resulting in a change in the effective rotor shape of resorufin. The rotational
decay times of resorufin in NOP exhibit an electrolyte concentration dependence in solutions
containing TBAB and TOAB.
The induced orientational anisotropy function, R(t) is generated from the fluorescence
transients polarized parallel, I||(t), and perpendicular, I⊥(t), to the vertically polarized excitation
pulse according to Eq. 2.2. It is the functional form of R(t) that contains information that can be

54

related to the local environment of the reorienting species. Chuang and Eisenthal have shown
that R(t) can contain up to five exponential decay components.31 The most commonly observed
functional forms are one- and two-component decays. The relationship between the number of
decays and the time constant(s) recovered from R(t) depends on the orientation of the excited
and emitting transition dipole moments relative to the molecular structure. For the transition
moments oriented along the long (x) molecular axis, the functional form of R(t) is related to the
Cartesian components of the solute rotational diffusion constant according to Eqs. 2.3 and 2.4,
where Dx, Dy, and Dz are the Cartesian components of the rotational diffusion constant, D. We
assign the x-axis to be the long axis in the chromophore π-system plane of resorufin, where y is
taken to be short in-plane axis and z is the out-of-plane axis (Fig. 2.1). For a prolate rotor, the
dominant motion is rotation about the x-axis (Dx > Dy = Dz) and for the oblate rotor the dominant
rotation is about the z-axis (Dz > Dx = Dy). The two-component decay for resorufin in NOP
changes to a single exponential decay when a quaternary ammonium salt is added to the solution.
This change in decay functionality corresponds to a change in the rotor shape of the reorienting
moiety. We assert that this change is caused by interactions between the chromophore and the
quaternary ammonium ions, and rationalize our findings as follows. The notion that the anionic
chromophore can interact with the cationic ammonium species is based on simple ionic
considerations. The change in rotor shape associated with this putative interaction is based on
the relative sizes of the species present in solution. The chromophore is the smallest of the
molecules in solution with a hydrodynamic volume of 165 Å3. The NOP solvent molecules have
hydrodynamic volumes of 219 Å3,32 and the organization of the comparatively large solvent
molecules around the chromophore mediates its reorientation dynamics, leading to the observed
two-component anisotropy decay. With the addition of electrolyte, any association between the

55

resorufin anion and the ammonium cation will perturb the chromophore local environment. The
hydrodynamic volumes of the non-aqueous electrolytes are 299 Å3 for TBAB and 571 Å3 for
TOAB,32 therefore association of the chromophore and electrolyte will give rise to a substantial
change in the volume and shape of the reorienting species. A visual representation of our
assertion (Fig. 3.2) shows the rotational dynamics of resorufin change from the motion of an
oblate rotor to a prolate rotor after forming a complex with a cation with a large hydrodynamic
volume.
The hydrodynamic volume of the electrolyte relative to that of the chromophore can have
a major influence on the local environment experienced by the reorienting moiety and we have
shown that this effect is dependent on the concentration of the electrolyte. As the electrolyte
concentration increases, the single exponential anisotropy decay exhibits an increasing
reorientation time constant for both TBAB/NOP and TOAB/NOP solutions (Fig. 3.3). There are
several possible explanations for this observation.

z

+
Resorufin anion

x
y
Complex

Ammonium cation

Figure 3.2 Illustration of the resorufin anion reorienting as an oblate rotor with diffusion
primarily about the z-axis in neat NOP and sweeping out a volume representative of a
prolate rotor after formation of a complex with a tetrabutyl- or tetraoctylammonium
cation.

56

Figure 3.3 Reorientation time of the reorienting moiety as a function of electrolyte
concentration (log10 scale) for resorufin in TBAB/NOP (filled circles) and TOAB/NOP
(open circles). Error bars represent ± 1 standard deviation.

57

It is instructive to examine these data using the modified DSE model for rotational
diffusion.27-29 The modified DSE equation relates the reorientation time to the properties of the
solute and solvent (Eq. 3.1). In the context of this equation, the electrolyte concentrationdependent increase in the reorientation time constant, τOR, could be the result of an increase in
local viscosity, an increase in the hydrodynamic volume of the reorienting moiety, or a decrease
in temperature. The third possible explanation is ruled out because the experiments are
performed at constant temperature. The friction factor, f, and shape factor, S, terms may also
change with a change in electrolyte concentration but they would do so in concert with either η
or V, respectively. The terms η and V can be evaluated within the framework of experimental
data and existing, well established models. Balabai et al. showed that the reorientation times of
resorufin exhibit a concentration dependence in electrolyte/methanol and electrolyte/DMSO
solutions when normalized to the solvent viscosity.22 Their results point to a change in the
hydrodynamic volume of the reorienting entity in the presence of electrolyte (or, more
accurately, a change in V/S), and this finding can be accounted for in the context of complex
formation between the anionic resorufin chromophore and the cationic electrolyte. For the
experiments we report on here, the cations are quaternary amines with relatively large
hydrodynamic volumes compared to the chromophore volume. If there is a direct contact ion
pair complex formed, such a species would experience significantly slower rotational diffusion
owing to its larger hydrodynamic volume. Based on our observation of the dependence of
reorientation time on electrolyte concentration, we can use these data to estimate the equilibrium
constant for the complex.
The hydrodynamic volume of the putative complex is given by the sum of the
hydrodynamic volumes of resorufin and the cationic electrolyte. By using the DSE equation to

58

determine the effective hydrodynamic volume of the reorienting moiety as a function of
electrolyte concentration, we can estimate the relative contributions of the free and complexed
chromophore to each observed reorientation time. This interpretation assumes that the complex
formation is a dynamic equilibrium with a characteristic exchange time on the order of the
reorientation time or faster. For the determination of the complex equilibrium constant, we
assume that the resorufin-electrolyte complex is characterized by 1:1 stoichiometry (Eq. 3.2),
keq 

R  E
 R  E 

(3.2)

where [R–E] is the concentration of resorufin-electrolyte complex, and [R] and [E] are the
concentrations of resorufin and the electrolyte, respectively. In order to solve for the equilibrium
constant, keq, we use the concentrations of electrolyte in Table 3.1 and designate the ratio of
resorufin-electrolyte complex to free resorufin to be (1–x)/x, where (1–x) and x are used as
weighting factors for the complexed and free forms of resorufin, respectively. In this model, we
determine the quantities x and 1–x at each electrolyte concentration using Eq. 3.3,

τ OR
= ( x)τ R + (1 − x)τ R − E

(3.3)

where τR represents the reorientation time for uncomplexed resorufin and τR–E is the
reorientation time for the resorufin–electrolyte complex. The key to the accuracy of this model
is the ability to evaluate τR and τR–E.
There are two approaches to the evaluation of τR and τR–E. The first is to use modified
DSE model parameters to achieve agreement, or alternatively to use an empirical factor to
achieve agreement between model and experiment. Regardless of the methodology used, the
functional dependence of the model used on the hydrodynamic volume of the reorienting moiety
does not change.
59

Table 3.1 Reorientation times, τOR, of resorufin in neat NOP and NOP with various
concentrations of TBAB or TOAB electrolytes.
Electrolyte
Concentration
(M)

Neat NOP30
τOR (ps)
fast
slow

0

347 ± 19 1432 ± 172

NOP w/
TBAB
τOR (ps)

NOP w/
TOAB
τOR (ps)

5 × 10-4

474 ± 10

545 ± 6

1 × 10-3

449 ± 15

559 ± 8

5 × 10-3

479 ± 10

521 ± 16

1 × 10-2

593 ± 4

608 ± 15

5 × 10-2

669 ± 9

644 ± 16

1 × 10-1

700 ± 12

781 ± 17

In the first case, if we assume a stick limit solute-solvent boundary condition (f = 1), the
agreement between model predictions and experimental data is poor, which is not surprising.
For the vast majority of polar and ionic species in water or other polar solvents, the measured
reorientation time constants exceed those predicted by the modified DSE model in the stick limit.
The reasons for this behavior have not been resolved fully, but it is typically held that strong,
associative solute-solvent interactions are responsible. In this case, reaching agreement between
experiment and model is best achieved through the use of an empirical correction factor. For the
experiments reported here, we find that the empirical factor required to achieve agreement
between model and experiment for τR is (f/S = 2.49). The molecular dimensions of resorufin
lead to a shape factor, S, of 0.768, implying a value of f = 1.91, which exceeds the limits of the
model. Using this same correction factor and the sum of the hydrodynamic volumes of resorufin
and TOAB allow for the calculation of τR–E. The theoretical reorientation time calculated for
free resorufin using f/S = 2.49 is 555 ps, and the calculated value of τR–E for complexation with
TOAB is 2476 ps. Using these values for τR and τR–E, and the τOR data given in Table 3.1, we
60

apply Eqs. 3.2 and 3.3 to extract the equilibrium constant, keq = 2 ± 1 for the resorufin-TOAB
complex. Applying the same data analysis to resorufin in TBAB/NOP solution yields a value of
keq = 11 ± 8. For both systems, there is significant uncertainty associated with the
experimentally determined values of keq, but the key point is that these values are modest.
Resorufin forms a stronger complex with the tetrabutylammonium cation than with the
tetraoctylammonium cation, consistent with screening effects associated with the length of the
cation alkyl chains of the tetraoctylammonium cation. We have examined the steady-state
spectra of resorufin in NOP as a function of electrolyte concentration and have also used semiempirical calculations (Hyperchem®) to evaluate the likely intermolecular interactions that are
responsible for the formation of the complexes we have observed.
The existence of the complex has little effect on the steady-state absorption and emission
spectra of resorufin in TBAB/NOP and TOAB/NOP solution. The normalized absorption
spectra (Fig. 3.4a) and emission spectra (Fig. 3.4b) exhibit only minor spectral shifts with the
addition of TOAB. There are subtle changes in the vibronic band intensities for both the
absorption and emission spectra, which are suggestive of the interaction of the chromophore with
the cation, but the subtle intensity variations of these bands do not allow for a detailed
understanding of the nature of the interaction(s). In conjunction with semi-empirical calculations
(not shown), where the most stable complex is between the resorufin terminal carbonyl groups
and the tetraalkylammonium cation, we can infer that the spectral changes shown in Fig. 3.4 are
related primarily to the chromophore C=O groups.

61

Figure 3.4 Normalized (a) absorption and (b) emission spectra of resorufin in neat NOP
(black), 10-3 M TOAB (red), 10-2 M TOAB (blue), and 10-1 M TOAB (green).
62

The semi-empirical calculations indicate that for interactions between the cation and the
chromophore, the closest approach of the cation is ca. 4.5 Å if the approach is along the
chromophore z-axis (perpendicular to the π-plane). For the cation approaching along the
chromophore x-axis (long in-plane axis), the optimum interaction distance is calculated to be ca.
3.2 Å. We hasten to point out that these results should not be taken as proof of any complex
geometry, but rather that chromophore-cation interactions in the π-plane of the chromophore are
favored over interactions perpendicular to the chromophore π-plane. This result is consistent
with the reorientation data, where the chromophore reorients as a prolate rotor in the presence of
the electrolyte. Complexation with the chromophore end group(s) would serve to diminish
rotation about the z-axis relative to rotation about the x-axis, making rotation about the x-axis the
dominant rotational motion (Fig. 3.2).
Aqueous solutions. In contrast to the results obtained for resorufin in electrolyte/NOP
solution, the reorientation dynamics of resorufin in aqueous electrolyte solutions exhibit a
decrease in the reorientation time constant with increasing electrolyte concentration (Fig. 3.5).
The reorientation time constant ranges from 130 ps at 5 × 10-3 M LiClO4 to 100 ps for 10-1 M
LiClO4 (Table 3.2). A comparison of these data to the reorientation time constant for resorufin in
water (133 ps, Table 3.2), shows that the low concentrations of electrolyte have a negligible
effect on the reorientation of resorufin.30 Previous studies on the reorientation of resorufin in Li+
containing electrolyte solution show the reorientation times in water increasing, but the effect of
concentration is much less significant in water compared to other solvents, such as methanol and
DMSO.20,22 The reorientation time of resorufin increased by almost 400% for neat DMSO vs.
1.0 M LiNO3/DMSO and 15% for water vs. 1.0 M LiNO3/water solution.20 The relatively subtle

63

change for resorufin in water that we report here suggests that the concentration dependence is
not linearly dependent on electrolyte concentration (Fig. 3.5).

Figure 3.5 Reorientation time as a function of electrolyte concentration (log10 scale) for
resorufin in aqueous LiClO4. Errors bars represent ± 1 standard deviation.

64

Table 3.2 Reorientation times, τOR, of resorufin in water and various concentrations of
LiClO4 (aq).
Electrolyte
Concentration
(M)

Water30
τOR (ps)

0

133 ± 11

LiClO4 (aq)
τOR (ps)

5 × 10-3

126 ± 4

1 × 10-2

131 ± 4

5 × 10-2

100 ± 3

1 × 10-1

102 ± 3

Our data are consistent with complexation between the resorufin chromophore and Li+,
where for low concentrations the free anionic chromophore is observed and for higher Li+
concentrations, a Li+-resorufin complex is present. We attribute the faster reorientation time of
the Li+-resorufin complex to be a consequence of the complex being a neutral moiety, where
solute-solvent interactions are different than those with the anionic chromophore. It is important
to consider that the Li+ ion will exist in a hydrated form in aqueous solution and the fate of the
waters of hydration is not clear when complexation with resorufin occurs. Regardless of the fate
of the waters of hydration, the formation of a Li+-resorufin complex gives rise to a neutral
reorienting species in solution. The absorption and emission spectra of resorufin, with and
without Li+ present indicate that the primary interaction occurs between Li+ and the ground state
chromophore (Fig. 3.6).

65

Figure 3.6 Normalized (a) absorption and (b) emission spectra of resorufin in water
(black), 5 × 10-3 M LiClO4 (red), 10-2 M LiClO4 (blue), 5 × 10-2 M LiClO4 (green), and 10-1
M LiClO4 (orange).
66

It is clear that the details of the interactions between the comparatively small Li cation
and resorufin in water are fundamentally different than those observed for the same chromophore
with quaternary ammonium salts in NOP, but our findings do provide consistent evidence for the
propensity of the resorufin anion to form ionic complexes with a number of cations in different
solvents. This finding is consistent with earlier reports of Na+-resorufin interactions in the series
of butanols.33 Resorufin is a useful probe for the interrogation of ionic interactions in a variety
of different systems.
3.4 Conclusions
We have examined the effect of electrolyte concentration on the rotational diffusion
dynamics of the anionic chromophore resorufin in aqueous and NOP solutions. There is a
change in the chromophore rotor shape from oblate to prolate for reorientation in NOP upon the
addition of quaternary ammonium salts to the solution. These findings are consistent with
interactions between the anionic chromophore and the cationic quaternary ammonium species,
and from the dependence of the reorientation time constants on electrolyte concentration we can
estimate the equilibrium constant for the complex to be 2 ± 1 for TOAB and 11 ± 8 for TBAB.
The absorption and emission spectra in conjunction with semi-empirical calculations indicate
that the ionic complexation occurs along the chromophore long in-plane axis with the carbonyl
(=O) functionalities, and this finding is consistent with the change of effective rotor shape
reported for the reorientation experiments.
Qualitatively different behavior is observed for resorufin in water. The addition of
LiClO4 to aqueous resorufin solutions causes a decrease in the measured reorientation time
constant. These data are also consistent with the formation of an ionic complex, but in this
system where the cation is comparatively small, the dominant change in reorientation dynamics

67

is the result of the complex being a nominally neutral species. For both aqueous and NOP
systems, we find that resorufin complexes relatively efficiently with cationic species present in
solution, with the detailed manner in which this complexation phenomenon is manifested
depending on the specific cationic species used and its size relative to that of the chromophore.
The next two chapters describe the rotational diffusion dynamics in interfacial systems
containing water and NOP.

68

REFERENCES

69

REFERENCES

1. Ben-Amotz, D.; Scott, T. W. Journal of Chemical Physics 1987, 87, 3739.
2. Ben-Amotz, D.; Drake, J. M. Journal of Chemical Physics 1988, 89, 1019.
3. Blanchard, G. J.; Cihal, C. A. Journal of Physical Chemistry 1988, 92, 5950.
4. Blanchard, G. J. Journal of Physical Chemistry 1988, 92, 6303.
5. Wiemers, K.; Kauffman, J. F. Journal of Physical Chemistry A 1999, 104, 451.
6. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 9328.
7. Gayathri, B. R.; Mannekutla, J. R.; Inamdar, S. R. Journal of Fluorescence 2008, 18, 943.
8. Madden, P.; Kivelson, D. Journal of Physical Chemistry 1982, 86, 4244.
9. Kivelson, D.; Spears, K. G. Journal of Physical Chemistry 1985, 89, 1999.
10. Dutt, G. B.; Doraiswamy, S.; Periasamy, N.; Venkataraman, B. Journal of Chemical Physics
1990, 93, 8498.
11. Hartman, R. S.; Alavi, D. S.; Waldeck, D. H. Journal of Physical Chemistry 1991, 95, 7872.
12. Spears, K. G.; Steinmetz, K. M. Journal of Physical Chemistry 1985, 89, 3623.
13. Gudgin Templeton, E. F.; Quitevis, E. L.; Kenney-Wallace, G. A. Journal of Physical
Chemistry 1985, 89, 3238.
14. Blanchard, G. J. Journal of Chemical Physics 1987, 87, 6802.
15. Kenney-Wallace, G.; Paone, S.; Kalpouzos, C. Faraday Discussions 1988, 85, 185.
16. Blanchard, G. J. Journal of Physical Chemistry 1989, 93, 4315.
17. Blanchard, G. J. Journal of Physical Chemistry 1991, 95, 5293.
18. Alavi, D. S.; Waldeck, D. H. Journal of Physical Chemistry 1991, 95, 4848.
19. Zhou, P. W.; Song, P.; Liu, J. Y.; Shi, Y.; Han, K. L.; Het, G. Z. Journal of Physical
Chemistry A 2008, 112, 3646.
20. Hartman, R. S.; Konitsky, W. M.; Waldeck, D. H. Journal of the American Chemical Society
1993, 115, 9692.

70

21. Balabai, N.; Waldeck, D. H. Journal of Physical Chemistry B 1997, 101, 2339.
22. Balabai, N.; Kurnikova, M. G.; Coalson, R. D.; Waldeck, D. H. Journal of the American
Chemical Society 1998, 120, 7944.
23. Argaman, R.; Huppert, D. Journal of Physical Chemistry B 2000, 104, 1338.
24. Argaman, R.; Molotsky, T.; Huppert, D. Journal of Physical Chemistry A 2000, 104, 7934.
25. Dutt, G. B.; Ghanty, T. K. Journal of Physical Chemistry B 2003, 107, 3257.
26. Esquivelzeta-Rabell, M.; Peon, J.; Cuevas, G. Journal of Physical Chemistry B 2009, 113,
8599.
27. Debye, P. Polar Molecules Chemical Catalog Co.: New York 1929.
28. Hu, C. M.; Zwanzig, R. Journal of Chemical Physics 1974, 60, 4354.
29. Perrin, F. Journal of Physics and Radium 1934, 5, 497.
30. Hay, C. E.; Marken, F.; Blanchard, G. J. Journal of Physical Chemistry A 2010, 114, 4957.
31. Chuang, T. J.; Eisenthal, K. B. Journal of Chemical Physics 1972, 57, 5094.
32. Edward, J. T. Journal of Chemical Education 1970, 47, 261.
33. Rasimas, J. P.; Blanchard, G. J. Journal of Physical Chemistry 1996, 100, 11526.

71

Chapter 4
Detection and Characterization of a Liquid|Solid Interfacial Compositional
Gradient of Water Nano-Droplets in Wet N-Octyl-2-Pyrrolidone
4.1 Introduction
The study of liquid-solid interfaces has attracted significant interest because of the central
role such interfaces play in areas ranging from tribology through chemical sensing and
separations to electrode processes and reactions. The examination of the liquid phase at and near
such interfaces typically relies on the use of techniques that were designed for the study of bulk
liquids and, as such, there is limited information available on the organization of the interfacial
region. Electrochemical studies have, of course, provided exquisitely detailed information on the
organization of solutions in contact with electrode surfaces, and there is a well-established body
of theoretical understanding of the electric double layer and short range organization at the
electrode surface itself.1-3 It is well established that the spatial extent of surface charge effects is
on the order of tens to hundreds of nanometers, at most. Beyond this distance it is held that the
liquid phase behaves in the same manner as the corresponding bulk liquid.
Many processes of central importance to chemical and physical events that occur at the
liquid-solid interface can extend over µm length scales. Diffusion, precipitate formation, and the
existence of temperature gradients are three examples. It is thus of fundamental interest to
understand the limit(s) to the validity of the assumption that liquids located at the liquid-solid
interface behave as bulk liquids over distances on the order of microns. It is the purpose of this
work to investigate two chemical systems; one that is expected to behave as a bulk liquid and
one that has the potential to exhibit anomalous behavior. The two systems studied are ethylene
glycol (EG) and N-octyl-2-pyrrolidone (NOP), each at a borosilicate glass interface. EG is
miscible with water in all proportions and is expected to exhibit properties of a bulk liquid in the
72

presence or absence of water, whereas amphiphilic NOP is not completely miscible with water.
NOP readily phase separates from water at room temperature and has a lower consolute
temperature below 0 °C.4,5 The solubility of water in NOP is not reported explicitly, although
phase diagrams indicate a single phase may exist at water concentrations less than 40% (w/w).5,6
The low miscibility of water in NOP, as well as the possibility of the formation of secondary
structures in the presence of water provides an opportunity for anomalous behavior.
While there are a variety of ways to study the liquid-solid interface, examining such an
interface with a depth resolution of less than one µm and an operating depth range on the order
of 100 µm, and providing spatially localized information on viscoelastic properties can be a
significant challenge. We use a confocal scanning microscope equipped with time-resolved
detection gear to resolve the fluorescence lifetime and reorientation dynamics of a chromophore
in the liquid phase with ca. 50 ps time resolution. Using this instrument we have demonstrated
previously the ability to obtain fluorescence lifetime and anisotropy decay images for planar
supported lipid bilayer structures.7,8 This is the first report utilizing this instrument primarily for
its ability to achieve depth resolution, as opposed to lateral resolution. We have found that the
lifetime and anisotropy decay dynamics of a rhodamine chromophore behave in the expected
manner, more or less as a bulk liquid, when measured as a function of distance from the
solvent|glass interface for the solvent EG. For the interface between NOP containing water (wet
NOP) and glass, we resolve a depth dependent gradient in fluorescence anisotropy decay
dynamics that persists for tens of microns into the liquid phase. Using the information from
these two interfacial systems, we are able to identify the presence of a compositional
inhomogeneity in the wet NOP system in the form of a nano-emulsion. We consider the possible
origins of this interfacial gradient.

73

The contribution of NOP as a minor component in industrial products is well-established,
however, there is limited information on NOP as a major component contributing to the uptake
and solvation of polar molecules. Longer chain N-alkyl-2-pyrrolidones including NOP exhibit
surface activity in aqueous solutions and are effective as co-surfactants forming micelles with
anionic surfactants.4,9 The work presented in chapter 2 on the decay dynamics of chromophores
solvated in neat NOP illustrated the role of solvent-solvent interactions in mediating solutesolvent interactions and the impact on rotational diffusion.10 The solvation of chromophores by
NOP was markedly different than that seen for water. In the binary mixture of NOP and water
studied here, we are interested in the composition of wet NOP and the interactions that influence
the solvation and local environment experienced by the fluorescent probe molecule.
4.2 Experimental Methods
The chromophore lissamine rhodamine B sulfonyl chloride (LRSC) (Fig. 4.1a, ≥ 99%
Acros Organic) was used as received. N-octyl-2-pyrrolidone (NOP) (Fig. 4.1b, 98% Aldrich)
and ethylene glycol (EG) (≥ 99%, Macron Fine Chemicals) were used as received. Water used
in these studies was purified in-house with a Milli-Q filtration system (Millipore). For steadystate absorbance and emission measurements, the chromophore concentration was 10-4 M in
water and dry NOP. The concentration of LRSC in EG was 10-5 M. Upon exposure to water, the
chromophore is likely hydrolyzed to the sulfonic acid. We refer to the chromophore as LRSC
throughout because we have not verified the hydrolysis reaction for each system. The sample
holder for optical measurements on the microscope was an Attofluor® cell chamber (Invitrogen)
fitted with a round borosilicate glass coverslip (25 mm diameter, No. 1.5 Thomas Scientific). In
order to contain the liquid samples in a smaller area, a glass microscope slide (1 mm thick, 1.7
cm diameter, Globe Scientific) with a bored hole (inner diameter of 5.1 mm) was placed in the

74

sample holder. Solutions of 10-4 M LRSC in dry NOP were exposed to water for 3 hours and
then approximately 150 μL was deposited in the center of the bored microscope slide onto a
clean glass surface. Excess liquid filled the space between the glass surface and the microscope
slide. The microscope slide and coverslips were cleaned prior to use by immersing in piranha
solution (a 3:1 ratio of concentrated sulfuric acid and 30% hydrogen peroxide – caution strong
oxidizer!) for 30 minutes. An alternative method of cleaning the borosilicate glass slides was
rinsing with ethanol prior to exposure to UV-generated ozone for 15 minutes.

a
O

+
N

N

–

SO3

b

SO2Cl
O

N

Figure 4.1 Structures of (a) lissamine rhodamine B sulfonyl chloride (LRSC) and (b) Noctyl-2-pyrrolidone (NOP).

75

Steady-state absorbance measurements were made with an ATI Unicam double-beam
UV-visible absorbance spectrophotometer over the range of 400 nm to 700 nm, at a spectral
resolution of 1 nm for all measurements. Steady-state spontaneous emission spectra were
recorded using a Jobin-Yvon Fluorolog III spectrometer over the range of 550 nm to 700 nm.
The excitation wavelength was set to 540 nm for LRSC in water and NOP emission spectral
acquisition.
The fluorescence lifetime and anisotropy decays were collected with the time-resolved
fluorescence imaging instrument (Fig. 4.2) described previously.8 A brief description of the
instrumental set-up is provided here. The light source for the system is a CW mode-locked
Nd:YVO4 laser that produces 13 ps pulses at 1064 nm with an 80 MHz repetition rate (Spectra
Physics Vanguard). The Nd:YVO4 laser second or third harmonic output excites a cavitydumped dye laser (Coherent 702-2), with output characterized by 5 ps pulses at wavelengths
between 430 nm and 850 nm. The repetition rate of the laser is adjustable between 80 kHz and
80 MHz using the cavity dumping electronics. The dye used to excite the samples was
pyrromethene 567 (Exciton), providing excitation wavelengths between 550 nm and 595 nm.
The average power at the sample is 1 mW or less.
A vertically polarized pulse of light is input to an inverted optical microscope (Nikon
Eclipse Ti-U) through a confocal scanner (Becker & Hickl DCS-120). The excitation pulses
pass through the microscope objective and the focused beam can be localized at one position or
rastered across the surface of the sample using the galvo-drive unit (Becker & Hickl GDA-120)
in the confocal scanner. The emission collected by the microscope objective is split into parallel
and perpendicular polarized signal components in the confocal scanner, passed through long-pass
and bandpass wavelength selection filters, and detected using two avalanche photodiode

76

detectors (ID-Quantique ID100). Time-correlated single photon counting (TCSPC) detection
electronics (Becker & Hickl SPC-152) are used for signal processing and collection. SPCM
software (Becker & Hickl) for a Windows-based PC is used to control the data acquisition
parameters.

Figure 4.2 Time-resolved fluorescence lifetime and anisotropy imaging instrument with
time-correlated single photon counting (TCSPC) detection.

77

An oil immersion microscope objective (Nikon, 100X magnification, NA 1.45, working
distance 0.13 mm, depth of field ca. 0.45 µm) was used for all data acquisition, and the
experimental setup is schematized in Fig. 4.3. Depth profiles were generated by collecting
individual data sets in triplicate with the beam parked at a fixed position. The positions along the
x-axis were fixed using the SPCM galvanometer control window and the depth along the z-axis
was controlled mechanically. The time-resolved fluorescence data were analyzed using
MicroCal Origin Pro v. 9.0 to obtain fluorescence lifetime and anisotropy decay time constants.
Depth into film, z

~500 µm

NOP
glass

Side view

z

100x objective

y

x

Figure 4.3 Schematic of sample and optical excitation/collection configuration for
NOP|glass and EG|glass interfaces. (NOP|glass interface shown.)
Particle size determinations were performed by dynamic light scattering (DLS) using a
Malvern Zetasizer Nano ZS with a 632.8 nm light source for dry NOP and NOP exposed to
water. All measurements were made at 20 ± 1 °C. The wet NOP was prepared by adding dry
NOP to a layer of water in a glass vial, where NOP was 6.67% (v/v). The NOP layer was
removed and analyzed by DLS.

78

4.3 Results and Discussion
An important area of understanding in interfacial science is the role that the solid
(interface) may play in mediating organization of the liquid phase(s) in contact with it. This is a
well explored and understood area in the context of solution|electrode interfaces, for example,
where control over the potential of the electrode surface allows for the evaluation of short range
electrostatic or Coulombic forces on solution phase local organization. It is understood that the
spatial extent of the electric double layer is less than one micron, typically much less.1,2,11 We
report in this paper on the existence of a gradient in the fluorescence anisotropy decay dynamics
of LRSC in wet NOP present at a glass surface, as a function of distance from the surface. This
gradient is anomalous because of its spatial extent and we are concerned with understanding the
fundamental reason(s) for its existence. Achieving this understanding will allow for such
gradients to be predicted and, possibly, the extent and nature of the gradient to be controlled for
pre-determined purposes.
The instrumentation we have used to acquire information on interfacial organization and
dynamics is designed to acquire images in the plane of the sample (x,y) (see Fig. 4.3).7,8 For the
work we report here, the sample is uniform in the x,y plane and we have examined the depth (z)
dependence of chromophore lifetime and dynamics. Because this imaging instrument is
confocal, the z-resolution (depth of focus) is sub-µm, and the data have been acquired at 10 µm
z-intervals. Each data point presented is the average of at least five individual measurements,
with the uncertainties being reported as the standard deviation (±1σ) of the average of the
individual acquisitions.
The fluorescence lifetimes and anisotropy decays plotted as a function of distance from
the NOP|glass interface are shown in Fig. 4.4. There are several important features contained in

79

these data. The first is that at the interface, the chromophore is adsorbed on or interacting with
the glass support and its fluorescence lifetime and anisotropy decay time constants are different
than for the liquid NOP layer. This is not surprising given the polar nature of the chromophore
and the interface and the comparatively less polar NOP solvent. The second feature of note is
that, to within the experimental uncertainty, the fluorescence lifetime is constant as a function of
depth into the NOP layer. We believe that this is likewise not a surprising result. The third
feature contained in these data is the depth dependence of the anisotropy decay time constant.
The existence of a depth dependence is reproducible and unexpected. The aim of this work is to
describe the physical and chemical basis for this finding.

80

Figure 4.4 Dependence of LRSC in NOP fluorescence lifetime (●) and anisotropy decay
time constant (○) on distance from the glass interface. The line drawn through the
fluorescence lifetime data is intended as a guide to the eye. The glass support was cleaned
with piranha solution and rinsed with water prior to use. Refer to Fig. 4.3 for the
experimental configuration and the x, y, z coordinate designation.

81

Before turning to a detailed interpretation of these findings, it is instructive to consider
several potential causes. We presume that the depth dependence of the anisotropy decay time
constant seen in Fig. 4.4 is the result of either an electric field gradient or a compositional
gradient. One of the first issues that must be addressed is whether or not this effect is a result of
contamination. The cleaning of the glass surface could, in principle, leave residue that could
diffuse slowly into the NOP solvent system. We cleaned the glass support used for the data
reported in Fig. 4.4 using piranha solution, followed by rinsing with deionized water (vide infra).
If there were any residue remaining from the washing process it could potentially diffuse from
the surface and create a compositional gradient, the form of which would change over time. For
this reason, we performed the same study using a glass support that had been rinsed with ethanol
and then cleaned using a UV/ozone cleaner. Any residual contaminant profile for this interface
would be different than for the interface cleaned with piranha solution and rinsed with water.
We show the results obtained for the ethanol/ozone prepared surface in Fig. 4.5. These results
show an identical trend to those shown in Fig. 4.4, and from this comparison we conclude that
contamination resulting from preparation of the support surface does not contribute to the depth
dependence of the LRSC anisotropy decay we observe.

82

Figure 4.5 Dependence of LRSC in NOP fluorescence lifetime (●) and anisotropy decay
time constant (○) on distance from the glass interface. The line drawn through the
fluorescence lifetime data is intended as a guide to the eye. The glass support was cleaned
using UV-generated ozone and rinsed with ethanol prior to use. Refer to Fig. 4.3 for the
experimental configuration and the x, y, z coordinate designation.

83

Given the inherently charged nature of the glass interface, it is important to consider any
potential role of Coulombic/electrostatic forces in producing the results shown in Figs. 4.4 and
4.5. The spatial extent of Coulombic effects is related to the propensity of the medium to screen
charge, and for polar systems containing electrolytes the spatial extent of the Coulombic effects
that give rise to the electric double layer and the diffuse double layer regions are typically under
100 nm. For nonpolar systems that contain very few charged species, the spatial extent of
surface charge effects can be longer, but in any event such effects operate on the sub-micron
length scale, and the effect we observe persists for tens of microns.
For these reasons, and the extent of the gradient in the anisotropy decay time we observe
experimentally (Figs. 4.4 and 4.5), we assert that neither contaminant-related nor Coulombic
effects play a significant role in the phenomenon we observe. By process of elimination, the
explanation for the anisotropy decay gradient we observe must be related to a compositional
gradient in the liquid (NOP) phase. We have exposed NOP to deionized water with the
expectation that the solution will reach an equilibrium water concentration in NOP after a few
hours. Because the solvent system we use is a binary system, there is the opportunity for a
compositional gradient to exist. There is potential for the composition of the liquid near the
NOP|glass interface to be affected by adsorption of NOP as has been observed for other
nonionic12-14 and ionic15-17 surfactants. As a result of adsorption, micelles, hemi-micelles,
monolayer or bilayer structures may form at the solid surface depending on the concentration and
interaction with surface charge.18 However, the spatial extent of the secondary structures would
only be on the order of nanometers and would not account for micron scale gradient we observe.
Additionally, NOP and longer chain N-alkyl-2-pyrrolidones are known to form micelles only in

84

presence of co-surfactants.4,9,19 We return to discussion of the observed compositional gradient
after evaluating the generality of the phenomenon.
If the anisotropy decay gradient we observe is indeed related to a compositional gradient
in the wet NOP phase, then the replacement of NOP with a different solvent system, particularly
one which is miscible with water in all proportions, should alter the gradient. We show in Fig.
4.6 the fluorescence lifetime and anisotropy decay data for LRSC in EG as a function of distance
from the EG|glass interface. The fluorescence lifetime data for LRSC in EG resemble those for
LRSC in wet NOP, exhibiting depth-independent lifetimes for the chromophore away from the
surface, and a different lifetime for the chromophore interacting with the glass interface (depth =
0). The anisotropy decay data for LRSC in EG exhibit a depth independence to within the
experimental uncertainty, in sharp contrast to the data for NOP. The behavior of LRSC at the
EG|glass interface is expected and this finding points to the NOP system exhibiting a
compositional gradient. We turn next to the possible reasons for the observed compositional
gradient.

85

Figure 4.6 Dependence of LRSC in EG fluorescence lifetime (●) and anisotropy decay time
constant (○) on distance from the glass interface. The line drawn through the fluorescence
lifetime data (solid) and the anisotropy decay data (dashed) are intended as guides to the
eye. The glass support was cleaned using piranha solution and rinsed with water prior to
use. Refer to Fig. 4.3 for the experimental configuration and the x, y, z coordinate
designation.

86

We consider the following discussion of these anisotropy decay data, and the
physical/chemical basis for them in the context of known models of rotational diffusion. The
fluorescence lifetime and anisotropy decay data are derived from polarized fluorescence
transients. The sample is excited with a polarized picosecond pulse of light and the emission
from the sample is collected for polarizations parallel (I||(t)) and perpendicular (I⊥(t)) to that of
the excitation pulse. The fluorescence lifetime (Ifl(t)) and anisotropy decay functions (R(t)) are
formulated according to Eqs. 4.1 and 4.2.
I fl=
(t ) I|| (t ) + 2 I ⊥ (t )

R (t ) =

I|| (t ) − I ⊥ (t )
I fl (t )

(4.1)

(4.2)

For all cases, we observe a single exponential decay of Ifl(t), indicating the absence of anomalous
excited state relaxation phenomena. The chemical information of interest is contained in the
functional form and time constant(s) of the decay of R(t). In this work the chromophore is not
tethered to either a solvent molecule or to the glass interface and its motion can be considered in
the context of a free rotor. Chuang and Eisenthal20 have developed the theoretical framework for
relating the Cartesian components of the rotational diffusion constant, D, and the angle between
the excited and emitting transition dipole moments, to the pre-factors and time constants of the
exponential decay(s) contained in the experimental R(t) data. The chromophore, LRSC, is
characterized by a nominally planar π system and the S1 ← S0 transition is polarized along the
chromophore ring system long axis (x).21,22 For these conditions, a single exponential decay of
R(t) is consistent with reorientation as a prolate rotor, and a two-component exponential decay is
consistent with reorientation as an oblate rotor. For all of the measurements we report here R(t)
decays as a single exponential, and the anisotropy decay time constant is given by τOR = (6Dz)-1.

87

Even with the limited information provided by a single exponential decay component, where
only one Cartesian component of the rotational diffusion constant is sensed, we can speak to the
possible reasons for the gradient we observe.
The anisotropy decay time constant is related to the size of the reorienting moiety, the
viscosity of the surrounding medium and the temperature of the system through the modified
Debye-Stokes-Einstein (DSE) model,23

τ OR
=

1
ηVf
=
6 D k BTS

(4.3)

where η is the viscosity of the bulk solvent, V is the hydrodynamic volume of the reorienting
entity, f is a term describing the frictional interaction between the reorienting moiety and the
surrounding medium (0 < f ≤ 1), kBT is the thermal energy term and S is a shape factor to account
for the non-spherical shape of the chromophore.24 We calculate the hydrodynamic volume of
LRSC to be 560 Å3 using the method of van der Waals increments,25 and the shape factor S has
been determined for rhodamine chromophores to be S = 0.9.21 Using this information and the
bulk solvent viscosity of η = 5.6 cP for NOP,26,27 we obtain an estimate of τOR = 862 ps from Eq.
4.3. This estimate is significantly lower than the anisotropy decay time constants we have
observed experimentally for the NOP|glass interface. The data in Fig. 4.4 show τOR varying from
ca. 2100 ps near the glass surface to ca. 1800 ps at 100 µm from the interface.
It is fair, of course, to question the accuracy or validity of the DSE model for this
application. We can gauge the accuracy of the DSE model by comparing its predictions to
experimental data for a system that is not seen to behave in an anomalous manner. For LRSC in
EG we observe τOR for distances of 10 – 100 µm from the interface of 2320 ± 38 ps. Taking the
viscosity of EG to be 16.1 cP at 25 °C,28 with f = 1 (stick limit) and S = 0.9,21,22 we calculate τOR

88

= 2435 ps. The slight difference between experiment and theory is compelling evidence for the
accuracy and applicability of the modified DSE model for these data. While the LRSC/EG
interactions and the LRSC/NOP interactions may differ, it is the z-dependent trend in the data
that is of primary importance here, and we consider how that trend can be explained in the
context of Eq. 4.3.
In considering the discrepancy between the simple estimate from the DSE model (Eq.
4.3) and the experimental results, there are a limited number of ways to account for the data.
Specifically, the temperature remains constant, we presume the chromophore does not change
shape as a function of proximity to the glass surface, and the viscosity of the NOP solvent does
not vary with proximity to the surface. We are thus left, within the framework of the DSE
model, to conclude the anisotropy decay gradient is reflective of a gradient in the hydrodynamic
volume of the reorienting entity. We recognize that the gradient in τOR corresponds to changes in

ηVf and that these terms cannot be separated cleanly. Despite this limitation, we assert on
physical grounds that the most plausible change is occurring with the effective hydrodynamic
volume of the reorienting species. Deviations from hydrodynamic theory as described by the
modified DSE equation have been reported previously in binary systems, where specific solutesolvent interactions and changes the hydrodynamic volume have been cited as reasons for
nonconformity to model behavior.29-32 Stevenson et al. reported a compositional dependence in
the reorientation times of 4-benzylamino-7-nitrobenzylfurazan (BBD) in 1-propanol/water binary
systems, where the hydrodynamic volume of the reorienting entity was found to change based on
the inclusion of interacting solvent molecule(s).32 Specific solute-solvent interactions, such as
hydrogen bonding, can account for reorientation times greater than model predictions by strong
association of solute and solvent molecule(s). Hydrogen bonding interactions provide for this

89

adjustment in hydrodynamic volume because the duration of hydrogen bonds can be on the same
order as the rotational diffusion time.
In the case of the wet NOP system, the change in hydrodynamic volume of the
reorienting entity can be accounted for in the context of the chromophore residing in a primarily
aqueous environment that is surrounded by NOP molecules, and the reorienting entity is
(LRSC⋅nH2O), with n being the depth-dependent quantity. In this model, the reorienting entity
would be a nominally spherical chromophore-water complex (S = 1). To account for the
reorientation times observed experimentally, we add n H2O molecules (21 Å3 each), as shown in
Table 4.1. The results of this model show that there is a monotonic trend in the number of water
molecules surrounding the chromophore, decreasing with increasing distance from the interface.
The number of water molecules in this model ranges from ca. 36 to ca. 48, a number that we
believe to be consistent with absorbance and dynamic light scattering (DLS) data (vide infra).
We show in Fig. 4.7a the calculated number of water molecules surrounding the LRSC
chromophore, n, as a function of distance from the interface, and in Fig. 4.7b, we show the
calculated diameter of the chromophore-water entity as a function of distance from the interface.
We note that while there may appear to be a slight offset in the data for the interface cleaned
using piranha solution and water compared to the interface cleaned using UV-generated ozone
and ethanol, any difference between the two sets of results can be accounted for in the context of
differences of ± 1 water molecule in the system. We take this to be acceptable agreement
between the two data sets.

90

Table 4.1 Estimated number of water molecules surrounding LRSC (model prediction for
number of waters, n).a
τOR calc (ps)

Vtotal (Å3)

Vwater (Å3)

n

d (nm)

1824

1316

756

36

1.36

1853

1337

777

37

1.37

1882

1358

798

38

1.37

1911

1379

819

39

1.38

1941

1400

840

40

1.39

1970

1421

861

41

1.39

1999

1442

882

42

1.40

2028

1463

903

43

1.41

2057

1484

924

44

1.42

2086

1505

945

45

1.42

2115

1526

966

46

1.43

2144

1547

987

47

1.43

2173

1568

1008

48

1.44

a

Number of water molecules, n, between 36 and 48 is used to determine the hydrodynamic
volumes of water (Vwater) and a total volume (VLRSC + nVwater), which is used to calculate model
reorientation times, (τOR calc). Diameters, d, are calculated assuming a spherical hydrodynamic
volume.

91

Figure 4.7 (a) Calculated number of water molecules in the immediate proximity of LRSC
as a function of distance from the NOP|glass interface, for the interface cleaned with
piranha solution and water (●) and cleaned with ethanol and UV-generated ozone (○). (b)
Calculated diameters of the reorienting entities containing the chromophore and the
number of water molecules indicated in (a), as a function of distance from the NOP|glass
interface.

92

It is useful to note that the number of water molecules predicted by this model is
dependent on the assumptions made, such as the local viscosity of the NOP. Local viscosity, as
sensed by a rotating chromophore, is not necessarily the same as the bulk solvent viscosity,
which is mediated by solvent-solvent intermolecular interactions. Also, the details of the
solvent-solute coupling for the chromophore-water microenvironment are not addressed directly
in this model. Despite these limitations, all data reported are interpreted using the same set of
assumptions. Thus, the exact number of water molecules predicted by the model is subject to
some uncertainty, the trend seen in these data is fully valid and is consistent with the
chromophore sampling a nano-emulsion environment characterized by a compositional gradient
normal to the plane of the glass support.
An important consideration is why the chromophore within the water droplet does not
simply reorient as though it resided in bulk water with the NOP|water interface being decoupled
from the LRSC/water interactions. To evaluate this issue is it instructive to consider the relative
sizes of the water nano-droplets and the bare chromophore. As noted above, the calculated
hydrodynamic volume for LRSC is 560 Å3, and if this volume is taken to be spherical it
corresponds to a diameter of 1.02 nm. The water nano-droplets are measured by DLS to have a
diameter of 1.26 ± 0.25 nm (vide infra), indicating a difference in diameter of ~0.24 nm. The
implication of this comparison is that the sizes of the bare chromophore and the water nanodroplets are sufficiently close that the decoupling of the chromophore/water and water/NOP
interactions is not possible. In other words, the relative sizes of the chromophore and water
nano-droplets are similar enough that the chromophore/water moiety needs to be treated as a
single reorienting species.

93

There are two testable predictions of this model. The first is that the chromophore finds itself in
a local environment influenced strongly by the presence of water. The absorption spectrum of
LRSC is sensitive to the presence of water. We show in Fig. 4.8 the absorption spectra of LRSC
in water, dry NOP, NOP saturated in water, and LRSC in NOP after extraction from water. The
latter two spectra are overlapped. As can be seen from these data, the LRSC chromophore finds
itself in a primarily aqueous environment in the water-saturated NOP solution. These data are
consistent with the explanation put forth for the anisotropy decay gradient, which asserts water
molecules surround LRSC forming reorienting entities with hydrodynamic volumes greater than
the chromophore alone.

Figure 4.8 Normalized absorption spectra of LRSC in water (black), LRSC in dry NOP
(red), LRSC in NOP after equilibration with water (blue), and LRSC in NOP after
extraction from water (green).
94

The second prediction of the model is that, if these data are the result of a water nanoemulsion, there should be detectable droplets present in the solution. We have used DLS to
evaluate whether or not this is the case. DLS is a spectroscopic technique used to determine the
particle size distribution of a colloidal suspension. A laser light source illuminates the sample
and light scattered from the dispersed particles is detected. The intensity of scattered light is
recorded as a time-correlation function that is related to movement of the particles via Brownian
motion.33 Particle size distributions are then calculated from the fluctuation in motion of
different size particles. We show these results in Fig. 4.9, indicating the presence of species with
a diameter of 1.26 ± 0.25 nm in the water-containing NOP solution. An absence of DLS signal
is seen in this same size region for dry NOP (Fig. 4.9). For all DLS measurements, no
chromophore was present. Comparing these results to those shown in Table 4.1 and Fig. 4.7b,
we assert that this represents excellent agreement between experiment and model. We
understand the correspondence between the experimental data and predictions of the model as
follows. While the results of the model calculation are in agreement with the experimental data
to within the experimental uncertainty, the values predicted by the model appear to be
consistently higher than those determined by DLS. The DLS measurement samples the bulk
solvent system whereas the reorientation data are for the solution near the interface, where the
anisotropy decay gradient is observed. Even at a depth of 100 µm, which is our experimental
limit with the time-resolved measurement, the trend in size of the chromophore-water entity
appears to be decreasing. Thus a bulk (DLS) measurement of the size of the water domains
would provide a value that is expected to be smaller than what we observe at and near the glass
interface based on the existence of the anisotropy decay gradient.

95

Figure 4.9 Dynamic light scattering (DLS) data showing the size distribution of water
nano-droplets present in wet NOP after exposure to water (●) and dry NOP (○), showing
the absence of species in the same size range as the water nano-droplets. For all DLS
measurements no chromophore was present.

96

The experimental anisotropy decay data for LRSC in wet NOP are consistent with a
gradient in the average size of chromophore-water cluster increasing with proximity to the glass
interface and this finding is not expected for a simple solution phase system, such as EG. It is
important to consider the implications of this finding. The first implication is that NOP and
water, at least up to saturation concentration, exist as a colloidal system. When viewed in this
context, it is reasonable to expect there to be a distribution of colloidal “nano-droplets” (i.e.
water and water + chromophore) suspended in NOP. In fact, there is precedent for the existence
of such nanoscale heterogeneity being induced in liquid crystalline systems by the presence of a
small amount of water.34 One possible explanation for the existence of a gradient in the size of
the colloidal suspension of water nano-droplets is that there is a difference in the surface energy
for the NOP|glass and NOP|water interfaces, and the size gradient in the water nano-droplets
represents a gradient in the Laplace pressure of the nano-droplets to compensate for the
difference in surface energies of the two interfaces. At the point where the NOP, water nanodroplet and glass are in contact, there is a single Laplace pressure that characterizes the resulting
structure, and that Laplace pressure is different than for the water nano-droplet in the NOP
phase. Diffusion mediated transport of water nano-droplets occurs and water molecules are
“shed” from the nano-droplet to achieve the equilibrium Laplace pressure. This explanation
requires that the shedding of water molecules occurs on a time scale similar to that of nanodroplet diffusion, giving rise to a gradient in nano-droplet size over the length scale we observe.
A prediction of this explanation is that different interfaces with wet NOP would give rise to
different nano-droplet gradients, and that for the NOP|water planar interface there should not be
a water nano-droplet size gradient. It is important to keep in mind that the range of the

97

compositional gradient is on the order of ± ten water molecules in a cluster, and the use of
rotational diffusion allows the resolution of this subtle compositional gradient.
Another issue that is of some significance is that for the NOP|glass system, there is a
limited amount of water present in the system and, for thermodynamic reasons, the concentration
should be uniform throughout the solution. For this reason, the dielectric constant of the system
should be nominally constant, consistent with a depth-independent fluorescence lifetime of the
chromophore. This is not expected to be the case under the condition of an excess of water being
available to the NOP phase, and we consider this situation in the next chapter.
The issue of hindered Brownian diffusion of nanoscale droplets near a solid surface may
play a role in these data as well. Near a solid surface, Brownian diffusion in the liquid phase
becomes anisotropic, with motion parallel to the surface remaining more similar to that in the
bulk than motion normal to the surface.35-38 Interfacially-mediated hindered Brownian diffusion
of a nanoparticle near a solid surface can be understood in terms of the surrounding liquid being
the source for momentum due to fluctuations in density. As the solid surface is approached, such
fluctuations diminish to the level of phonons in the solid, leading to slower diffusion near the
surface. While the length scale of this effect (microns) is too short to account for the anisotropy
decay gradient we observe (Fig. 4.4), it can play a role in mediating the rate at which nanodroplets diffuse from the surface. Also, to the extent to which translational and rotational
diffusional motion are coupled,39 such interfacial effects could be responsible for the functional
form of the anisotropy decay gradient data we see in Fig. 4.4 near the wet NOP|glass interface.
This effect is not expected to play a role at the liquid-liquid interface which we consider in the
next chapter.

98

4.4 Conclusions
We have reported on the existence of a gradient in the anisotropy decay time for the
chromophore LRSC in NOP that contains water. The anisotropy decay gradient scales with the
distance from a glass interface (z), and this gradient is not seen for LRSC in EG in the proximity
of a glass interface. We understand these data in the context of a compositional gradient in the
wet NOP solvent system, where the species we detect are LRSC solvated by water droplets in the
NOP phase. The chromophore-containing water droplets are found to vary in size
(hydrodynamic volume) as a function of distance from the glass surface. Anisotropy decay data
in conjunction with the modified DSE model point to a progression in the number of water
molecules associated with a chromophore molecule with increasing proximity to the glass
surface. The size of the water nano-droplets is estimated from these data to be on the order of
1.3 – 1.4 nm in diameter, and dynamic light scattering of the colloidal suspension reveals the
presence of nano-droplets that are ca. 1.26 nm diameter, on average. We view this as excellent
agreement because the DLS measurements sample all colloidal droplets in the suspension,
whereas the anisotropy decay data sample only those droplets in close proximity to the glass
surface. The existence of the gradient is consistent with a difference in the surface energies of
the wet NOP nano-droplet interface and the NOP|glass interface. No analogous organization is
found in a solvent system (ethylene glycol) that does not exhibit phase segregation issues. This
work points to one possible means of controlling colloidal suspension size in applications where
such dimensional issues are important. In the next chapter, we examine the fluorescence
lifetimes and anisotropy decays for a NOP|water|glass system.

99

REFERENCES

100

REFERENCES

1. Pletcher, D. A First Course in Electrode Processes, 2nd Edition; RSC Publishing: Cambridge,
UK, 2009, and references therein.
2. Bockris, J. O. M.; Devanathan, M. A. V.; Muller, K. Journal of the Royal Society, Series A,
Mathematical and Physical Sciences 1963, 274, 55.
3. Helmholtz, H. Annalen der Physik 1853, 165, 211.
4. Rosen, M. J.; Zhen, H. Z.; Gu, B.; Murphy, D. S. Langmuir 1988, 4, 1273.
5. Lou, A. J.; Pethica, B. A.; Somasundaran, P.; Yu, X. Journal of Colloid and Interface Science
2002, 256, 190.
6. Adamy, S. T. Langmuir 1995, 11, 3269.
7. Setiawan, I.; Blanchard, G. J. Journal of Physical Chemistry B 2014, 118, 3085.
8. Setiawan, I.; Blanchard, G. J. Journal of Physical Chemistry B 2014, 118, 537.
9. Rosen, M. J.; Gu, B.; Murphy, D. S.; Zhu, Z. H. Journal of Colloid and Interface Science
1989, 129, 468.
10. Hay, C. E.; Marken, F.; Blanchard, G. J. Journal of Physical Chemistry A 2010, 114, 4957.
11. Conway, B. E.; Bockris, J. O. M.; Ammar, I. A. Faraday Transactions 1951, 47, 756.
12. Levitz, P.; Elmiri, A.; Keravis, D.; Van Damme, H. Journal of Colloid and Interface Science
1984, 99, 484.
13. Levitz, P.; Van Damme, H.; Keravis, D. Journal of Physical Chemistry 1984, 88, 2228.
14. Levitz, P. E. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2002, 205,
31.
15. Cases, J. M. Bulletin De Mineralogie 1979, 102, 684.
16. Strom, C.; Hansson, P.; Jonsson, B.; Soderman, O. Langmuir 2000, 16, 2469.
17. Strom, C.; Jonsson, B.; Soderman, O.; Hansson, P. Colloids and Surfaces APhysicochemical and Engineering Aspects 1999, 159, 109.
18. Paria, S.; Khilar, K. C. Advances in Colloid and Interface Science 2004, 110, 75.

101

19. Zhu, Z. H.; Yang, D.; Rosen, M. J. Journal of the American Oil Chemists Society 1989, 66,
998.
20. Chuang, T. J.; Eisenthal, K. B. Journal of Chemical Physics 1972, 57, 5094.
21. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2001, 105, 9328.
22. Dela Cruz, J. L.; Blanchard, G. J. Journal of Physical Chemistry A 2002, 106, 10718.
23. Debye, P. Polar Molecules Chemical Catalog Co.: New York 1929.
24. Perrin, F. Journal of Physics and Radium 1934 5, 497.
25. Edward, J. T. Journal of Chemical Education 1970, 47, 261.
26. MacDonald, S. M.; Watkins, J. D.; Bull, S. D.; Davies, I. R.; Gu, Y.; Yunus, K.; Fisher, A.
C.; Page, P. C. B.; Chan, Y.; Elliott, C.; Marken, F. Journal of Physical Organic Chemistry
2009, 22, 52.
27. MacDonald, S. M.; Watkins, J. D.; Gu, Y.; Yunus, K.; Fisher, A. C.; Shul, G.; Opallo, M.;
Marken, F. Electrochemistry Communications 2007, 9, 2105.
28. CRC Handbook of Chemistry and Physics, 71st Edition; Lide, D. R., Ed.; CRC Press: Boca
Raton, FL, 1990-1991.
29. Beddard, G. S.; Doust, T.; Hudales, J. Nature 1981, 294, 145.
30. Dutt, G. B.; Doraiswamy, S. Journal of Chemical Physics 1992, 96, 2475.
31. Bhattacharyya, S.; Bagchi, B. Journal of Chemical Physics 2001, 115, 9065.
32. Stevenson, S. A.; Blanchard, G. J. Journal of Physical Chemistry A 2006, 110, 3426.
33. Berne, B. J.; Pecora, R. Dynamic Light Scattering With Applications to Chemistry, Biology,
and Physics; General Publishing Company, Ltd: Toronto, 2000.
34. Schroder, U.; Wadahawan, J. D.; Compton, R. G.; Marken, F.; Suarez, P. A. Z.; Consorti, C.
S.; de Souza, R. F.; Dupont, J. New Journal of Chemistry 2000, 24, 1009.
35. Kihm, K. D.; Banerjee, A.; Choi, C. K.; Takagi, T. Experiments in Fluids 2004, 37, 811.
36. Banerjee, A.; Kihm, K. D. Physical Review E 2005, 72, 042101.
37. Feitosa, M. I. M.; Mesquita, O. N. Physical Review A 1991, 44, 6677.
38. Hosoda, M.; Sakai, K.; Takagi, K. Physical Review E 1998, 58, 6275.
39. Kurban, M. R. Journal of Chemical Physics 2011, 134, 034503.

102

Chapter 5
Characterization of Liquid|Liquid|Solid Multi-Interfacial Compositional
Gradients
5.1 Introduction
The study of liquid interfaces has garnered interest in industrial, biological, and
environmental research because of their broad physical and chemical significance.1,2 Interactions
between dissimilar molecules and molecular organization at liquid-liquid and liquid-solid
interfaces are important for many processes, including chemical separations, extractions,
interfacially-mediated polymerization reactions, and drug delivery, for example. Advances in
spectroscopic measurement techniques have allowed for greater understanding of the solvation
phenomena and intermolecular interactions that occur at liquid interfaces despite the difficulty in
selectively probing the liquid-liquid interfacial region (ca. 1 nm thick).3-12 Second order
nonlinear optical techniques can provide the requisite interface selectivity, but with these
techniques there is little ability to investigate organization and dynamics away from the interface
by more than a few nanometers. Understanding and controlling interfacial processes, such as
electrochemical phenomena, that occur within a few molecular diameters of liquid-liquid and
liquid-solid interfaces requires in-depth knowledge of interfacial composition, organization, and
dynamics, which has spurred both experimental studies and theoretical simulations.13-17 The
three-phase boundary, where a liquid-liquid interface comes into contact with a solid surface, has
been studied as a reaction center for electrosynthesis and ion transfer in a two-phase flow regime
and in microdroplets.18-22 Liquid interfaces and three phase boundaries have the potential for use
in numerous areas. The characterization of these interfaces and their properties, such as the
relative surface energies of the interface constituents and the thickness of thin film interfaces,
can pose a significant challenge in addition to inherent challenges in differentiating interfacial
103

and bulk phenomena. Bulk properties are often influenced by thin film geometry and the
presence of support materials. For these reasons, we have undertaken the examination of a thin
film interface between two liquids using spatially-resolved time-domain fluorescence lifetime
and anisotropy decay measurements to gain insight into the physical and chemical properties of
such interfaces.
In this work, we utilize the rotational diffusion dynamics of lissamine rhodamine B
sulfonyl chloride (LRSC) to characterize a thin film N-octyl-2-pyrrolidone (NOP)|water interface
supported on glass. We measure the rotational diffusion behavior of this chromophore as a
function of (micron-scale) distance from the interface and as a function of distance from the
glass support using time-correlated single photon counting (TCSPC) detection coupled to an
inverted confocal microscope to obtain time- and position-resolved fluorescence lifetime and
fluorescence anisotropy decay data. Lateral resolution of this system is on the order of the
diffraction limit (ca. 250 nm) and the depth of focus is less than 1 µm for a 100X objective.
In the previous chapter, we examined a compositional gradient in the size of water nanodroplets that persists for tens of microns from the glass surface into the wet NOP layer. The
discovery of the gradient in LRSC rotational diffusion dynamics as a function of distance from
the NOP|glass interface was the first report, to our knowledge, of the existence of such a gradient
in a nano-emulsion, and this gradient can be understood in the context of the relative surface
energies of the NOP|water nano-droplet and the NOP|glass interfaces.23 In this work, we have
constructed a system where there is both a NOP|water interface and a NOP|glass interface,
nominally orthogonal to one another, and we have examined the spatial-dependence of the
chromophore fluorescence lifetime and anisotropy decay time in the proximity of these
interfaces. The data reveal a gradient in the anisotropy decay time constants normal to the

104

NOP|glass interface consistent with the results from the study in chapter 4 and an absence of the
water nano-droplet size gradient normal to a NOP|water interface. We do find evidence for a
gradient in the concentration of water nano-droplets normal to the NOP|water interface, and
these findings are consistent with the nano-emulsion size being mediated by the surface energies
of the NOP|water and NOP|glass interfaces.
5.2 Experimental Methods
The chromophore lissamine rhodamine B sulfonyl chloride (LRSC) (Fig. 4.1a, ≥ 99%
Acros Organic) and N-octyl-2-pyrrolidone (NOP) (Fig. 4.1b, 98% Aldrich) were used as
received. The water for these studies was purified in-house with a Milli-Q filtration system
(Millipore). The chromophore concentration was 10-4 M in water and in dry NOP. Upon
exposure to water the chromophore is likely hydrolyzed to the sulfonic acid. We refer to the
chromophore as LRSC throughout because we have not verified the hydrolysis for each system.
For optical measurements, the sample holder was an Attofluor® cell chamber (Molecular
Probes) fitted with a round borosilicate glass cover slip (25 mm diameter, No. 1.5 Thomas
Scientific). A glass microscope slide (1 mm thick, Globe Scientific) was modified to fit into the
Attofluor® cell (inner diameter 18.5 mm) and a hole was bored through the center (inner
diameter of 5.1 mm) to form a well to contain liquid samples. The NOP|water interface was
formed by adding approximately 150 μL of Milli-Q water to the cell chamber and then
depositing ~0.01 μL of NOP containing 10-4 M LRSC to the well in the center of the microscope
slide. A side-by-side interface with a thickness of ~500 μm formed inside the well and was in
contact with the surface of the glass coverslip. The glass coverslip and glass microscope slide
were cleaned in piranha solution (a 3:1 ratio of concentrated sulfuric acid and 30% hydrogen
peroxide – caution strong oxidizer!) for 30 minutes prior to each use.

105

Fluorescence lifetime and anisotropy decay time constants were collected with the timeresolved fluorescence imaging instrument described in section 4.2. Pyrromethene dye (Exciton)
was used for sample exication in the range of 550 nm to 595 nm. An oil immersion objective
with magnification 100X (Nikon), numerical aperture 1.45, working distance 0.13 mm and depth
of field ca. 0.45 µm was used for all data acquisition, and the experimental setup is schematized
in Fig. 5.1. Depth profiles were generated on the NOP side of the interface by collecting single
measurements with the beam parked at a fixed position. Measurements were made in the NOP
phase because LRSC preferentially partitions into the NOP phase (Fig. 5.2). The positions along
the x-axis were fixed using the SPCM galvanometer control and the depth along the z-axis was
controlled mechanically. Measurements along the z-axis were collected in 10 μm increments.
Fluorescence lifetime and anisotropy decay time constants were determined using MicroCal
Origin Pro v. 9.0.
Depth into film, z
Distance from interface, x
~500 µm

NOP

water
glass

Side view

z

100x objective

y

x

Figure 5.1 Schematic of sample and optical excitation/collection configuration for
NOP|water|glass interface.

106

Figure 5.2 Time-resolved images of partitioning of LRSC initially deposited into the
aqueous phase (bottom) and migrating into the NOP phase (top). In each image, the
cuvette on left contains LRSC and the cuvette on the right is NOP|water with no
chromophore to demonstrate the existence of the phase separation.

107

5.3 Results and Discussion
As noted above, in the preceding chapter we examined a depth-dependent gradient in the
anisotropy decay time constant for the chromophore LRSC in a thin film of wet NOP supported
on borosilicate glass. Our data were consistent with the depth-dependent anisotropy decay
reflecting subtle compositional changes in the micro-scale organization of a water-in-NOP nanoemulsion. In the work presented here, we examine whether or not an analogous gradient can be
seen for the liquid-liquid NOP|water interface and within the NOP|water|glass multi-interfacial
region.
We have performed fluorescence lifetime and anisotropy decay measurements as a
function of depth into the NOP layer (z) (normal to the plane of the glass support) and as a
function of distance from the NOP|water interface (x) to resolve the spatial dependence of
chromophore reorientation. The complete data sets are tabulated in the appendix (Tables 5.1 and
5.2), while selected data are plotted for illustration purposes. We apply the same treatment to
our experimental data that we described in the previous chapter. We find that the anisotropy
decay time constant for LRSC in NOP exposed to water is consistent with a reorienting moiety
that has a greater volume than that of the chromophore alone. The reorientation data and the
steady-state absorbance spectral data point to the chromophore residing in significantly aqueous
environment(s) in (wet) NOP, and the reorientation data can be modeled in the context of the
LRSC chromophore residing in a nanoscale aqueous region within the NOP. Such a system is
described in the context of a nano-emulsion, and in this case the number of water molecules
surrounding the chromophore varies in a regular manner, increasing with proximity to the glass
support. The experimental anisotropy time constants and the hydrodynamic volume of the
reorienting entity were used to estimate the number of water molecules in closest proximity to

108

the chromophore in the context of the modified Debye-Stokes-Einstein (DSE) equation (Eq.
4.3).24-26 We found the number of water molecules to vary in number from ~30–50. The
presence and approximate diameters of such nanoscale droplets was consistent with dynamic
light scattering (DLS) data for these samples. There is some amount of uncertainty associated
with the exact number of water molecules predicted by our model, however, the trends in the
time-resolved fluorescence anisotropy and steady-state spectral data are fully valid and
consistent with the chromophore sampling an emulsion environment.
Table 5.1 Fluorescence lifetime, τfl, (ps) of LRSC as a function of depth into the NOP phase
of a NOP|water system and distance from the NOP|water interface.
Depth into
NOP (μm)

Distance from the NOP|water interface (μm)

0

12
2964 ± 7

23
2942 ± 1

34
2931 ± 3

45
2919 ± 3

56
2912 ± 3

10

2732 ± 19

2745 ± 1

2742 ± 3

2737 ± 3

2730 ± 1

20

2661 ± 13

2675 ± 3

2689 ± 3

2683 ± 3

2671 ± 3

30

2618 ± 1

2628 ± 1

2653 ± 1

2656 ± 2

2648 ± 1

40

2581 ± 3

2594 ± 1

2619 ± 1

2631 ± 2

2630 ± 2

50

2561 ± 2

2569 ± 1

2588 ± 2

2606 ± 3

2614 ± 2

60

2545 ± 7

2553 ± 1

2570 ± 3

2591 ± 1

2602 ± 1

70

2543 ± 3

2544 ± 1

2556 ± 1

2576 ± 2

2590 ± 1

80

2540 ± 1

2533 ± 3

2543 ± 2

2562 ± 2

2577 ± 3

90

2539 ± 1

2526 ± 2

2535 ± 1

2548 ± 3

2569 ± 1

100

2540 ± 5

2522 ± 2

2528 ± 1

2540 ± 3

2560 ± 2

109

Table 5.2 Fluorescence anisotropy decay time constants, τOR, (ps) of LRSC as a function of
depth into the NOP phase of a NOP|water system and distance from the NOP|water
interface.
Depth into
NOP (μm)

Distance from the NOP|water interface (μm)

0

12
1461 ± 15

23
1514 ± 17

34
1594 ± 14

45
1763 ± 11

56
1696 ± 12

10

1596 ± 21

1609 ± 15

1661 ± 5

1695 ± 11

1667 ± 12

20

1577 ± 17

1603 ± 20

1584 ± 10

1653 ± 14

1668 ± 10

30

1539 ± 2

1564 ± 15

1527 ± 5

1600 ± 5

1613 ± 11

40

1517 ± 4

1553 ± 14

1523 ± 5

1561 ± 19

1558 ± 2

50

1491 ± 15

1534 ± 11

1489 ± 16

1531 ± 1

1516 ± 18

60

1467 ± 19

1499 ± 13

1469 ± 11

1491 ± 26

1462 ± 8

70

1478 ± 9

1473 ± 8

1432 ± 5

1472 ± 6

1437 ± 4

80

1468 ± 14

1463 ± 19

1421 ± 10

1453 ± 11

1409 ± 8

90

1474 ± 18

1466 ± 18

1402 ± 5

1427 ± 10

1385 ± 18

100

1477 ± 7

1438 ± 4

1381 ± 19

1385 ± 3

1353 ± 3

The data we report here, for the NOP|water|glass system are consistent with those
reported for the NOP|glass interface, where a gradient in the anisotropy decay time is seen
normal to the NOP|glass interface for distances ca. 40 µm and greater from the NOP|water
interface (Fig. 5.3). In contrast to the gradient normal to the NOP|glass interface, there is not a
size gradient in the anisotropy decay data normal to the NOP|water interface (Fig. 5.4). The
number of water molecules in closest proximity to LRSC and contributing to the hydrodynamic
volume of the reorienting entity is found to vary in number from ~20–30.

110

Figure 5.3 (a) Reorientation time and (b) calculated number of water molecules as a
function of distance from the NOP|glass interface. Refer to Fig. 5.1 for the experimental
configuration and the x, y, z coordinate designation.

111

Figure 5.4 (a) Reorientation time and (b) calculated number of water molecules as a
function of distance from the NOP|water interface. Refer to Fig. 5.1 for the experimental
configuration and the x, y, z coordinate designation.

112

Given the presence of water in the NOP, it is presumed that the water domains are at least
the same size as those seen for the NOP|glass system in the preceding paper, however, we
observe smaller water domains. For this reason it may appear anomalous that the anisotropy
decay time constants for the NOP|water|glass system are faster than they are for the system with
only the NOP|glass interface. This is, however, an expected result for an emulsion. There is a
substantial literature on the subject of emulsions and it is known from this literature that the
viscosity of the fluid depends inversely on the average size of the emulsion inclusion
domains.27,28 We would expect this effect to become significant only under conditions where
there is a relatively high density of water nano-droplets. Thus the faster reorientation times seen
for the system characterized by both water and glass interfaces implies that the average size of
the water domains for this system are larger than for the system with only the glass interface.
It is instructive to examine the spatial dependence of the fluorescence lifetime for this
system. For the case of the single NOP|glass interface we observed no positional dependence of
the fluorescence lifetime save for surface-chromophore interactions.23 In contrast, the addition
of the NOP|water interface normal to the NOP|glass interface gives rise to position-dependent
fluorescence lifetime data in the NOP phase as a function of distance from both interfaces (Fig.
5.5). There is a steep decline in fluorescence lifetime moving away from the glass support while
the changes in lifetime as a function of distance away from the NOP|water interface are modest.
The fluorescence lifetime of LRSC is sensitive to the polarity of its environment, and spatially
resolved lifetime data can be used to infer variations in the dielectric constant (ε) of the system.
There is a phenomenological relationship between rhodamine fluorescence lifetime and dielectric
constant29 and we use this relationship to estimate the spatial dependence of the dielectric
constant (Fig. 5.6). The fact that we observe a spatial dependence to ε for this system, with two

113

distinct interfaces, while we do not observe a corresponding spatial dependence in fluorescence
lifetime for the case where there is only the NOP|glass interface is indicative of a compositional
gradient in the NOP|water|glass system. We assert the compositional gradient we observe
normal to the NOP|water interface is not a gradient in the size of the water nano-drops in NOP,
but rather a gradient in the concentration of water nano-droplets in the NOP.

114

Figure 5.5 (a) Fluorescence lifetime as a function of distance from the NOP|glass interface
for three different distances from the water interface. (b) Fluorescence lifetime as a
function of distance from the NOP|water interface for six different distances from the
NOP|glass interface. Experimental error bars representing the standard deviation are
shown for each data point. Refer to Fig. 5.1 for the experimental configuration and the x,
y, z coordinate designation.
115

Figure 5.6 (a) Dielectric constant as a function of distance from the NOP|glass interface for
three different distances from the water interface. (b) Dielectric constant as a function of
distance from the NOP|water interface for six different distances from the NOP|glass
interface. Refer to Fig. 5.1 for the experimental configuration and the x, y, z coordinate
designation.

116

It is important to consider the form of the lifetime gradient data and, in particular, to
compare it to the absence of a gradient in the two-phase (NOP|glass) system.23 In that work we
did not observe a dielectric constant gradient in the wet NOP liquid phase because there was a
fixed concentration of water in the system, determined by exposure of the NOP to water initially
and by the amount of adsorbed water at the glass surface. While there is variation in the
diameter of the water nano-droplets resulting in a size gradient, for any given region of the
sample, the total amount of water present remains nominally the same. For this reason, there is
no expectation of a dielectric gradient in the NOP|glass system studied in chapter 4. For the
three phase system, where (excess) water is in contact with both the NOP and glass phases, it is
possible for there to be not only a size gradient, but also a water nano-droplet concentration
gradient. Close to the NOP|water interface, the size of the water nano-droplets will be the same
as those that are distant from the interface because the NOP|water surface energy is the same in
both cases. The region of the intersection of the two interfaces is the meniscus of the
NOP|water|glass system formed as the denser and hydrophilic water molecules make contact
with the glass surface near the three phase boundary. While we do not have a ready means of
evaluating the profile on this meniscus, the water forming the meniscus may contribute to the
effects we observe. Near the NOP|glass interface, the water nano-droplet size is different than
the droplet size for distances greater than 10 μm for the reasons discussed in association with the
anisotropy data, but due to the presence of a meniscus in the NOP|water|glass system, the total
amount of water accessible to this interface is not limited. Thus, there can be a relatively higher
concentration of large water droplets near the NOP|glass interface, leading to a dielectric
gradient along the z-axis that is more pronounced than the gradient for the NOP|water interface
along the x-axis, and this is precisely the situation we observe experimentally (Figs. 5.5 and 5.6).

117

Taken collectively, the lifetime and anisotropy decay time constant data indicate that
there is infiltration of water into NOP resulting from the presence of the NOP|water interface.
The water concentration gradient exists normal to the NOP|water interface as is indicated from
the fluorescence lifetime data, but there is not a gradient in the anisotropy decay data to support a
size gradient. The nano-emulsion we observe normal to the NOP|water interface is comprised of
a distribution of water nano-droplets that maintain a constant average size. This is an expected
result based on surface energy considerations, where the NOP|water interface is characterized by
a surface energy essentially the same as that of the water nano-droplets in bulk NOP. Normal to
the NOP|glass interface and parallel to the NOP|water interface there is a gradient in the size of
the water nano-droplets, manifested in the anisotropy decay gradient, which can also be
understood in terms of interfacial surface energies.
5.4 Conclusions
We have investigated a system comprised of two interfaces, a NOP|water interface and a
NOP|glass interface, nominally perpendicular to one another using the fluorescent probe
molecule LRSC. Characterizing this two-interface system by means of spatial variations in the
fluorescence lifetime and anisotropy decay constant has shown that the (wet) NOP phase exists
as a nano-emulsion and that there are different compositional gradients, with one perpendicular
to the NOP|glass interface and one perpendicular to the NOP|water interface. Normal to the
NOP|water interface is a gradient in the concentration of water nano-droplets of constant average
size, and normal to the NOP|glass interface there exists a gradient in the average size of the water
nano-droplets. We understand the reason for different types of water nano-droplet gradients in
the context of differences in the surface tension for the NOP|glass and NOP|water interfaces. A

118

clear implication of these findings is that nano-emulsion size gradients can be controlled through
the identity of the solid support with which the nano-emulsion is in contact.

119

REFERENCES

120

REFERENCES

1. Interfacial Nanochemistry: Molecular Science and Engineering at Liquid-Liquid Interfaces;
Watarai, H.; Teramae, N.; Sawada, T., Eds.; Kluwer Academic/Plenum Publishers: New York,
2005.
2. Perera, J. M.; Stevens, G. W. Analytical and Bioanalytical Chemistry 2009, 395, 1019.
3. Kovaleski, J. M.; Wirth, M. J. Journal of Physical Chemistry 1995, 99, 4091.
4. Burbage, J. D.; Wirth, M. J. Journal of Physical Chemistry 1992, 96, 5943.
5. Wirth, M. J.; Burbage, J. D. Journal of Physical Chemistry 1992, 96, 9022.
6. Watarai, H.; Funaki, F. Langmuir 1996, 12, 6717.
7. Ishizaka, S.; Kim, H. B.; Kitamura, N. Analytical Chemistry 2001, 73, 2421.
8. Walker, D. S.; Richmond, G. L. Journal of Physical Chemistry C 2008, 112, 201.
9. Walker, D. S.; Brown, M.; McFearin, C. L.; Richmond, G. L. Journal of Physical Chemistry
B 2004, 108, 2111.
10. Steel, W. H.; Walker, R. A. Nature 2003, 424, 296.
11. Messmer, M. C.; Conboy, J. C.; Richmond, G. L. Journal of the American Chemical Society
1995, 117, 8039.
12. Fedoseeva, M.; Richert, S.; Vauthey, E. Langmuir 2012, 28, 11291.
13. Nagatani, H.; Fermin, D. J.; Girault, H. H. Journal of Physical Chemistry B 2001, 105, 9463.
14. Dryfe, R. A. W.; Ding, Z. F.; Wellington, R. G.; Brevet, P. F.; Kuznetzov, A. M.; Girault, H.
H. Journal of Physical Chemistry A 1997, 101, 2519.
15. Reymond, F.; Fermin, D.; Lee, H. J.; Girault, H. H. Electrochimica Acta 2000, 45, 2647.
16. Benjamin, I. Chemical Physics 1994, 180, 287.
17. Benjamin, I. Chemical Reviews 1996, 96, 1449.
18. MacDonald, S. M.; Watkins, J. D.; Gu, Y.; Yunus, K.; Fisher, A. C.; Shul, G.; Opallo, M.;
Marken, F. Electrochemistry Communications 2007, 9, 2105.

121

19. MacDonald, S. M.; Watkins, J. D.; Bull, S. D.; Davies, I. R.; Gu, Y.; Yunus, K.; Fisher, A.
C.; Page, P. C. B.; Chan, Y.; Elliott, C.; Marken, F. Journal of Physical Organic Chemistry
2009, 22, 52.
20. Collins, A. M.; Zhang, X. H.; Scragg, J. J.; Blanchard, G. J.; Marken, F. ChemPhysChem
2010, 11, 2862.
21. Collins, A. M.; Blanchard, G. J.; Hawkett, J.; Collison, D.; Marken, F. Langmuir 2011, 27,
6471.
22. Collins, A. M.; Blanchard, G. J.; Marken, F. Electroanalysis 2012, 24, 246.
23. Hay, C. E.; Marken, F.; Blanchard, G. J. Langmuir 2014, 30, 9951.
24. Chuang, T. J.; Eisenthal, K. B. Journal of Chemical Physics 1972, 57, 5094.
25. Perrin, F. Journal of Physics and Radium 1934 5, 497.
26. Hu, C. M.; Zwanzig, R. Journal of Chemical Physics 1974, 60, 4354.
27. Pal, R. AIChE Journal 1996, 42, 3181.
28. Chanamai, R.; McClements, D. J. Colloids and Surfaces A: Physicochemical and
Engineering Aspects 2000, 172, 79.
29. Snare, M. J.; Treloar, F. E.; Ghiggino, K. P.; Thistlethwaite, P. J. Journal of Photochemistry
1982, 18, 335.

122

Chapter 6
Conclusions
6.1 Conclusions
Throughout this research we sought to gain a better fundamental understanding of the
molecular structure and organization in liquid solutions and in proximity to liquid interfaces. We
utilized time-correlated single photon counting (TCSPC) to measure the fluorescence lifetime
and anisotropy decay of fluorescent probe molecules in the solvents, water and N-octyl-2pyrrolidone (NOP), separately, with addition of electrolyte, and in combination with each other.
Our results have enhanced our knowledge of the solvation properties of NOP and have
implications in the utilization of NOP|water systems in interfacial transport processes.
In chapter 2, we reported on the rotational diffusion dynamics of two structurally distinct,
polar chromophores in water and in NOP. Initial steady-state absorption and emission results
showed that resorufin and 6-[N-(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino] hexanoic acid
(NBDHA) exhibit different solute-solvent interactions in NOP. Resorufin underwent a change in
dipole moment upon excitation, while no analogous change was observed for NBDHA indicating
different solvation environments. The fluorescence anisotropy of each of these probe molecules
provided more information on the local solvation environment. Both probe molecules were
found to reorient as prolate rotors in water exhibiting one-component anisotropy decays and as
oblate rotors in NOP exhibiting two-component anisotropy decays. This change in rotor shape
indicates the rotational motion changes from out-of-plane rotation in water to in-plane rotation in
NOP. The more restrictive in-plane rotation occurring in NOP was attributed to solvent-solvent
interactions, as well as the size of the solvent relative to the solute. The similar behavior of the
two structurally different chromophores in water and in NOP is evidence that the solvent-solvent

123

interactions in NOP mediate the rotational motion and play a more significant role than the
solute-solvent interactions.
In chapter 3, we assessed the effect of electrolyte on the rotational diffusion dynamics of
resorufin in water and in NOP. The addition of a quarternary ammonium electrolyte to NOP
disrupted the solvent-solvent interactions discussed in the previous study. We arrived at this
conclusion through the observation that resorufin exhibited a change in rotor shape from oblate
in pure NOP to prolate in NOP containing tetrabutyl ammonium bromide (TBAB) or tetraoctyl
ammonium bromide (TOAB). The reorientation times of the anionic chromophore were
concentration dependent, which points to ion pair formation between the resorufin anion and the
ammonium cations. Equilibrium constants for complex formation (2 ± 1 for TOAB and 11 ± 8
for TBAB) were consistent with a stronger complex forming between TBAB and resorufin
compared to the TOAB-resorufin complex, which is expected due to steric hindrance issues
associated complex formation with the tetraoctyl ammonium cation.
In aqueous electrolyte solutions of lithium perchlorate (LiClO4), resorufin exhibited inplane rotation as was the case in pure water and concentration dependent reorientation times
were observed. The complex formation between the resorufin anion and a Li cation, which is
relatively small, does not make a significant impact on the rotational dynamics of resorufin.
These results showed the solute-electrolyte interactions have an influence on rotational motion
that can disrupt solvent-solvent interactions. The solute-solvent and solvent-solvent interactions
studied in water and in NOP set the stage for the next step of the research which was forming
NOP|water interfaces.
In chapter 4, we utilized a TCSPC fluorescence anisotropy and lifetime confocal imaging
instrument to generate depth profiles as a function of distance from a NOP|glass interface. Using

124

lissamine rhodamine B sulfonyl chloride (LRSC) as a probe molecule in wet NOP, a
fluorescence anisotropy decay gradient was revealed. This anisotropy decay gradient, which
showed decreasing reorientation times with increasing distance from the glass surface, was
determined to result from a compositional gradient due to the presence of water nano-droplets in
the NOP phase. In this study, NOP was exposed to water prior to the analysis and some mutual
solubility allowed for water to infiltrate the NOP phase. The decreasing trend in reorientation
times as a function of distance from the NOP|glass interface corresponds to the changing
molecular environment experienced by the LRSC as the size of the water inclusion decreases.
Absorbance spectra of LRSC in water, NOP, and NOP exposed to water support our theory that
LRSC resides in an aqueous environment, while dynamic light scattering data (DLS) confirmed
the approximate particle size of the nano-droplets (ca. 1.26 nm diameter). Depth profiles
collected as a function of distance from an ethylene glycol (EG)|glass interface showed no
analogous anisotropy gradient. The absence of a gradient in EG, which is miscible with water in
all proportions, is significant because it rules out other possible phenomena and supports the
theory of a compositional gradient in NOP. These findings demonstrated that the properties of
NOP allow for uptake of small amounts of water and the water contained in the nano-emulsion
varies in size based on proximity to a liquid-solid interface. The results of this study allowed for
the interpretation of findings in a phase separated system of NOP|water.
In chapter 5, we measure the rotational diffusion dynamics of LRSC in a two interface
system composed of NOP and water on a glass support. Fluorescence anisotropy and lifetime
were collected in the NOP phase as a function of distance from a NOP|water interface and a
NOP|glass interface. The two interfaces situated normal to one another formed a three phase
liquid|liquid|solid boundary and both exhibited fluorescence lifetime gradients as a function of

125

distance from the interfaces, while a fluorescence anisotropy gradient was only recorded normal
to the NOP|glass interface. The anisotropy gradient confirms the result of the previous study,
which reported the presence of a compositional gradient resulting from water nano-droplets of
decreasing size in the NOP phase. The lifetime gradients indicate dielectric gradients emanating
from each interface. A slight lifetime gradient exists normal to the NOP|glass, while the gradient
normal to the NOP|water interface is more pronounced. The presence of a lifetime gradient
normal to NOP|water, where no anisotropy gradient exists, is taken to be a sign that the water
nano-droplets remain a constant average size, but are present in different concentrations with a
greater concentration near the NOP|water interface. The constant average size extending from
the NOP|water interface is an expected result because the surface tension between water and
NOP should remain constant from the NOP|water interface to the water nano-droplets in the bulk
NOP.
The results of this research have given us a more in depth understanding of the solvent
systems, water and NOP, and their interactions in a liquid-liquid system. Rotational diffusion
measurements have allowed us to establish the organization surrounding chromophores in neat
liquids and in nano-emulsions. The discovery of water nano-droplets in NOP with size and
concentration gradients has implications in containment and transport of polar molecules through
nonpolar barriers.
6.2 Future Directions
This research has laid the groundwork for future studies on the unique properties of
NOP|water interfaces and emulsions. The discovery of water nano-droplets present in the NOP
phase is influential in the direction of future work on NOP containing systems. Our
measurements were performed on borosilicate glass cleaned with oxidizing agents making the

126

surfaces hydrophilic. An interesting follow-up would be to change the character of the solid
support by siliconization of glass, for example, to make the surface hydrophobic in order to
assess the effect of surface character on droplet formation and size. In our studies, the
hydrophilic glass support essentially acted as a reservoir for water to move into the NOP phase,
so it is expected a hydrophobic surface would limit the availability of water in a wet NOP|glass
system.
The polar rhodamine chromophore used in the studies of NOP|glass and NOP|water|glass
systems in chapters 4 and 5 was solvated in an aqueous environment that was surrounded by
NOP molecules. The presence of water in NOP necessitates that the amphiphilic NOP molecules
likely surround the water by forming inverse micelle-like structures. Incorporation of a nonpolar
chromophore in the acyl chains of the NOP secondary structures would provide another
perspective on the local environment in a binary system. Alternately, a study could be done on
the oil-in-water structures that may be present in the water phase of the NOP|water|glass system.
The solubility of NOP in water is documented (0.12% (v/v) (20 °C))1 and, therefore, the water
phase of the three phase system is likely to contain NOP. The secondary structures formed by
the amphiphilic NOP molecules is not known. Although no data were collected, observations
were made of fluorescent droplets in the water phase of the three phase system. We know from
our measurements in the wet NOP presented in chapter 4 that the polar rhodamine molecules
partition into an aqueous environment leading to the possibility of multilayer micellar structures
of NOP in the water phase of the NOP|water|glass system. Obtaining rotational diffusion data on
a polar chromophore within one of these droplets would provide information on the local
environment of the secondary structure.

127

At the end of chapter 4, we alluded to a phenomenon in which particles near solid
surfaces exhibit slower diffusion due to hindered Brownian motion. Delving deeper into this
phenomenon in the wet NOP|glass system would provide more information about the formation
and movement of the water nano-droplets near the solid support. Confocal laser scanning
microscopy has been used for imaging optical slices of microtubes containing water previously.2
Park et al. exploited the lateral and depth resolution of confocal microscopy to scan at different
depths and create a 3D map of particle velocities. In our confocal imaging system, we have a
high magnification oil immersion objective (100X, NA 1.45) capable of adequate depth
resolution for imaging different focal planes of a colloidal suspension.
Beyond altering the current system to gain new information, there are additional avenues
to take this research which include studying transfer processes. Marken et al. showed coupled
ion and electron transfer occurs at the three phase boundaries using electrochemical techniques.35

Our confocal imaging setup could enhance what is known about the local environment during

these processes by coupling spectroscopic and electrochemical techniques. In chapter 5, we
showed fluorescence lifetime and anisotropy gradients in a three phase NOP|water|glass system.
The replacement of glass with a conductive and optically transparent solid support, such as
fluorine tin oxide (FTO) and incorporation of a redox couple would be the first steps for making
simultaneous electrochemical and spectroscopic measurements.
This work as a whole has characterized the local organization of NOP from strong
solvent-solvent interactions in neat solution to accommodation of water droplets and formation
of two phases with water. Our studies on the properties of NOP will be a resource for future
work on NOP and its inclusion in new industrial formulations. Additionally, it is our hope that
the research presented here on depth profiles using the fluorescence lifetime and anisotropy

128

confocal imaging instrument will be applied to gain fundamental understanding of new
interfacial systems.

129

REFERENCES

130

REFERENCES

1. Login, R. B. Journal of the American Oil Chemists Society 1995, 72, 759.
2. Park, J. S.; Choi, C. K.; Kihm, K. D. Experiments in Fluids 2004, 37, 105.
3. Marken, F.; Webster, R. D.; Bull, S. D.; Davies, S. G. Journal of Electroanalytical Chemistry
1997, 437, 209.
4. Wadhawan, J. D.; Compton, R. G.; Marken, F.; Bull, S. D.; Davies, S. G. Journal of Solid
State Electrochemistry 2001, 5, 301.
5. MacDonald, S. M.; Watkins, J. D.; Gu, Y.; Yunus, K.; Fisher, A. C.; Shul, G.; Opallo, M.;
Marken, F. Electrochemistry Communications 2007, 9, 2105.

131