5

Particle Size Distribution and Zeta Potential

Based on Dynamic Light Scattering:
Techniques to Characterize Stability
and Surface Charge Distribution
of Charged Colloids
SANAT KARMAKAR

ABSTRACT

Size distribution and zeta potential based on dynamic light scattering

(DLS) are the essential characterizations prior to any applications of
some charged colloids. Nowadays, these are the routine methods used
in a wide variety of research areas, such as, stability of nanoparticles,
colloids, bio-colloids and polymer science etc. In this chapter, the basic
techniques, principles and data analysis will be discussed in details.
First, theory, experiment, data analysis and inference from DLS
techniques will be discussed. Next, we shall discuss zeta potential as a
tool to characterize the stability and surface charge distributions of
charged colloids. Some examples of usages of these techniques in
material science will also be discussed.

Key words: Dynamic light scattering, Zeta potential, Colloids, Surface

charge, Electrophoresis, Debye length

Light scattering from a matter is widely used non-invasive technique

with applications in numerous scientific disciplines[1]. The specific

Soft matter and Biophysics Laboratory, Department of Physics, Jadavpur University,

Kolkata – 700032, India
* Corresponding author: E-mail: [email protected]
118 Recent Trends in Materials: Physics and Chemistry

properties of the molecules can be studied depending on the light source

and the detector used. In 1859, the scattering of light, observed by
John Tyndall in colloids, is one of the earliest light scattering
experiments[2]. Later in 1904, the Lord Rayleigh explained the blue
colour of the sky, which is caused by the light scattering from smaller
atmospheric particles with respect wavelengths of light. This result
established the fact that the refractive index of the medium plays a
vital role in light scattering phenomenon. In the early twentieth century,
Gustav Mie (1908) established the theory to investigate the light
scattering from absorbing and non-absorbing media containing particles
that are larger compared to the wavelength of light[3]. The Mie theory
also takes into account particle shape and the contrast in refractive
index between particles and the medium in which particles are
suspended. However, Peter Debye suggested that the scattering of
particles can be studied independently of assumptions on mass, size, or
shape as a function of angle, which is often referred to as Rayleigh–
Debye scattering[4]. Detection and analysis of absolute mean intensity
as a function of scattering angle is known as static light scattering (SLS).
On the other hand, intensity fluctuations of scattered light provide
information on the dynamics of the systems. The measurement and
analysis of such intensity fluctuations to obtain the material properties
is termed as dynamic light scattering (DLS). In the present chapter we
shall discuss only the various aspects of DLS.

The DLS technique provides us the information about the material

properties at “long wavelength” limited by the wavelength of light. This
technique is a consequence of the interaction of light with particles and
often known as photon correlation spectroscopy (PCS) and quasi–elastic
light scattering[5]. DLS mostly uses the Mie theory of scattering to
estimate the particle size[3]. Information, such as, size of the particles,
molecular weight, diffusion constant, interaction length, second virial
coefficient etc are obtained. This technique is also used to obtain the
electrostatics and the dynamic behaviour of charged particles in a
solution. Zeta potential is one such property of charged colloids which
can be measured from a technique known as laser Doppler velocimetry
along with phase analysis light scattering based on DLS[6]. In contrast
to DLS, other scattering techniques, such as x-ray or neutron provide
us information at much shorter length scale typically of inter-atomic
distances (Table 1)[7]. Scattering of light can occur whenever there are
variations of refractive index within the materials, whereas, x-ray (much
shorter wavelength compared to visible light) scattering depends on
contrast in the electron density of the materials. Spatial variation of
refractive index of the materials leads to static scattering where one
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 119

measures the intensity as a function of scattering angle. The intensity

fluctuations of scattered light with time give us information on the
dynamics of the particles in solution. In dynamic light scattering (DLS),
the intensity fluctuations with time are measured to obtain the size
information.

DLS is widely used to analyze size distributions of nano-particles,

proteins, latex and colloids[8]. DLS provides us with an understanding
of stability as well as the dynamics of nano and micron-sized particles.
Nowadays, this technique has drawn a significant attention for
biomedical applications[9]. The dynamic light-scattering technique has
a number of advantages over other experimental methods. For instance,
it is possible to perform experiments with a wide range of sample buffer
and wide range of temperature as well as concentrations. DLS is also a
non-invasive technique that requires comparatively low amounts of
sample and provides reliable estimates of the quality of preparation
rapidly.

Table 1: Different scattering techniques for the determination of the structure of the
materials at various length scale.

In this chapter, we will discuss the basic principle of dynamic light

scattering and its uses to characterize micron or even nano-sized particles.
First, we discuss the detailed experimental technique to determine the
size distributions of particles and then we discuss the measurement of
electrophoretic mobility for charged colloids. Electrophoretic mobility
provides information about the zeta potential which is a good
approximation of surface potential of the charged particles in solution.
The knowledge of zeta potential also provides us the electrostatic
interactions and stability of the particles in a solution.
120 Recent Trends in Materials: Physics and Chemistry

1. DYNAMIC LIGHT SCATTERING TO CHARACTERIZE THE

DIFFUSION OF PARTICLES IN SOLUTION

The relation between the diffusion behavior of particles in solution and

dynamic light scattering was established by a series of seminal
discoveries. The theory of Brownian motion was explained by
Smoluchowski assuming the kinetic model of colliding spheres[10]. This
theory results how fluctuations give rise to an appreciable effect when
the effect of an individual collision is negligible. On the other hand, a
similar result was obtained by Einstein assuming the probabilistic aspect
instead of kinetic consideration. The mean square displacement of the
particles estimated from Brownian motion is directly proportional to time.
Further, Langevin theory and more generalized theory given by Fokker
and Planck establish the diffusion equation from Brownian motion.
Later, Einstein also established that Brownian motion can be mapped
into the diffusion problem. The relationship between the diffusion
coefficients of particles to their translational friction along with the
discovery from Sir George Stokes (1845) suggested that the friction
exerted by a moving particle is proportional to its radius and to the
viscosity of the solvent surrounding particles[11]. The density and the
concentration fluctuations of the particles in a solution due to Brownian
motion lead to intensity fluctuations in the scattered light when a
monochromatic light beam encounters the system.

Dynamic light scattering (DLS) measures the particle size in sub-

micron regime from the intensity fluctuations of the scattered lights.
Translational diffusion is strongly dependent on the temperature and
viscosity of the medium in which particles are dispersed. The root mean
square displacement of the particle relates the diffusion constant which
in turn allows us to estimate the particle size via Einstein stokes
relation[12]. The intensity fluctuations of scattered light determine the
intensity-intensity correlation function which can be used to obtain the
diffusion coefficient and hence the particle size in the system. Over the
years, there have been a number of experimental and theoretical
approaches to determine the diffusion coefficient of particles in a solution
which eventually lead to the development of modern DLS instruments[13-
17]
. Let us discuss the theory of DLS, which will tell us how the measured
intensity autocorrelation is related to diffusion constant as well as particle
size distributions.

1.1 Theory of Dynamic Light Scattering

When a monochromatic light typically laser beam is incident on a

sample, scattering of light occurs due to local fluctuations of refractive
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 121

index of the medium. Usually the refractive index of the particles is

different from the solvent. Light is primarily scattered by particles
suspended in a solution executing Brownian motion. Intensity
fluctuations of scattered light are measured and analyzed to obtain
diffusion constant and hence the hydrodynamic radius of the particle.
The movement of particles in a solution depends on their size,
temperature, and solvent viscosity[18]. It has been well established that
for a fixed temperature and solvent viscosity, the larger particles diffuse
slowly compared to smaller particles. As the intensity fluctuations of
the scattered light occur as a result of local particle density fluctuations
due to Brownian motion, the intensity fluctuation also appears to be
very rapid for smaller particles relative to larger particles as described
in Fig. 1. The continuous Brownian fluctuation of the molecules causes
frequency shift of the monochromatic incident light and the amount of
shift depends on the size of particles. This phenomenon is known as
Doppler broadening. The spread of frequencies, ∆ω, is dependent
on the diffusion coefficient D as ∆ω = Dq2, where q is the scattering
vector.

In a typical scattering experiment, intensity fluctuations, measured

at two different times t and t + τ, look like a statistical noise. However, If
τ is much smaller than the characteristic time of fluctuation, these
intensities measured in the time interval τ are correlated. Therefore, we
define intensity auto correlation function as
�
1
��(�0 ) �(�0 + �)� = lim � �(�)�(� + �)�� = �(�) (1)
� →∞ �
�0

Since t0 is Wthe initial or starting time and can be taken as t0 =0.

Average of intensity – intensity data was taken over the duration of
the experiments. In DLS, the amplitude fluctuations of the electric field
can be described in term of normalized correlation function.

���� (�, � ) ��� (�, � + �)�

�1 (�, �) = (2)
< |��� (�, �)|2 >

where is the scattering amplitude.

Now, consider light is incident on a system of N particles. Now the

electric field of the incident light can be represented as ���� = � � −� � �� .� � .
The electric field of the scattered light from the j th particle at position
rj(t) is given by �� (�) = �� � −� (��� − � � ).� � (�) = �� � −� �.� � (�) , where q is the
122 Recent Trends in Materials: Physics and Chemistry

4 � sin �
scattering vector of magnitude . For system of N particles we
�
can obtain the scattering amplitude as
�

��� (�) = � �� � −� �.� � (�) (3)

� =1

Normalized field autocorrelation function

−� �.� � (�) �
〈∑�
� =1 �� � ∑�=1 �� � −� �.� � (�+�) 〉
�1 (�) = −� �.� � (�) 2 (4)
〈| ∑�
� =1 �� � | 〉

�1 (�) = 〈� −� �.Δ� (�) 〉= �(�, �) (5)

The particle distribution function can be defined as

�(�, �) = 〈�( � − {�� (�) − �� (0)}〉

Now the scattering amplitude is the Fourier transform of �(�, �) .

Therefore, the intensity autocorrelation function is the Fourier
transformation of the power spectrum, and therefore the DLS
measurements can be equally well performed in the spectral domain[19].
∞
�(�, �) = � � −�� .� �(�, �)� 3 � (6)
0

According to the theory of Brownian motion, �(�, �) satisfies the

diffusion equation.

��(�, �)
= �∇2 �(�, �) (7)
��

where, D is the diffusion constant. By Fourier transforming the

equation 7 equation we get

��(�, �)
= �q2 �(�, �) (7a)
��

Solution of the equation 7a gives

2�
�(�, �) = � −�� (8)

2
Therefore, �1 (�) = � −�� � .. The electric field correlation, �1 (�) decays
exponentially and is dependent on a decay constant, à = �� 2 , for
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 123

monodisperse particles undergoing a Brownian motion. G(τ) is related

to g1(τ) via the Siegert relation based on an approximation that the
photodetector detects only the scattered light and that the photon
counting is a random Gaussian process. Therefore, the intensity-intensity
autocorrelation function can be written as
2�
�(�) = � + �|�1 (�)|2 = � + �� −2�� (9)

where β is the coherence factor that depends on detector area, optical

alignment, and the scattering properties of macromolecules. A is the
baseline and can be normalized to a value 1. It is important to mention
here that, for a polydisperse sample, g1(τ) cannot be represented as a
single exponential decay rate Γ, but can be expressed as the intensity
weighted average of distribution of decay rate G(Γ). In this case,
∞

�1 (�) = � �(Γ)� −Γ� �Γ . The equation (9) is used to fit the measured intensity
0

autocorrelation function to obtain the diffusion constant D. The size of

the particles in terms of hydrodynamic radius (Rh) has been estimated
using the Einstein - Stoke relation.

��
�ℎ = (10)
6 ���

Where kT being the thermal energy (k: Boltzmann constant, T:

temperature in ºK) and be the coefficient of viscosity of the medium (k
= 1.38 × 10–23 Joule/Kelvin). Hydrodynamic radius refers to the radius
of the sphere that diffuses with the same velocity as that of the particle.
��
Now the frictional coefficient C can be expressed as = �
.. The frictional
coefficient can also be written using equation (10) as

� = 6���ℎ (11)

Theory of DLS assumes only spherical particles. Therefore, deviation

from spherical shape can be envisaged from measured frictional
coefficient (equation (11)) when compared with the frictional coefficient
of unsolvated volume of the particle (actual radius sphere).

1.2 Experiment

In a typical DLS experiment, laser light is used to illuminate the sample

cuvette. Red He-Ne or semiconductor laser of wavelength 630 nm and
power of 2–4 mW is used. The sample must be filtered to eliminate the
124 Recent Trends in Materials: Physics and Chemistry

Fig. 1: Schematic representation of the relation between particle size and intensity
fluctuations and hence the correlation function.

unwanted larger particles. In the dilute limit most of the laser beam
passes through the sample and some are scattered by the particles. The
scattered light is then detected by either the photo detector or avalanche
photo diode. If the concentration of the particle is very large, then too
much scattered light will be detected. In this situation, an attenuator is
placed before the sample (as shown in Fig. 2) to reduce the incident
laser intensity to a desired intensity. In a DLS experiment, scattered
intensities are measured at a fixed scattering angle, typically at 90º or
173º (i.e., back scattered light). It is important to mention that the
rotational diffusion of molecules will affect the observed autocorrelation
function. The effect of rotational diffusion on the observed
autocorrelation function can be ignored when scattered light is detected
at 173º (i.e., the back scattered light) compared to the detection at 90º.
Therefore, by detecting the back scattered light only the translational
diffusion D can be obtained. Detection of back scattered light also allows
minimizing multiple scattering by the sample and hence it is possible to
use high concentration sample for DLS measurement. Further, scattering
from unwanted large dust particles present in the solution is mostly in
the forward direction compared to the smaller particles which scatter
almost equally in all directions. Modern light scattering instruments,
such as zetasizer nano ZS (Malvern Instrument, UK) have provision
for detection of scattered light at 90º as well as 173º.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 125

Fig. 2: Schematic representation of scattering geometry and experimental setup for

dynamic light scattering experiment. Here ���� �� and �����
�� are the incident and
scattered wave vectors and q is the resultant scattering vector whose magnitude
is given in the figure.

The scattered signals from the detector are then fed to the digital
correlator. The rate of intensity fluctuations is monitored by the correlator
which compares the scattered intensity at successive time intervals.
Correlator is connected to a computer which provides the correlation
curve G (τ) vs. τ. Sample properties, such as dispersant viscosity and
refractive index are required to calculate the size distribution. For better
DLS result, the sample should be transparent to light. However, for
efficient scattering, it is essential to have sufficient contrast in the
refractive indices between the particle and the dispersant. The
concentration of the sample depends on the particle size and refractive
index of the dispersant. Therefore, the choice of concentration of the
sample is important in order to obtain efficient scattering from particles.
Excessive diluted particle dispersion will give us a poor signal to noise
ratio. However, the lower limit of concentration depends on the laser
power, detector sensitivity and optical configuration of the instrument.
Concentrated particle dispersion should also be avoided in order to
restrict multiple scattering. High concentration of sample will hinder
free diffusion of the particles due to interparticle interaction, leading to
restricted Brownian motion. Therefore, in order to determine the
optimum desired concentration, one would like to perform a series of
size measurement as a function of concentration. The experiments should
be performed over a certain range of the concentration for which average
distributions of particle size do not alter significantly.
126 Recent Trends in Materials: Physics and Chemistry

Figs. 3: Interpretation of the correlation curve obtained from intensity fluctuations

is illustrated in Figs. a-b shows the typical size distribution obtained from
cumulant analysis of the correlation data. The analysis essentially provides
the mean size and polydispersity index. Time, when decay of the correlation
function starts indicates the mean size of the particle. Therefore, decay starts
at larger time for larger particles than smaller particles. Fig. c corresponds to
much larger particles compared to (a). However, the rate of decay, i.e., the
slope or gradient of decay determines the polydispersity of the size distribution.
Larger the slop greater the polydispersity of the sample, as reflected in the
width of the size distribution depicted in (c). The width of the size distribution
is a measure of polydispersity and quantitative measure of polydispersity is
discussed section 1.3. The value of Y- intercept is normalized to one for ideal
signal.

1.3 Interpretation of Data

The correlation function (equation 9) contains all information regarding

the diffusion of particles in a solution. In order to obtain reliable
information on diffusion coefficient, two approaches are widely used to
fit the correlation function – monomodal distribution and nonmonomodal
distribution methods. For a distribution with mono-disperse particles,
the correlation curve appears smooth and single exponent decay
function. However, for polydisperse sample or sample with different
particle sizes, the correlation curve is a sum of many experimental
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 127

decays. For instance, we obtain two decay constant, corresponding to

two different diffusion constants for bimodal distributions. The measured
intensity auto correlation function is fitted to equation (9) to obtain the
translational diffusion constant (D).

The cumulant analysis method provides mean values of the diffusion

coefficient. This method was widely popular due to its ease and
reliability [20] , and was accepted by the International Standards
Organization (ISO) in 1996 and again in 2008 (ISO, 2008).

In DLS, average size (z average) and polydispersity index (PDI) are

calculated from the cumulant analysis of intensity autocorrelation
functions. PDI measures the extent of polydispersity of the sample which
can be envisaged from the width of the size distribution. Now, equation
(9) yields
2
�(�) = ��1 + �� −2�� � � = �(1 + �� −2Γ� ) (12)

Where, à = ��2. To account for the polydispersity or the peak

broadening, we can expand the argument of the exponential part as
2
�(�) = �(1 + �� (−2Γ�+ � � ) ) (13)

The intensity autocorrelation function is then fitted to a linearized

expression obtained from equation 13.

1 1 �
�(�) = ln{�(�) − �} = ���� − Γ� + � 2 = �0 − �1 � + �2 � 2 (14)
2 2 2

Here a1 measures the intensity weighted average of mean size

(Z D) and a 2 measures the variance of the distribution and hence
2�2
polydispersity via the relation ��� = � 2 , If we assume a single size
1
population of a Gaussian distribution, the PDI can be directly written
�2
in term of standard deviation of the Gaussian distribution ��� = .
�� 2

Typical size distributions obtained from the DLS are shown in Fig.
3b. For analysis the viscosity and the refractive index of the dispersant
must be known with a great accuracy. Broader the distribution, more
the polydispersity of the sample. The width of the size distribution,
estimated from the rate of decay is a measure of polydispersity. Unlike
the cumulant analysis, non-monomodal methods are more suitable for
polydisperse systems and do not assume a certain type of distribution of
128 Recent Trends in Materials: Physics and Chemistry

diffusion properties. There are several methods developed for broad

monomodal or multimodal distributions of data. The nonnegative least
squares (NNLS) method[21] and constrained regularization method for
inverting data (CONTIN)[22] are two most popular data analysis methods.

In the analysis of correlation statistics of time series data, the plot of

autocorrelation function is often referred to as correlogram. Let us discuss
the qualitative information obtained from the correlogram (Fig. 3). The
smaller particle correlation curve decays at very short time scale
compared to the larger particles (Figs. 3a-c). However, the slope of the
decay curve depends on the polydispersity of the particle (Fig. 3c). If
the particles are highly monodisperse, the decay is very fast, i.e., the
correlation curve becomes very stiff. Therefore, the rate of decay, i.e.,
the slope of the correlation curve provides the information about the
polydispersity of the sample. The quality of data is usually determined
from the signal to noise ratio. The intercept on the Y-axis in the
correlogram gives information about the signal to noise ratio. For an
ideal signal, the Y-intercept is normalized to a value one (Fig. 3a).
Therefore, any value of Y-intercept greater than 0.9 is the best data
and below 0.6 refers to a bad quality data. The photon count rate of the
detector is often useful to monitor the stability as well as the quality of
data. Usually attenuator is adjusted to optimize the count rate for good
quality data. The Mie theory can be used to calculate the volume
distribution and number distribution of the particle. Assumptions that
one makes here are all particles are spherical in shape, homogeneous
and equivalent density. This data are often useful when we need to
compare the relative population or the amount of materials in separate
peaks.

For unstable particle dispersion, the random Brownian motion will

be disturbed due to sedimentation of the particles. As DLS measures
the random movement of Brownian particles, it is not applicable to the
non-random movement leading to sedimentation of particles. Therefore,
upper size limit is strongly depends on the onset of sedimentation and
number fluctuations of particles. However, DLS measurement is still
applicable, if the rate of diffusion is must faster than the rate of
sedimentation. A clear evidence of aggregates or sedimentation is seen
in the tail of the correlogram (Fig. 4). An increase in correlation or the
rapid fluctuations at high delay time, as seen in Fig. 4 is an indication
of sedimentation or aggregation of particles. It is important to mention
that if we observe a gradual decrease in the stable count rate over
multiple measurement of the same sample , this indicates the loss of
particles due to sedimentation. Such a sample may not be suitable for
DLS measurement.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 129

Fig. 4: Multimodal distribution of particles. The sample is too polydisperse and the
clear signature of aggregation of particles is seen from the tail of the
correlogram, marked by a circle. Base line (Y- intercept) in the correlogram is
below 0.8, indicating poor signal to noise ratio. It is important to mention that
the large fluctuations of number of particles in the measurement volume can
cause elevated baselines, leading to Y-intercept values > 1. This can happen
when the particle size is large and concentration of sample is too low.

1.4 Factors Affecting the Hydrodynamic Radius

The factors which alter the Brownian motion or translational diffusion

will invariably change the hydrodynamic radius. One of the most obvious
factors is the temperature which needs to be constant throughout the
experiment. Any temperature gradient will lead to convection flow,
which in turn ruin the correct interpretation of the size distribution.
The effect of temperature on the mean diameter of large unilamellar
vesicles (can be considered as a biocolloids) is shown in Fig. 5. Viscosity
is also a crucial parameter, which could change the rate of translational
diffusion and hence the particle size. This is why both temperature and
coefficient of viscosity appear in the Einstein-Stokes relation. Other
factors that could change the hydrodynamic radius are ionic strength
of the medium and shape of the particles. The fluid layer associated
with the particles might be influenced by the surface structure of the
particles. Therefore, the surface structure could change the diffusion
speed and hence the hydrodynamic radius. Needless to say that
polydispersity is an important factor in interpreting the size distribution
data. We will now discuss the polydispersity of the particle and its
physical significance.
130 Recent Trends in Materials: Physics and Chemistry

Fig. 5: Temperature dependence of large unilamellar vesicles, made from phospholipid,

on the mean hydrodynamic radius.

1.5 Polydispersity

Polydispersity is a measure of dispersion or the standard deviation of

size distributions. The quantitative description of polydispersity, called
polydispersity index (PDI) which indicates the extent of polydispersity
of distribution is defined in section 1.3. In the case of polymer chemistry,
polydispersity index is defined as the ratio of weight average molecular
weight to the number average molecular weight[23]. The term PDI is,
therefore, often used to indicate the width of the molecular distributions
of polymers[24]. However, in the context of DLS, width of the size
distribution is a measure of PDI. As shown in Fig. 3c, the slope of the
correlogram and width of the size distribution provide the idea of
polydispersity of the sample. The cumulant analysis of correlogram
described in section 1.3 gives the expression of PDI. PDI greater than
0.7 is considered to be too polydisperse to measure the size distribution.
PDI<0.05 is purely monodisperse and monomodal distribution is obtained.
Usually the highly monodisperse distribution is obtained only for
standard sized latex polystyrene particles.

The major drawback that we encounter in DLS is the size distribution

from non-spherical particles. Einstein–Stoke relation assumes the
hydrodynamic radius of spherical particles only. Therefore, DLS
measures equivalent spherical particles which have the same average
translational diffusion as that of the non-spherical particles. It is
important to mention that DLS never provide actual size of the particle
as it measures only the hydrodynamic radius associated with the particle.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 131

For accurate size information, one needs to adopt other experimental

techniques, such as transmission electron microscopy. However, DLS
can readily be used for several biomedical applications, such as, detecting
aggregation of recombinant proteins, protein-protein, protein-RNA,
protein-small molecule interactions[9] and many more. We shall now
describe two specific applications where DLS is used to study interaction
between molecules.

1.6 Interaction of Monovalent Ions with Unilamellar Phospholipid

Vesicles

Unilamellar vesicle is an enclosed lipid bilayer shell and serves as an

excellent model system of biological membranes. Here we describe the
size distribution of large unilamellar vesicles (LUV) made from mixture
of zwitterionic phospholipid dioleoyl phosphatidylcholine (DOPC) and
negatively charged phospholipid dioleoyl phosphatidylglycerol
(DOPG) [25,26]. The importance of the study of this system will be
discussed later in this chapter. LUV are prepared in aqueous solvent
using an extrusion technique (LiposoFast from AVESTIN (Canada)
as described by Hope et al.[27]. Multilamellar vesicles suspension is
extruded through polycarbonate membranes having pore diameters
of 100 nm. This results in the formation of well defined size of LUV.
This method produces vesicles of diameter ~100 nm, as measured by
dynamic light scattering (Fig. 6). Size distribution of LUV containing
ions was obtained at room temperature (25ºC) with Zetasizer Nano ZS
(Malvern Instruments, UK). The Zetasizer Nano uses 4 mW He-Ne
laser of wavelength 632.8 nm.

Fig. 6: Typical size distribution LUV with mean diameter 100 nm. PDI obtained from
the distribution profile is 0.06 which indicates LUV are fairly monodisperse.
132 Recent Trends in Materials: Physics and Chemistry

Various concentrations of NaCl are explored starting from 10 mM to

500 mM. The average size in terms of the effective hydrodynamic radius
of LUV for different NaCl concentration is shown in Fig. 7. PDI is also
presented for each salt concentration in the inset of Fig. 7. The diffusion
constant is calculated from the Einstein-Stokes equation (equation 10)
and is found to be 1.25 × 10-12m2/s which is in agreement with the value
obtained from the simulation study[28]. It is evident from the Fig. 7 that
the average radius slightly decreases as the salt concentration is
increased. Above 100 mM average size again increases and eventually
stays constant with a value similar to that without salt concentration.
The presence of ions in the medium in which negatively charged LUV
are dispersed, can, in principle, alter the hydrodynamic radius by
changing the thickness of the electrostatic double layer also known as
Debye length. It is interesting to note that the average size of LUV
decreases with the increase of salt in the intermediate concentration
range. The adsorption of ions on LUV compresses the electrostatic double
layer, resulting in a decrease in the hydrodynamic diameter. At a higher
salt concentration >100 mM, charge screening effect overcomes the
adsorption process and therefore, no more decrease in hydrodynamic
radius have been observed. This result indicates the evidence of screening
effect at higher concentrations of salts. Therefore DLS technique can
provide us significant insight into the ion-membrane interaction.

Fig. 7: The Average size (d nm) of LUV composed of DOPC-DOPG (4:1) mixture for
different concentration of NaCl. Polydispersity index (PDI) is shown in the
inset table. Error bars were obtained from 5 independent measurements.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 133

1.7 Interaction of Antimicrobial Peptide with Negatively Charged

Unilamellar Phospholipid Vesicles

Antimicrobial peptides (AMP) are the unique and diverse group of

molecules which are part of innate immune response found in all animal
and human bodies. The NK-2 is one such potent antimicrobial peptide
derived from cationic core region of NK-lysin found in natural killer cells.
The interaction of NK-2 with phospholipid membranes has been studied
using DLS in order to get insights into the mechanisms of pore formation
and membrane disruption[29]. The DLS technique can, in principle, show
the evidence of membrane–membrane interaction mediated by NK-2 or
the leakage of LUV induced by NK-2. DLS technique shows the evidence
of strong interaction between the phospholipid membranes and the
peptide. The increase in polydispersity and eventual aggregation of
vesicles induced by NK-2 at higher peptide to lipid ratio is clearly seen
from the correlation curve as shown in Fig. 8. In case of LUV without
NK-2, PDI found in the distribution to be 0.04 which suggests that the
vesicles are fairly monodisperse. The correleation curve also shows
monomodal distribution which can be fitted to single exponential decay.
However, in the presence of NK-2 multimodal distribution of the size
arises due to strong membrane-membrane interaction induced by NK-2
which eventually leads to aggregation of vesicles. Polydispersity index
found in this case is pretty large (0.9) which signifies that the sample is
highly poydisperse. Signature of aggregations can also be seen from the
fluctuations in the tail of the correlation curve.

Fig. 8: Variation of average size of LUV, exposed to various concentrations of NK-2.

Inset shows the correlation curve and size distributions of LUV in the presence
and absence of NK-2. Schematic picture of LUV shown in the box indicates the
aggregation of LUV in the presence of NK-2.
134 Recent Trends in Materials: Physics and Chemistry

We shall now discuss another experimental technique, zeta potential

measurement, based on dynamic light scattering and laser Doppler
velocimetry. This technique is particularly useful and relevant for
charged colloidal particles.

2. ZETA POTENTIAL

This technique is primarily being used to study the properties of particles

in dispersions/emulsions and at the interfaces or biointerfaces[6]. It is
used to estimate the surface charge, surface potential of the colloidal
particles and to determine electrophoretic mobility, isoelectric point of
charged system. Before going to the definition and significance of zeta
potential, we shall discuss the electrostatic behavior of charge surface
in a solvent, starting from the Poisson Boltzmann equation, for
understanding the concept of zeta potential and the processes involved
and to aid in manipulating the events to best advantage. The electrostatic
behavior of the charged surface can be envisaged from the charged
colloidal dispersion. Colloidal dispersion is a heterogeneous system
formed by disperse phase and dispersion medium. Usually, one substance
is dispersed as fine particles in an another substance, known as
dispersion medium. Depending on the states of the dispersed phase and
dispersion medium, a variety of colloidal systems are formed. In order
for two phases to coexist in contact, there must be a region through
which the intensive properties, such as, mechanical, electrical, optical,
kinetic properties of the system change. For stability, the system must
possesses an interfacial free energy such that work must be done to
extend or enlarge the boundary. There exist numerous practical
situations in which the energetic balance of interfacial regions must be
modified in order to make use of the unique characteristics of a system[30].
For example, the presence of ions, excess electrons or inorganic group
in one or both states leads to a distribution of electric charges in a
nonuniform way at the interface. The stability of the system is always
determined by the charge distribution at the region between two
adjoining coexisting phases[31]. The importance of the charge distribution
at the solid-liquid interface is manifold. The electrostatics as well as
electrodynamics of charge particles in a liquid provide the basic
understanding of many natural phenomena, particularly in the field of
colloidal chemistry and electrochemistry. The knowledge of the
distribution of charges and dipoles in the interfacial region plays vital
role in understanding the mechanisms, such as, electrode kinetics,
electrocatalysis, corrosion, adsorption, crystal growth, colloid stability
and flow behaviour (both of colloidal suspensions and of electrolytes
through porous media). Zeta potential can provide useful information
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 135

to understand a number of interesting situations and events occurring

at interfaces.

A most relevant colloidal system to discuss the zeta potential is charge

particles dispersed in an aqueous or a non-aqueous solvent. One such
example is the dispersion of nano-sized polystyrene latex particles in
an aqueous solvent. Mono-disperse small or large unilamellar vesicles
made from charged phospholipids can also be an example of bio-colloids.
We shall discuss the electrostatic behavior of such vesicles in details
later in this chapter.

Fig. 9: Schematic representation of origin of zeta potential

2.1 Origin of Surface Charge and Stability Criteria

The van der Waal forces alone cannot determine the stability of particles
in an aqueous solvent. If only attractive van der Waals forces were
present, we might expect all dissolved particles to stick together
(coagulate) immediately and precipitate out of solution as a mass of
solid material. This does not happen to a stable dispersion of particles,
as particles in the solutions acquire charges, preventing from coalescing
by electrostatic repulsion. There are three different ways particles in an
aqueous solvent can acquire charges[32].
136 Recent Trends in Materials: Physics and Chemistry

(i) Ionization or dissociation of surface group. For example, the

dissociation of the acid group (or basic group) on the surface of
the particle leaves behind negatively (or positively) charged
surface.
(ii) Adsorption or binding of ions from solution onto a previously
uncharged surface. For instance, adsorption of anions (or cations)
makes the zwitterionic lipid bilayer surfaces negatively (or
positively) charge.
(iii) Differential loss of ions from the crystal lattice.

The magnitude of the final surface charges will depend on the

strength of the acidic or basic group of the surface or the amount of ions
adsorbed to the surface and the pH of the solution. Whatever the
charging mechanisms, final surface charges of the co-ions will be
balanced by oppositely charged counterions, as shown in Fig. 9. Usually
some counterions are bound or closely associated around the oppositely
charged particles and form Stern or Helmholtz layer. Other counterions
close to the surface undergo rapid thermal motion and are free to diffuse.
Stern layer together with a diffused layer constitute electrostatic double
layer. The potential at the surface of the electrostatic double layer, also
known as slipping plane, is termed as zeta potential (Fig. 9). The
magnitude of the zeta potential gives an indication of the potential
stability of the colloidal system. The binding affinity of counter ions
with oppositely charged particles can also be estimated from the
measured zeta potential. If the particles in a dispersion possess large
zeta potential (> ± 30 mV, ‘>’ is with respect to magnitude of the zeta
potential and sign indicate the type of charge of particles), the strong
repulsive forces prevent particles to aggregate, resulting in a stable
dispersion. However, low (< ± 30 mV) zeta potential indicates unstable
dispersion.

There are certain situations where particles in a dispersion may

adhere and undergo a process of sedimentation, flocculation, coagulation
and phase separation as described in Fig. 10. It is now necessary to
discuss the responsible intermolecular forces that prevent from
aggregation and keep the system stable. Four scientists Derjaguin,
Landau, Verwey, and Overbeek developed a theory, known as the DLVO
theory which discusses the stability of colloidal systems in terms of van
der Waal attraction and electrostatic double layer repulsion[32]. At very
short length scale, i.e., within the few nanometer separation, contribution
of interaction potential due to solvent is negligible and only van der
Waal (VA) and double layer repulsion (VR) are important. The VA and VR
essentially operate over a much larger distance. Dependence of these
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 137

Fig. 10: Various processes that particles in a solution can undergo due to inter-particle
interaction. These processes can be explained using DLVO theory.

�
potentials on the particle separation D is given by �� = − and
12 � �2
�� = 2��� � 2 � −�� , where, A is the Hamaker constant, ε, are permeability
of the medium and radius of particle, respectively. κ is the inverse Debye
length, which is a function of ionic composition of the system and � is
the zeta potential. DLVO theory suggests that the stability of colloids is
determined by the competition between van der Waal attraction and
double layer repulsion that exists between the particles as they approach
each other due to Brownian movement. If van der Waal attraction
dominates over the double layer repulsion, particles may adhere and
eventually undergo flocculation. However, if particles have sufficiently
high repulsive interaction to overcome the barrier of van der Waal
attraction, the dispersion becomes stable. The combine effect of these
two forces canbe understood clearly in Fig. 11.

Unlike electrostatic double layer interaction, van der Waal (VDW)

attraction is insensitive to the electrolyte concentration in the solution
and the pH. Therefore, VDW can be taken as constant over all electrolyte
concentrations. Fig. 11 shows schematically both VDW and double later
interaction potentials that exist between two similarly charged surfaces
or colloidal particles in a 1:1 electrolyte solution.

The minimum potential energy at contact (D ~ 0) is called the primary

minimum which corresponds to thermodynamically equilibrium state
138 Recent Trends in Materials: Physics and Chemistry

Fig. 11: Stability of colloidal suspension in a dispersant. Dashed curves indicate van
der Waal (VDW) attraction and double –layer repulsion. Resultant interaction
suggested by the DLVO theory is plotted as a function of charge density ó of
the particles. σ can be varied by changing the electrolyte concentration in the
solution[32].

for a colloidal system. For highly charged surfaces in dilute electrolyte

(i.e., high Σ and long Debye length), the strong long-range repulsion
occurring at short distances (between 1 and 5 nm), leads to a high energy
barrier (many kT). In more concentrated electrolyte solutions, a
significant secondary minimum, usually beyond 3 nm, appears before
the energy barrier enters into the attractive potential region. It may so
happen that the energy barrier before reaching the primary minimum
is so high during any reasonable time period that the particles prefer to
reside either at the weaker secondary minimum or remain totally disperse
in the solution. In the latter case, the colloid is referred to as being
kinetically stable. The small energy barrier can be envisaged for low
charge density or potential which leads to slow aggregation, known as
coagulation or flocculation. However, rapid coagulation starts to happen
when the energy barrier falls in the negative potential energy (Attractive
regions, see Fig. 11). This happens below a certain charge or potential,
or above some concentration of electrolyte, known as the critical
coagulation concentration. The colloid is now referred to as being
unstable. At sufficiently low surface charge or potential the interaction
curve approaches the pure attractive type and two surfaces now attract
each other strongly at all separations. In summary, if the electrical
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 139

double layer of a colloidal system overlap, i.e., if κ-1 exceeds the average
interparticle separation, then the system will usually be stable, given a
moderate amount of charge on the particles.

This is clearly evident from the above discussions, the surface charge,
potential as well as added electrolyte concentration play a crucial role
in determining the stability of the colloidal dispersion. The electrostatics
of particles in an aqueous solution and their stability can be obtained
directly from the measurement of zeta potential which is a good
approximation to a surface potential at moderate electrolyte
concentration. We shall now discuss a brief theory of potential and charge
distribution in the electrostatic double layer.

2.2 The Potential and Charge Distribution in Electrical Double Layer

The theory of diffused electrical double layer was first described

independently by Gouy (1910) and Chapman (1913). In this model, a
layer of charge is assumed to be smeared out uniformly over a plane
surface immersed in an electrolyte solution. A practical example of such
system is the charged membrane in an electrolyte solution. This charged
surface will have electrostatic potential, called surface potential �0 . The
compensating ions are considered as point charges embedded in a
continuum dielectric medium. The charge distribution and the potential
ø in this system are described by the Poisson equation.
�
∇2 � = − (15)
�
Where, ε(Dielectric constant × permittivity of the free space) is the
permittivity of the medium. In equilibrium, the electrical forces and the
diffusive forces of the ions must balance out, i.e., the electrochemical
potential must be constant everywhere.

∇�� = − �� �∇� (16)

The term in the left hand side of equation (16) describes the diffusive
force of the ith type of ions with valency zi and chemical potential µi .
The right hand side is typical electrical force, as electric field (conservative
field) can be written as negative gradient of potential. For a flat electrical
double layer, both the potentials are constant along the planes parallel
to the wall (i.e., Ψ(y,z)= constant, considering the double layer wall is
along the Y-Z plane). So equation (16) can be written in terms of one
dimensional Poisson equation.
140 Recent Trends in Materials: Physics and Chemistry

�� � ��
��
= − �� � ��
(17)

Now the chemical potential per ions of ith type is given by

�� = �0 + �� �� �� (18)

where ni is the number of ions of ith type per unit volume. From eqs.
(17 and 18) we get
��� � � 1 �� � 1 ��
��
=
� � ��
=− ��
�� � ��
(19)

Solution of equation (19) with a boundary condition that � = 0 at

bulk, i.e., �� = ��0 . Where ��0 is number density of ions of a particular
type i at bulk. Therefore, Boltzmann distribution of each type of ion is
given by

�� = ��0 exp (−�� ��/��) (20)

Equation (20) gives the local concentration each type of ion in the
vicinity of the double layer region, provided the correct sign of zi is
taken in the equation. For positive ions �� = +�� and �� < ��0 and for
negative ion �� > ��0 . Volume charge density near the charged surface
is given by

� = � �� �� � (21)
�

where the summation is over all species of ions present in the solution.
Using equation (20) in equation (21) we get
� = � �� �� � = � ��0 �� � exp (−�� ��/��)
. Now equation (15) in one dimension
� �
be written as

� 2 �(�) � 1
= − = − � ��0 �� � exp (−�� ��/��) (22)
�� 2 � �
�

Equation (22) is a non linear second order differential equation and

can be solved analytically as well as numerically. Let us now examine
an approximate solution for the electrical energy associated with double
layer which is much smaller than thermal energy, i.e., �� �� ≪ ��. This
approximation is known as Debye-Hückel approximation. Therefore,
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 141

the argument of the exponential term is small and can be expanded

using Taylor series (� −� ≈ 1 − �) keeping only the linear term.

� 2 �(�) 1
= � (��0 �� � + �� 2 � 2 ��0 �/��) (23)
�� 2 � �

The 1st term in the right hand side of the equation (22) is zero, in
order to preserve the electroneutrality in the bulk. Therefore, equation
(22) can be written as

� 2 �(�)
= �2� (24)
�� 2

�2 2 2 � 1/2
Where � 2 = � �� 2 ��� . The parameter � = �� ∑� � � � � � , known
��� � ���
as Debye-Hückel parameter or inverse Debye length, is a function of
electrolyte concentration and has a great significance in colloid science.
To solve equation (24) to obtain �(�) as a function of x, multiply equation
��
(24) by
by 2 ��
and integrating the entire equation.

�� � 2 �(�) �� 2
�2 2
�� = � 2 � �
�� �� ��
� �� 2
Or � �� � �� � �� = � 2� 2 ���(for � = ∞, � = 0 ) (25)
��
Or,
��
= − �� (26)

The negative sign is chosen when Ψ is positive and decreases to zero

at the bulk solution. The solution of the equation (26) gives

� (�) = �0 exp (- κx) (27)

The exponential decay, as given in equation 27, is not applicable to

many colloidal systems near the surface of the electrical double layer.
Therefore, we need to seek for a general solution of equation (22).

�� 2 2 �� ��
�� � � �� = − � � ��0 �� � exp �− � ��
�� � ��
�

�� 2 2�� � � ��
or,� � = ∑ ��0 [exp �− � − 1] (28)
�� � ��
142 Recent Trends in Materials: Physics and Chemistry

As the charge distribution is symmetric for most electrolytes, the number

of ions is exactly same as the number of counter ions which adsorb in the
form of electrical double layer. Equation (22) turns out to be

� 2 �(�) 2�0 ��
= sinh(���/��) (29)
�� 2 �
0 0
Where �0 = �+ = �− and �+ = − �−

�� 2 4�0 �� ���
� � = (cosh − 1)
�� � ��
��
= −(8�0 ��/�)1/2 sinh ���/�� (30)
��
�� 2���
or = − sinh ���/2�� (31)
�� ��

Again the negative sign indicates that falls to zero at a distance

far from the wall. Solution of equation (30) can be obtained by integrating
the equation and expressed in the following form

��� ���0
tanh = tanh exp (−��) (32)
4�� 4��

For small values of p, sinh � ≈ tanh � ≈ �. In this case equation 31

reduces to 27. The approximate formula is always an overestimate of
���
the potential. Far from the electrical double layer one assumes � (= 4��
)
is small and tanh � ≈ � and equation (32) reduces to

��� ���0
= tanh exp(−��)
4�� 4�� (33)
4��
�= ��
� exp (−��) (34)
��� 0
�� � 0 exp � �−1
2��
here, � = tanh 4��
= ��� 0 (35)
exp � �+1
2��

���0
The approaches unity for high values of . One expects the
��
4��
potential far from the surface to become provided the surface
��
potential is sufficiently high. Typical value of such potential ~ 100 mV/
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 143

z. The equations (27), (32) and (33) describe the behavior of electrostatic
potential from the wall of the charged surface. The effect of electrolyte
concentration appears through the parameter κ which is the measure
of thickness of electrical double layer. Putting values of all parameters
sin expression of κ, it can be written as � = 3.288����� nm-1. ���� is the
molar concentration of ions. κ Increases with increasing concentration
of ions (see Fig. 12). As a result the potential from the charged surface

falls rapidly as shown in Fig. 12. As measures the thickness of the

double layer, the compression of double layer occurs as is increased.
The decrease in LUV radius with increasing NaCl concentration, as
presented in Fig. 7 might be the consequence of the compression of
electrical double layer. We shall discuss the zeta potential study of the
LUV later.

Fig. 12: Variation of potential obtained from the Gouy Chapman theory.

2.3 Surface Charge Density

The surface charge density (σ 0), i.e., amount of charge per unit area
can be determined from the following equation

∞ ∞ �2� ��
�0 = �0 ��� = � �0 �� 2
= -� � �� � at x = 0 (36)

For symmetrical electrolyte solution, equation 35 can be written using

equation 27

1/2
�� ��
�0 = �2��� � ��0 [exp �− � − 1]�
��
144 Recent Trends in Materials: Physics and Chemistry

However, for symmetrical electrolyte

4� 0 ��
�0 = �
sinh ���/2�� = 11.74����� sinh(19.46��0 ) µC cm-2 . Here �0 is
in volt and ���� is mole/litre. For very small potential, we can use equation
(26) to obtain σ0 as

�0 = ���0 (37)

From equation (36) it is evident that this system behaves like a

1
parallel plate capacitance with separation of between the two plates.
�

When charged particles move through the liquid under the action of
an electric field, three distinct electrokinetic effects depending on the
way the motion is induced come into the picture. These effects are
electrophoresis, electroosmosis and streaming potential. Zeta potential
is measured from the electrophoresis experiment. Before describing the
experiments, let us discuss the phenomenon of electrophoresis and the
working formula to determine zeta potential. Later we shall also describe
how electroosmosis affects the measurement of electrophoresis.

2.4. Electrophoresis

The movement of charged particles suspended in solution towards the

oppositely charged electrode when an electric field is applied across an
electrolyte solution is known as electrophoresis. The force acting on the
particles due to the application of electric field is known as the
electrophoretic force. The velocity of a particle in a unit electric field is
referred to as its electrophoretic mobility µ. Zeta potential is related to
the electrophoretic mobility by the Henry equation (32).

3��
�= (38)
2��(��)

where η and ε are coefficient of viscosity and permittivity of the

aqueous medium, respectively. The Henry function, �(��) depends on
the inverse Debye length (κ) and the radius of the particle a. Now, κ can
be defined as

2� 2 ����
�= � (39)
���
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 145

where, e is the electronic charge, k is the Boltzmann constant, and T

is the temperature. ���� is the concentration of electrolyte in the solution.
�(��) was estimated using the following equation.

1 for �� < 1
1
�(��) = � ln(��) + 1 for 1 < �� < 1000 (40)
6
1.5 for �� > 1000

The Debye length, ( � −1 ) is often taken as a measure of the “thickness”

of the electrical double layer (Fig. 13a). Typical variation of Debye length
and Henry function for a particle of diameter ~ 100 nm dispersed
in a various salt concentration, estimated from equations (39 and 40),
are shown in Fig. 13b. The argument of the Henry function, κa, measures
the ratio of the particle radius to electrical double layer thickness. For
moderate electrolyte concentration, electrophoretic mobility and zeta
potential are most commonly measured in aqueous media and with
which is known as the Smoluchowski approximation. For
small particles in low dielectric constant media f(κa) becomes 1.0 which
is referred to as the Hückel approximation. Particles in a non-aqueous
solvent generally use Hückel approximation. The corrected zeta
potential is obtained by putting the modified Henry function into the
eq. (38). Effective charge, Qeff, of the particle can also be estimated from
the measured electrophoretic mobility according to equation (41), where
aeff is the effective radius of the particle, including stern layer and can
be obtained from DLS.

����
�= (41)
6������

2.5. Determination of Electrophoretic Mobility from Dynamic Light

Scattering

The electrophoretic mobility of the particle is measured from the

scattering of light from particles that move in liquid under the influence
of an applied electric field. The basic technique is known as Laser Doppler
velocimetry (LDV) or electrophoretic light scattering (ELS). The charged
particles moving under the influence of the electric field E will attain
the terminal velocity v which is proportional to E.

� = �� (42)
146 Recent Trends in Materials: Physics and Chemistry

Figs. 13: Schematic picture of electrostatic double layer (a) Variation of Debye length
and Henry function �(��) for a particle of radius ~100 nm dispersed in a
various salt concentration (b)[26].

When light of frequency (νi) and wavelength λ is incident on a moving

particle in an applied electric field, the scattered light will undergo a
Doppler shift in frequency, ∆ν, as depicted in Fig. 14. The amount of
Doppler shift depends on the velocity of the particle, the wavelength of
the light in the medium and the scattering angle θ and can be expressed
as 2π∆ν = q.ν, where q is the scattering vector defined in Fig. 2.

Now,

�. � 4� n sin � � n sin 2�
Δ� = = � cos � = (43)
2� 2�� �

Where n is the refractive index of the liquid medium. From equations

(42 and 43) we obtain

� λΔ�
�= = (44)
� n E sin 2�
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 147

Fig. 14: Schematic diagram of scattering by moving particles under the influence of
electric field showing the frequency shift in scattered light.

Equation (43) shows that ∆ν is directly proportional to the particle

velocity ν. Therefore, the determination of electrophoretic mobility µ
requires the accurate measurement of ∆ν. Finally, the value of the mean
zeta potential, ζ, is obtained from the µ using equation (38). The factor
3/2 appears in the expression due to Hückel limit. In this limit �� ≪ 1
and Debye length overlaps with the interparticle separation. Now we
are ready to discuss detailed experimental method to obtain µ.

2.6. Measurement of Electrophoretic Mobility

A typical experimental geometry to measure µ in a scattering experiment

is shown in Fig. 15. A classical micro-electrophoresis system consists of
a cell with electrodes at either end to which a potential is applied.
Particles move towards the electrode of opposite charge, their velocity is
measured and expressed in unit field strength as their mobility. A laser
is used to illuminate the sample after it has passed through a beam
splitter. The beam splitter produces a reference beam which will later
interfere with the scattered beam. The velocity of the particles undergoing
electrophoresis is determined by comparing the beat frequency detected
with that of a reference frequency. The scattered beam is usually
detected at a small angle typically 10 to 15º. The resulting coherent
superposition between reference beam and the scattered beam is
transmitted to the detector (Usually photo detector, such as PMT or
APD) by the optical fiber. In order to enhance the Doppler shift the
reference beam is also modulated with an oscillating mirror before it
combines with the scattered beam to improve signal to noise ratio. The
mobility of the particles in an applied field will therefore produce a
148 Recent Trends in Materials: Physics and Chemistry

frequency shift away from that of the modulator frequency. All that is
required to determine ∆µ is to measure the frequency power spectrum
of the detector output signal. Detector sends the information to a digital
signal processor. Proper design and synchronization of the electronic
waveforms that drive the cell electrode voltage and optical modulator
allow the sign, as well as the magnitude, of the frequency shift to be
determined.

Fig. 15: Schematic experimental geometry for the measurement of zeta potential

The electroosmosis within the sample capillary cell will affect the
electrophoresis process. Electroosmosis is the movement of a liquid
relative to stationary charged surface. In order to avoid electroosmosis,
phase analysis light scattering (PALS) technique is adopted. PALS is a
technique that uses the same optical setup as conventional laser Doppler
electrophoresis. However, a different signal processing method is
employed. PALS measures the phase (frequency × time) shift between
scattered light and reference beam instead of frequency shift. The
detection of phase change is more sensitive to changes in mobility than
the traditional detection of frequency shift. Details of this technique
and its consequence in the context of electrophoresis measurement are
discussed elsewhere[15,33]. Here we discuss briefly the PALS technique to
measure phase shift.

The phase shift in the scattered light is proportional to the change in

the position of the particles. A reasonable contribution of electrophoretic
mobility to the particle motion can be obtained by sampling many
oscillations of frequency shifted signals (Fig. 16a). For high mobility
particles, enough data can be produced within the measurement time
for accurate determination of particle motion. This is true as high mobility
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 149

particles pick up a lot of energy from the electric field and move quickly.
However, the effect of electric field on low mobility particles is too weak
to produce enough data and often a fraction of oscillation can be obtained
within the measurement time (Fig. 16b). In such case, electrophoretic
mobility would not be possible to determine from standard Fourier
transform analysis. Therefore, the strength of the applied electric field
to the sample needs to be increased in order to compensate the electric
field. However, the increase in the strength of the applied electric field,
especially for high conductivity sample, leads to significant Joule heating
which affects the random motion of particles and can denature the
sample. Now a small change in frequency leads to a considerable large
change in phase between beat frequency and the reference frequency.
By measuring the phase differences, the changes of the position of the
particle can be determined more accurately. When the phase difference
between two signals (Figs. 16c-d) of different frequencies is plotted over
time (Fig. 16e), a gradient line is obtained. The PLAS measures the slop

Fig. 16: (a) For high mobility particles, enough data is collected within the
measurement time to produce an accurate determination of the particle
motion; (b) For low mobility particles, only a fraction of an oscillation may
have been completed within the measurement time. Schematic diagram
showing the phase difference between two signalsl; (c–d) of two different
frequencies ν and ν + ∆ν. Plot of phase difference between two signals ,
depicted in (c–d) with time shows straight line; (e) Slop or gradient of the line
determines the velocity of particles and the electrophoretic mobility; (f) is the
typical phase plot with time which is obtained due several times field reversal
during the measurement sequence.
150 Recent Trends in Materials: Physics and Chemistry

of the line, i.e., the rate of change of phase difference which is related to
the velocity of the particles. By measuring the velocity of the particles,
the zeta potential can be determined. Fig. 16f shows typical phase plot
that is obtained from the zeta potential measurement. The well-defined
alternating slopes of the phase difference with time that, are obtained
due to number of times rapid reversal of applied filed during the
measurement sequence. These slopes are averaged to determine the
mean phase difference and hence mean zeta potential.

There are several major advantages of PALS techniques over

standard measurement techniques. PLAS allows to measure high
conductivity sample at moderately low electric field to avoid the risk of
sample effect such as joule heating. Besides this, the low electric field
also prevents the electrolysis of the sample which might damage the
electrodes of the sample cell. It is also useful for low mobility particles
i.e., high viscosity and non aqueous applications. In a typical experiment
each measurement of zeta potential consists of several runs to minimize
the statistical errors. The typical output from the instrument (Malvern
zetasizer nano ZS) is shown as Fig. 17. In addition to the distribution

Fig. 17: Typical electrophoretic mobility and zeta potential distributions obtained from
a standard sample.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 151

profile, measurement will also provide data for frequency shift and phase
plot of the signal.

2.7. Factors Affecting Zeta Potential

The most important factor that affects zeta potential is pH. As pH of the
solution changes the effective charge state of the particles, zeta potential
value also changes. Therefore, zeta potential value without mentioning
the pH is meaningless. Fig. 18 shows the zeta potential at different pH.
The variation zeta potential with pH also enables us to determine the
isoelectric point (IEP) as indicated in Fig. 18. It is the point at which the
particles do not possess net charges. For example, if more alkali is added
to the particle suspension with negative zeta potential, then the particles
tend to acquire a more negative charge. Alternatively, if acid is added
to this suspension, the zeta potential will approach towards positive
value and eventually a point will be reached where the charge will be
neutralized. This point is called the IEP. Further addition of acid will
cause a buildup of positive charge, as depicted in the positive zeta
potential in Fig. 18. The IEP is indeed an important practical
consideration for many applications, especially in the field of colloid
science and biochemistry. For example, IEP is used for detection and
prevention of protein aggregation before, during, and after
purification[34,35]. It is normally the point where the colloidal system is
least stable[36].

Fig. 18: Dependence of zeta potential on pH of the solution. The point indicates by
IEP is the isoelectric point[37].

As discussed before (Fig. 13), the thickness of the double layer (κ–1)
can be estimated from the ionic strength of the medium. The double
layer becomes more compressed, as seen from decreasing the value of
κ–1 at higher ionic strength. The extent of decrease in κ–1 also depends
152 Recent Trends in Materials: Physics and Chemistry

upon the valency of the ions. Higher valency, such Ca+2 or Al+3 will
compress the double layer to a greater extent than that of monovalent
ions, such as Na+. The specific adsorption of ions onto a particle surface,
even at low concentrations, can have a dramatic effect on the zeta
potential of the particle dispersion. We shall discuss later this aspect
with an example of peptide adsorption onto negatively charged vesicles.
High ionic strength typically more than 500 mM concentrations of ions
affects zeta potential significantly. High ionic conductivity is not suitable
for the measurement which may lead to electrolysis of the electrode.
Therefore, at high ionic strength, the electrode may get damaged. The
best result we could get is for moderate electrolyte concentration[25]. Now
we shall discuss two specific examples for the measurement of zeta
potential. These systems are already discussed in the context of size
measurement using dynamic light scattering.

2.8. Zeta Potential of Unilamellar Vesicles Prepared from Charged

Phospholipids

The effect of various types of ions on model membranes has enticed

significant interest due to their immense biological applications[38]. For
example, trans-membrane potential may arise from the asymmetric
binding of Na+ and Cl- ions which in turn affect ion transport across the
bilayer[39]. In 1888, Hofmeister proposed a series which qualitatively
arranged ions according to their propensity to precipitate proteins[40].
The presence of ions in the aqueous medium plays a vital role in
modulating bilayer properties. Electrostatics of the ion-membrane
interaction can be best described by measuring zeta potential[41]. As
discussed before, zeta potential is a potential at the electrical double
layer and can be used to determine surface charge distributions using
Gouy Chapman theory. Therefore, the affinity of ions towards the
charged membrane and hence the binding constant of ions is
determined[25,26,42]. The Fig. 19 shows the result of zeta potential of LUV
composed of negatively charged lipids (already mentioned before) for
various alkali metal ions. The affinity of these ions with the negatively
charged membranes are inferred from the zeta potential and the order
of affinity is presented at the top of the figure 19. The formation of
electrical double layer around the LUV is also depicted in the Fig. 19.

As evidenced from Fig. 19, Li+ and Na+ ions show stronger adsorption
affinity relative to other alkali metal ions. In summary, the affinity of
ion adsorption to negatively charged membranes follows the sequence
Li+> Na+> K+> Cs+>Rb+ which is consistent with the Hofmeister series.
As expected, zeta potential increases with increasing ion concentration.
Measurement of zeta potential from electrophoretic mobility also becomes
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 153

Fig. 19: Effect of various monovalent alkali metal ions with negatively charged
phospholipid vesicles composed of DOPC-DOPG (4:1).

difficult at high salt concentration, as this may damage the gold

electrodes of the zeta cuvette. The behaviour of zeta potential for three
different lipid compositions is shown in Fig. 20. LUV, made from a neutral
lipid (DOPC), show zeta value close to zero, indicating very low ion
adsorption affinity. It is expected for neutral membranes as the zeta
potential can only be realized with respect to the charged membranes.
However, the zeta potential increases significantly in the case of
negatively charged LUV, indicating higher adsorption affinity of cations
towards negatively charged lipid than of that neutral lipid. We shall
now briefly describe how Gouy Chapman theory is used to determine
the binding constant of alkali metal ions in 1:1 electrolyte.

Fig. 20: Variation of zeta potential with different concentration of NaCl for three
different phospholipid vesicles, indicated in the figure[25].
154 Recent Trends in Materials: Physics and Chemistry

2.9. Electrostatics of Charged Membranes

The binding constant essentially measures the relative ratio of number

of ions that bind to the membranes to the number of free ions present in
the bulk. Total ion concentration in the solution is the sum of bound
ions and free ions. Therefore, the apparent binding constant is usually
obtained in many experimental techniques, such as isothermal titration
calorimetry and fluorescence spectroscopy techniques. The apparent
binding constant Kapp is defined as �� = ���� �� , where Xb is the extent of
ion binding per mole of lipid and Cf is the free ion concentration of the
solution. When the ions adsorb, the membrane becomes charged. Hence,
further adsorption of ions onto the membrane is restricted by electrostatic
repulsion. Therefore, Kapp is no longer a constant, but rather changes
with Cf. Therefore, the most relevant binding parameter would be the
intrinsic binding constant K int which is assumed to be directly
proportional to the concentration of ions (CM ) at the vicinity of the
membrane surface. The binding model essentially calculates Xb and CM
using Gouy Chapman theory as described before. The surface
concentration CM of monovalent cations at the vicinity of the membrane
solution interface, is described by the Boltzmann relation (also see
equation 20).

−���0
�� = �� exp � � (44)
��

Where CB is the bulk concentration of the cations. The surface charge

density σ is estimated from the Gouy Chapman theory. The theory
assumes that charges are located on the surface of the membranes and
are smeared uniformly over the entire surface. The zeta potential (ζ) is
defined as the potential at the shear plane or diffused layer. The
hydrodynamic plane of the shear is assumed to be located at a distance
x =2 Å from the charged surface of the membranes [42]. From the
knowledge of zeta potential, i.e., ψ(x= 2 Å)=ζ, the surface potential ψ0
was calculated. The adsorption of monovalent ions onto the membrane
alters the surface charge density (σ) and σ for each electrolyte
concentration can be obtained from the Gouy–Chapman relation
discussed the section 2.3.

Langmuir adsorption isotherm describes the simplest approach to the

binding of cations with the charged lipids, which relates to the surface
charge density with Kint from the knowledge that the maximum charge
density of the membrane in the absence of ions σ0 (~1.4 × 1014 charges
per cm2 "0.228 C/m2)[43] and CM.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 155

�0
�= (45)
1 + ���� ��

Therefore, using equation (45), the intrinsic binding constant can

be estimated. Results from the Gouy Chapman theory is presented in
Table 2.
Table 2: Binding parameters obtained from the zeta potential for a 100 mM salt
concentration[25].
Alkali Zeta Surface Surface Surface Intrinsic
metal potential potential charge concentra- binding
ions (mV) ψ0 σ tion CM constant
mV) (C/m2) (mM) Kint(M-1)
Li+ –29 ± 2 –36 ± 2 –0.028 ± 0.002 410 ± 40 1.4 ± 0.9
Na+ –34 ± 3 –43 ± 4 –0.034 ± 0.003 520 ± 80 0.6 ± 0.2
K+ –37 ± 1 –47 ± 2 –0.039 ± 0.002 620 ± 40 0.3 ± 0.1
Cs+ –37 ± 2 –47 ± 1 –0.039 ± 0.001 620 ± 10 0.3 ± 0.1
Rb+ 40 ± 1 –51 ± 1 –0.043 ± 0.001 720 ± 10 0.1 ± 0.1

The ideal concentration range that one should use to study the binding
phenomena of ions with the membranes is 10-100 mM. It is interesting
to note that even at moderate concentration of salts, the intrinsic binding
constant decreases with increasing salt concentrations[25]. The surface
potential as well as zeta potential are determined by the competition
between the binding of ions and charge screening effect. As the Debye
screening length decreases very rapidly with concentration, charge
screening effect increases with increasing concentration of ions and
eventually overcomes the binding of cations. This might be the reason
why higher concentrations of salt (> 200 mM) are not suitable for analysis
of binding phenomena from the measurement of zeta potential.

2.10 Interaction of Antimicrobial Peptide NK-2 with Various

Phospholipid Membranes: Zeta Potential Study

The interaction of NK-2 with negatively charged membranes has already

been illustrated in terms of size distribution obtained from DLS (see
section 1.6). Zeta potential was measured in LUV composed of different
lipids or lipid mixtures for various concentrations of NK-2. Results are
presented in Fig. 21[29]. It is clearly evident from Fig. 21 that the zeta
potential increases significantly with increasing NK-2 to lipid molar
ratio (NK-2/L) for all charged LUV. Although phospholipids such as
DOPC and a mixture of DOPC-DOPE are neutral, they exhibit low
negative value and increase to a positive zeta potential due to adsorption
of peptide onto the membrane surface. As NK-2 possesses high (+10 in
156 Recent Trends in Materials: Physics and Chemistry

pH 7.4) positive charge, the charge compensation occurs at different

NK-2/L for different lipid mixtures. The overcharge compensation occurs
at a different NK-2/Lipids ratio for all different lipid compositions which
indicates the entropic contribution besides the electrostatic interaction
between charged membrane and the peptide. The modified Gouy
Chapman equation can be used to estimate the intrinsic binding constant
of NK-2 with phospholipid membranes.

Fig. 21: Zeta potential of phospholipid LUV at various NK-2 to Lipids molar ratio.
Solid lines are intended only as guides to the points. Error bars indicate the
width of the distribution profile of zeta potential[29]

It is important to mention that change in the size distribution of

vesicles does not influence the zeta potential significantly. Zeta potential
is estimated from the Smoluchowski approximation, where the Henry
function f(κa) takes its maximum value 1.5. In other words, zeta potential
is determined from the electrostatic double layer formation which does
not depend on the size of the particle.

Apart from size distribution and zeta potential measurements, DLS

can also be used to measure rheological properties of a fluid by measuring
the scattered intensity from tracer particles embedded in the solution.
Dynamic light scattering microrheology is a passive characterization
technique that uses tracer particle and probes linear viscoelastic
properties, such as viscosity and viscoelastic moduli of low viscosity and
weakly structured fluids.

To conclude, we discuss two important non-invasive experimental

characterizations namely particle size distribution and zeta potential
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 157

based on dynamic light scattering along with laser Doppler velocimetry.

These characterizations are widely and routinely used in many
interdisciplinary fields of science including nanotechnology. In order to
interpret and explain the experimental results obtained from DLS and
zeta potential, it is necessary to understand the techniques as well as
theory behind. In the present chapter, we intend to provide some details
of DLS techniques and underlying theory.

ACKNOWLEDGMENTS

The author sincerely acknowledges the Department of Biotechnology

(DBT), Govt. of India (BT/PR8475/BRB/10/1248/2013).

BIBLIOGRAPHY

[1] Berne, B.J. and Pecora. R. (2000). Dynamic Light Scattering: With Applications
to Chemistry, Biology, and Physics. New York: Dover Publications. Inc.
[2] Tyndall, J. (1868). On the blue colour of the sky, the polarization of skylight,
and on the polarization of light by cloudy matter generally. Proc. Royal Soc.
Lond, 17: 223–233.
[3] Hergert, W. and Wriedt, T. ( eds. ) (2012). The Mie Theory: Baiscs and
Applications. Springer, 2012.
[4] Yeh, C. and Lee, W.C. (1978). Rayleigh-Debye scattering with focused laser
beams. Appl. Optics, 17: 1637–1643.
[5] Pecora, R. (ed. ) (2011). Dynamic light scattering: Applications of photon
correlation spectroscopy. Spinger.
[6] Hunter, R. (1981). Zeta Potential in Colloids Science: Principles and
Applications. Ottewill, R.L. and Rowell, R.H. (ed.), New York: Academic Press.
[7] Guinier, A. X-Ray Diffraction In Crystal, Imperfect Crystals, and Amorphous
Bodies. W.H. Freeman and Company, San Francisco and London.
[8] Bhattacharjee, S. (2016). DLS and zeta potential – What they are and what
they are not? J. Control. Release, 235: 337–351.
[9] Stetefeld, J., McKenna, S.A. and Patel, T.R. (2016). Dynamic light scattering: a
practical guide and applications in biomedical sciences. Biophys. Rev., 8: 409–
427.
[10] Smoluchowski, M. (1906) . Essai d’une théorie cinétique du mouvement
Brownien et des milieux troubles (Outline of the kinetic theory of Brownian
motion of suspensions. Bulletin International de l’Académie des Sciences de
Cracovie. pp. 577–602.
[11] Einstein, A. (1906). Zur Theorie der Brownschen Bewegung. Annalen Der Physik,
324: 371–381.
[12] Beale, P.D. and Pathria, R.K. (2011). Statistical Mechnics. 3rd ed., Elsevier
Ltd.
[13] https://www.lsinstruments.ch/technology/dynamic_light_scattering_dls/.
[14] http://www.horiba.com/scientific/products/particle-characterization/technology/
dynamic-light-scattering/.
[15] https://www.malvern.com/en/products/technology/dynamic-light-scattering.
158 Recent Trends in Materials: Physics and Chemistry

[16] https://www.brookhaveninstruments.com/nanobrook-90plus.
[17] http://pssnicomp.com/.
[18] Harding, S.E. and Jumel, K. (1998). Light Scattering. In: Dunn, B.M., Ploegh,
H.L., Speicher, D.W., Wingfield, P.T. and Coligan, J.E. (eds.), New York: John
Wiley & Sons, Inc.
[19] Pecora, R. (1964). Doppler Shifts in Light Scattering from Pure Liquids and
Polymer Solutions. J. Chem. Phys., 40: 1604.
[20] Koppel, D.E. (1972). Analysis of macromolecular polydispersity in intensity
correlation spectroscopy: The method of cumulants. J. Chem. Phys., 57: 4814.
[21] Morrison, I.D., Grabowski, E.F. and Herb, C.A. (1985). Improved techniques for
particle size determination by quasi-elastic light scattering. Langmuir, 1: 496–
501.
[22] Provencher, S.W. and Stepanek, P. (1996). Global analysis of dynamic light
scattering autocorrelation functions. Part. Syst. Char., 13: 291–294.
[23] https://www.agilent.com/cs/library/technicaloverviews/Public/5990-
7890EN.pdf.
[24] Rogošiæ, M., Mencer, H.J. and Gomzi, Z. (1996). Polydispersity index and
molecular weight distributions of polymers. Euro. Polym. J., 32: 1337–1344.
[25] Maity, P., Saha, B., Kumar, G.S. and Karmakar, S. (2016). Binding of
monovalent ions with negatively charged phospholipid membranes. Biophys.
Biochim. Acta, 1858: 706–714.
[26] Maity, P., Saha, B., Kumar, G.S. and Karmakar, S. (2016). Effect of counterions
on the binding affinity of Na+ ions with phospholipid membranes. RSC Adv., 6:
83916–839925.
[27] Hope, M.J.; Bally, M.B.; Webb, G.; Cullis, P.R. (1985). Production of large
unilamellar vesicles by arapid extrusion procedure: Characterization of size
distribution, trapped volume and ability to maintain a membrane potential.
Biophys. Biochim. Acta, 812: 55–65.
[28] Klasczyk, B. and Knecht, V. (2011). Validating affinities for Ion–lipid
association from simulation against experiments. J. Phys. Chem. A, 115:
10587–10595.
[29] Karmakar, S., Maity, P. and Halder, A. (2017). Charge-driven interaction of
antimicrobial peptide NK 2 with phospholipid membranes. ACS Omega, 2:
8859–8867.
[30] Myers, D. (1999). Surfaces, Interfaces and Colloids: Principles and applications.
John Wiley & Sons, Inc.
[31] Lebovka, N.I. (2014). Aggregation of charged colloidal particles. Adv. Polym.
Sci., 255: 57–96.
[32] Israelachvili, J.N. (2011). Intermolecular and surface forces. Elsevier and
Academic Press, 3rd Ed.
[33] Corbett, J.C.W., McNeil-Watson, F., Jack, R.O. and Howarth, M. (2012).
Measuring surface zeta potential using phase analysis light scattering in a
simple dip cell arrangement. Colloids Surf: Physicochemical and Engineering
Aspects, 396: 169–176.
[34] Bondos, S.E. and Bicknell, A. (2003). Detection and prevention of protein
aggregation before, during, and after purification. An. Biochem., 306: 223–231.
[35] Widmann, M., Trodler, P. and Pleiss, J. (2010). The isoelectric region of proteins:
A systematic analysis. PLOS One, 5: e10546.
[36] Singh, B.P., Menchavez, R., Takai, C., Fuji, M. and Takahashi, M. (2005).
Stability of dispersions of colloidal alumina particles in aqueous suspensions.
J. Colloid. Interface Sci., 291: 181–186.
Particle Size Distribution and Zeta Potential Based on Dynamic Light... 159

[37] http://pssnicomp.com/applications/isoelectric-point-iep/isoelectric-point-test/.
[38] Vorobyov, I., Olson, T.E., Kim, J.H., Koeppe, R.E., Andersen, O.S. and Allen,
T.W. (2014). On–induced defect permeation of lipid membranes. Biophys. J.,
106: 586–597.
[39] Lee, S.-J., Song, Y. and Baker, N.A. (2008). Molecular dynamic simulations of
asymmetric NaCl and KCl solutions separated by phosphatidylcholine bilayer:
Potential drops and structural changes induced by strong Na + lipid interactions
and finite size effects. Biophys. J., 94: 3565–3576.
[40] Kunz, W.; Henle, J. and Ninham, B.W. (2004). Zur Lehre von der Wirkung der
Salze (About the science effect of salt) Franz Hofmeister’s historical papers.
Curr. Opin. Colloid Interface Sci., 9: 19–37.
[41] Mclaughlin, S. (1989). Electrostatic properties of membranes. Annu. Rev.
Biophys. Biophys. Chem., 18: 113–136.
[42] Eisenberg, M., Gresafi, T., Riccio, T. and McLaughlin, S. (1979). Adsorption of
monovalent cations to bilayer membranes containing negatively phospholipids.
Biochemistry, 18: 5313.
[43] Elsa, Y.L., Yan, C.Y., Zhao, X. and Eisenthal, K.B. (2001). Surface potential of
charged liposomes determined by second harmonic generation. Langmuir, 17:
2038–2066.