-
DEPARTMENT OF CHEMISTRY & APPLIED BIOSCIENCES INSTITUTE FOR
CHEMICAL & BIOENGINEERING Prof. M. MORBIDELLI GROUP
STUDY OF THE EMULSION AND THE MINIEMULSION
POLYMERIZATION OF STYRENE IN THE PRESENCE OF A CHAIN TRANSFER
AGENT
SUPERVISOR: ALEXANDROS LAMPROU
ZURICH 2006
-
1
Cover page: SEM picture of hybrid polystyrene-silica
nanoparticles prepared by the miniemulsion
polymerization technique (photo taken at the Chemical Process
Engineering Research Institute in
Thessaloniki,Greece)
-
Table of Contents
i
TABLE OF CONTENTS Page
Introduction................................................................................................
1
1. Theoretical
part..........................................................................................
1
1.1. Molecular solutions, colloidal systems and
suspensions......................... 1
1.2. Polymer latexes and the concept of heterogeneous
polymerization...... 2
1.3. Emulsion Polymerization.. 3
1.3.1. Concerning the emulsion polymerization in general...
3
1.3.2. Surfactants.... 4
1.3.3. Qualitative mechanistic picture of emulsion
polymerization.. 4
1.4. Miniemulsion Polymerization... 6
1.4.1. Definition of a miniemulsion... 6
1.4.2. Coalescence by collisions and the role of surfactant...
7
1.4.3. Ostwald ripening and the role of the ultrahydrophobe
7
1.4.4. Preparation of a miniemulsion. 8
1.4.5. The critical stability of miniemulsions 9
1.4.6. Mechanistics in miniemulsion polymerization 10
1.5. Chain transfer 11
1.5.1. Main concept 11
1.5.2. Kinetics.... 11
1.6. Chain Length Distribution of a polymeric mixture 13
1.6.1. Average Molecular Weight.. 13
1.6.2. Average Degree of Polymerization.. 13
1.6.3. Polydispersity Index. 14
1.6.4. Instantaneous Properties.. 14
2. Devices and Characterization
Techniques............................................... 15
2.1.
Reactor........................................................................................................
15
2.2. Sonication 15
2.3. Particle size
analysis..................................................................................
16
2.3.1. Theory applied. 16
2.3.2. Organology, measuring and data interpretation... 17
2.4. Gel Permeation Chromatography 18
2.4.1. Principle of operation. 18
-
Table of Contents
2.4.2. Organology and data interpretation. 19
3. Experimental Part.. 21
3.1. Materials. 21
3.2. Experimental
procedure............................................................................
21
3.2.1. Emulsion polymerization. 21
3.2.2. Miniemulsion polymerization.. 23
4. Tasks 25
5. Bibliography... 27
ii
-
Theoretical Part
Introduction
The aim of the present laboratory exercise is the comparison of
two heterogeneous
polymerization techniques, namely emulsion and miniemulsion
polymerization.
To this end, two batch free-radical polymerizations of styrene
in the presence of a Chain
Transfer Agent (CTA), one in a conventional emulsion and one in
a miniemulson system will
be carried out in order that the differences both in terms of
system and product properties are
conceptualized. Special emphasis is to be given to the evolution
of conversion of monomer
and CTA with time, as well as to the Molecular Weight
Distribution (MWD) resulting in the
two cases.
Samples characterization will be conducted by gravimetry, Gel
Permeation
Chromatography (GPC) and Laser Diffraction for the sake of
monomer conversion, MWD
and Particle Size Distribution (PSD) determination
respectively.
1. Theoretical part
1.1. Molecular solutions, colloidal systems and suspensions
There is an important class of materials, intermediate between
the pure substances and
the molecularly dispersed systems, known as colloidal
dispersions or simply colloids, that
possess special properties of great practical interest. In these
systems, although one
component is microdispersed into another component, the size of
the dispersed species is
much larger than this of a molecular mixture (a conventional
chemical solution where the
solute and solvent molecules are of comparable size). On the
other hand, these dispersed
species are not large enough to rapidly sediment due to gravity;
they appear a sedimentation
rate in the order of 10-4 cm/s or even slower, being subject to
an irregular, random movement
(Brownian motion).
Colloids are therefore constituted by a finely dispersed
(discontinuous) phase,
homogeneously distributed in a dispersion medium (continuous
phase). Both the continuous
and the discontinuous phase may be solid, liquid or gaseous
(nine cases in total). The
properties of colloidal character appear when the dimensions of
the dispersed phase are
between 1 and 1000 nm, without however these limits being
strict. In some special cases, like
1
-
Theoretical Part
for example in emulsions, the dispersed entities may be larger.
When the dimensions of the
dispersed phase are smaller than 1 nm the colloidal behaviour
coincides with that of
molecular solutions. On the contrary, dispersions of particles
generally bigger than 1 m are
called suspensions and may sediment relatively fast if their
density and size are big enough.
1.2. Polymer latexes and the concept of heterogeneous
polymerization
Apart from the classical homogeneous polymerization techniques,
like bulk and solution
polymerizations, the heterogeneous polymerization techniques
employed for the synthesis of
polymer nanoparticles dispersed in a non-solvent, so called
polymer latexes (latex being a
word meaning droplet in ancient Greek), is much more favourable
in polymer science due
to a number of reasons. On the one hand, a general trend towards
non-organic volatiles
because of health, security and environmental reasons, makes
formulations of polymers in
healthy dispersants, in general water, necessary. A second
driver is the technological trend
towards high polymer contents at reasonable processing
viscosities that can only be obtained
by polymer dispersions. Thirdly, polymer dispersions are a
possibility to control or imprint an
additional length scale into a polymer bulk material, given by
the diameter of the particle.
Polymer dispersions are an omnipresent part of worldwide
commerce as well as scientific
polymeric studies. It is no exaggeration to say they are
important for our daily life mainly
because of their versatile properties and applications for the
production of high quality
polymers with special tailored properties in paper making, all
kinds of paints and coatings,
construction industry, adhesives, textile and leather
industries, medicine and pharmaceuticals
and so forth.
Since the technical term polymer dispersions describes a special
state of matter and not
a special chemical composition of the polymeric material, any
kind of polymer can be
obtained in such a form. There is a variety of techniques to
generate latexes, such as
dispersion, suspension, emulsion, miniemulsion and microemulsion
polymerization, each
featuring different characteristics.
It is a long-standing idealized concept in heterophase
polymerization to generate small,
homogeneous and stable droplets of monomer or polymer precursors
which are then
transformed to the final polymer latexes. Polymerization should
subsequently proceed inside
the dispersed species as in a hypothetical bulk state, where the
continuous phase is still good
to transport initiators, side products, heat and even the
monomer itself. This state we call
nanoreactors, since every droplet behaves as an independent
reaction vessel without being
2
-
Theoretical Part
seriously disturbed. The polymerization in such nanoreactors
takes place in a highly parallel
fashion, i.e. the synthesis is performed in ca. 1020
nanocompartments per liter, separated from
each other by the continuous phase.
1.3. Emulsion Polymerization
1.3.1. Concerning the emulsion polymerization in general
Emulsions are defined as microheterogeneous systems, constituted
by at least one non-
miscible liquid dispersed in another one in form of droplets,
which diameter is usually
between 50 and 5000 nm. Those systems feature a minimum
stability, that can however be
enhanced by adding suitable substances, like surfactants, finely
ground solids etc. Optically
examined, the emulsions appear depending on the size of the
dispersed droplets, milky
white- (when the droplets are small, typically between 1 and
1000 nm) or bluish-grey-
semitransparent (when the droplets are relatively big).
Emulsion polymerization is a very interesting technique with
some of its main features
being:
Particle size in-between 5 nm and 5 m. Mainly water soluble
initiators are used. Special substances, called emulsifiers, play a
pivotal role in the emulsion polymerization
process by keeping the droplets and the particles under
continuous, steady suspension.
The kinetics of the emulsion polymerization are significantly
different compared with all other techniques.
The final product is a stable latex. Emulsion polymerization
affords the means of increasing the polymer molecular weight
without decreasing the polymerization rate and altering the
systems temperature, something
that makes it practically very attractive.
There has to be noted however that strictly spoken, emulsion
polymerization is good for
the radical polymerization of a set of barely water-soluble
monomers, yet already heavily
restricted in radical copolymerizations, let alone other polymer
reactions. The reason for this
is the polymerization mechanism, where polymer particles are the
product of a kinetically
controlled growth, built from the centre to the surface, while
all the monomer has to be
transported by diffusion through the aqueous phase, as will be
discussed later.
3
-
Theoretical Part
The basic components of an emulsion polymerization system are
the monomer (or the
monomers in case of a copolymerization), the dispersing medium
(most commonly water), the
emulsifier and the water soluble initiator.
1.3.2. Surfactants
Both the nature and the quantity of the surfactant play an
important role in the reaction
evolution. Surfactants are in general molecules with a polar or
ionic group as head and a
hydrocarbon chain as tail, ranging between 10 and 20 carbon
atoms. These emulsifiers display
the following properties:
They are water soluble due to their polar head if their tail is
not too long. They lower the surface energy between an aqueous and
an oil phase, arranging
themselves at the interphase in such a way that their polar head
is in the aqueous phase,
while their hydrocarbon tail in the oil phase.
At low concentrations these surfactant molecules on the
interphase are in equilibrium with the surfactant molecules in the
bulk. Above a characteristic concentration, referred
to as the Critical Micelle Concentration (CMC), aggregates of
surfactant molecules are
formed in the bulk, referred to as micelles, which are in
equilibrium with free surfactant
molecules in bulk. Micelles are constituted by 20-100 surfactant
molecules, while their
typical diameter is 5-10 nm with their number in the solution
being a function of the
surfactant concentration.
1.3.3. Qualitative mechanistic picture of emulsion
polymerization
The physical picture of emulsion polymerization is based on the
original qualitative
picture of Harkins: When in the water-surfactant system a non-
or slightly- water soluble
monomer is added, the latter is distributed like this: A small
portion stays in the bulk as
independent molecules. A greater, yet relatively small portion
enters inside the micelles, as
proved by the fact that the micelles grow bigger in size while
the monomer is being added.
The greatest monomer portion however lies dispersed in form of
droplets, whose actual size
depends upon the shear rate and is typically between 1-10 m.
These droplets are therefore
much bigger than the micelles and although the typical micelles
concentration is 1018 per ml,
the monomer droplets number is 1010-1011, this being the
maximum. It also has to be noted
that the total micelles surface is much greater than this of the
monomer droplets.
The initiator lies in the aqueous phase, where the free radicals
for the initiation of the
polymerization appear. The rate of these radicals production is
typically of the order of 1013
4
-
Theoretical Part
radicals per ml per second. Undoubtedly, polymerization of the
monomer takes place in the
bulk as well, nevertheless this is not important due to the low
monomer concentration in the
bulk. But neither the monomer droplets are polymerization sites,
given the fact that the
initiators used are non-monomer soluble.
Therefore the actual polymerization reaction takes place almost
exclusively inside the
micelles, where the water soluble initiator and the organic,
non-water soluble monomer meet
each other. The micelles are actually favoured to be the
polymerization locus, both due to
their high concentration and their much greater surface to
volume ratio as opposed to the
monomer droplets. Despite the fact that the initiator is water
soluble, while the hydrophobic
monomers are being added they oblige the growing chain to seek
an organic, hydrophobic
environment and as such the inside of the micelles is most
convenient for the above
mentioned reasons. During the course of the polymerization, the
micelles become swollen
with monomer arriving initially from the bulk and later on from
the monomer droplets. Thus,
three types of particles exist now in the system: monomer
droplets, non active micelles where
no polymerization is taking place and active micelles, where
polymerization is taking place.
The system undergoes a major change when a small percentage of
the monomer is
converted to polymer. Just a small percentage, ca. 0.1%, of the
initial micelles is actually
activated. While these active micelles grow bigger in size,
containing both monomer and
polymer, they adsorb increasingly more emulsifier molecules from
the bulk. Soon enough the
emulsifier concentration in the bulk becomes lower than CMC,
thus some of the inactive
micelles are destabilized and destroyed to independent
molecules.
Figure 1. Simplified representation of an emulsion
polymerization system.
5
-
Theoretical Part
At a conversion around 2-15%, depending on each systems nature,
the active micelles
have grown too much in comparison to the initial ones, therefore
they are no longer
considered to be micelles, but polymer particles instead, or
better polymer particles swollen
with monomer. At this stage all inactive micelles have vanished
and actually all the emulsifier
content has been adsorbed on the polymer particles.
Consequently, the monomer droplets are
no longer stable and if the shear would cease, they would
aggregate. The polymerization is
continued in the polymer particles at the same rate, as the
monomer concentration in the latter
is kept constant via monomer diffusion from the initial monomer
droplets. During this critical
stage of the emulsion polymerization the number of the polymer
particles remains constant,
while the monomer droplets shrink as the polymer particles grow
bigger. This stage is
characterized by steady-state conditions.
Figure 2. The emulsion polymerization process
Finally, at a conversion of 50-80% the monomer droplets have
totally vanished and all
the non reacted monomer is actually contained in the polymer
particles. At this stage, the
polymerization proceeds with an ever decreasing rate, while the
monomer concentration in the
polymer particles is continuously decreasing as well.
Conversions of up to 100% are fairly
common. The final latex diameter is typically 50-200 nm, being a
size in-between this of the
initial micelles and the initial monomer droplets.
1.4. Miniemulsion Polymerization
1.4.1. Definition of a miniemulsion
A stable emulsion of very small droplets is for historical
reasons called a miniemulsion.
According however to a more strict definition, miniemulsions are
specially formulated,
6
-
Theoretical Part
critically stabilized dispersions with droplets ranging between
50 and 500 nm in size,
although they are rather characterized by a special mode of
operation and special properties
than by a specific size range. They are formed by applying high
shear to a system containing
water, oil, a surfactant and an osmotic pressure agent insoluble
in the continuous phase. It is
delineated that miniemulsions rely on the appropriate
combination of shear treatment and the
two stabilizing agents.
1.4.2. Coalescence by collisions and the role of the
surfactant
It is evident that destabilization of emulsions or miniemulsions
can occur by collision of
the droplets resulting to coalescence (a bimolecular process).
The handling of this problem is
the standard question in colloid science and is usually solved
by addition of appropriate
surfactants which provide the necessary colloidal stability to
the droplets against coalescence
by collision, controlled by the type and amount of the employed
surfactant.
1.4.3. Ostwald ripening and the role of the ultrahydrophobe
When the surface of a liquid is curved, the pressures at the two
sides of the interface are
not equal. This pressure difference is given by the
Young-Laplace equation:
1 1(r r
= + ) (1) where r stands for the radius of the droplet and for
the surface tension.
When an oil in water emulsion is created by mechanical agitation
of a heterogeneous
fluid containing surfactants, a distribution of droplet sizes
results. Even when the surfactant
provides sufficient colloidal stability to the droplets, the
fate of the droplet distribution is
dictated by thermodynamics and more precisely by the different
droplets size or Laplace
pressure, which increases with decreasing droplet size,
resulting in a net mass flux by
diffusion. If the droplets are not stabilized against diffusion
degradation (Ostwald ripening, a
monomolecular process) small ones will eventually disappear,
increasing the average droplet
size.
Unstable emulsions may however be stabilized by the addition of
small amounts of a
third component preferentially located in the dispersed oil
phase. The Ostwald ripening can
thus be efficiently slowed down as this hydrophobic agent serves
to build up an osmotic
pressure that counteracts the inherent Laplace pressure of the
droplet. For inducing osmosis
effectively it is important to choose an agent which can hardly
diffuse from one droplet to
another, being trapper in each droplet. The minimemulson
evolution is driven by the
7
-
Theoretical Part
competition between the osmotic pressure of the trapped species
and the Laplace pressure of
the droplets. This osmotic pressure is given by the classical
vant Hoff equation:
osm = C R T (2)
Let us note that for this ultrahydrophobic agent the term
costabilizer may also be used,
but the term cosurfactant is rather misleading as this substance
is not surface active, is not
placed on the droplets interface, nor does it lower the
interfacial tension or forms micelles
itself.
1.4.4. Preparation of a miniemulsion
Mechanical emulsification starts with a premix of the two
incompatible fluid phases,
resulting to a conventional emulsion by simple stirring, which
however is not capable to
transfer the energy sufficient in order to obtain small and
homogeneously distributed droplets.
A much higher energy for the communition of bigger droplets into
smaller ones is required,
ideally equal to the difference in surface energy corresponding
to the newly formed interface,
but actually significantly higher due to viscous dissipation.
That is:
Eminiemulsification > (3)
where is the surface tension and the new additional surface
created. As high force
dispersion device, today ultrasonication is used, that induces
droplet formation and disruption
under the influence of longitudinal density waves.
Figure 3. The principle of miniemulsion polymerization.
8
-
Theoretical Part
1.4.5. The critical stability of miniemulsions
The monomer droplets change quite rapidly in size throughout
sonication. With
increasing the time of the ultrasonication, the droplet size
decreases and therefore the entire
oil/water interface increases. Since a constant amount of
surfactant has to be distributed now
at larger interphase, the surface tension also increases. In the
beginning of the
homogenization, the polydispersity of the droplets is quite
high, but by constant fission and
fussion processes, the polydispersity decreases and the
miniemulsion reaches then a steady
state, where a dynamic equilibrium of droplet fission and fusion
rates is achieved. This also
means that miniemulsions come to the minimal particle sizes
under the applied conditions,
while they make use of the surfactant in the most effective way
possible.
Figure 4. Scheme for the formation of miniemulsion by
ultrasound
To put it straight, the resulting nanodroplets are at the
critical borderline between stability
and instability, called critically stabilized. Properly
formulated miniemulsions are
controlled by colloidal stability and neither Ostwald ripening,
nor insufficient mechanical
shearing is a problem, at least for a time scale of several
days.
Since the discussion concerns an equilibrium state, it is good
to make an attempt to
describe it from the scope of thermodynamics. The chemical
potential i of the ith droplet can
be expressed as the net pressure difference between its two
sides and is a function of its radius
2i Laplace osmir = = (4)
If the miniemulsification process is successful, the chemical
potential in each droplet will
equilibrate. The equality of the chemical potential, or else of
the net pressure in every droplet
does not represent a real thermodynamic equilibrium; on the
contrary, miniemulsions as
systems are only thermodynamically metastable.
Considering the dispersed phase of a miniemulsion as a system
consisting of n droplets
each of which with volume vi and chemical potential , the total
chemical potential of this
9
-
Theoretical Part
system would be the volume-weighted average of the chemical
potentials of all droplets, since
the chemical potential is an intensive thermodynamic
parameter:
1
1
n
ii
total n
ii
v
v
=
=
=
(5)
1.4.6. Mechanistics in miniemulsion polymerization
Droplet nucleation is by far the dominant mechanism for Styrene
miniemulsions. It
suggestes that the droplets formed during the miniemulsification
step are polymerized via
radicals which enter the monomer droplets, if the initiator is
water-soluble (i.e. the formation
of free primary radicals takes place in the water phase). Due to
the statistics of radical entry
and overall size, every single droplet is nucleated.
Monomer diffusion to the reaction sites is of no kinetic
importance, since there is already
the maximal monomer concentration at all reaction sites.
Moreover, the monomer diffusion is
balanced by the high osmotic background of the hydrophobe which
makes the influence of the
formed polymer chains less important. As a result, the growth of
the minidroplets is much
slower than the actual polymerization time.
Figure 5. The miniemulsion polymerization process
Therefore, since the polymerization is initiated in each of the
small stabilized droplets,
without major secondary nucleation or mass transport processes
involved, the particle number
and identity is preserved during polymerization. The final
polymer particles can essentially be
understood as nearly 1:1 polymerized copies of the original
droplets, the size of which is
given by the dispersion process and droplet stability, but not
any polymerization parameter. In
miniemulsion polymerization the concept of nanoreactors is
realized to a wide extent,
10
-
Theoretical Part
making miniemulsion polymerization particularly attractive for
the formation of polymeric
nanoparticles which are hardly accessible by other types of
heterophase polymerizations.
1.5. Chain transfer
1.5.1. Main concept
In many free-radical polymerization systems the molecular weight
of the polymer
produced may be lower than the one predicted if the growth of
macroradicals in these systems
is prematurely terminated by transfer of a hydrogen or other
atom or species to the
macroradical from some special compound present in the reaction
mixture (i.e. the monomer
itself, initiator, solvent or other species, as the case may
be). The donour species itself
becomes a radical in the process and the kinetic chain is not
terminated if this new radical can
add monomer. Although the rate of monomer consumption may not be
altered by this change
of radical site, since a chain-breaking reaction occurs the
initial propagating macroradical will
have ceased to grow and its size is less than it would have been
in the absence of the atom
transfer process. These types of radical displacement reactions
are called chain transfer
processes and the corresponding compounds involved chain
transfer agents.
1.5.2. Kinetics
If chain transfer reactions take place in the system another
equation has to be introduced
to the typical free-radical polymerization kinetic scheme of
initiation, propagation and
termination, that would be depicted as:
Stage Reaction Rate
Initiation 2 2ik
1I R ri = 2 f ki [I2] (6)
Propagation 1pk
n nR M R
++ [ ] [ ][ ]p pd Mr k Mdt R= = (7)
where: 1
[ ] [ nn
]R R =
= (8) In particular, if the polymerization is taking place in a
dispersed phase, then:
[ ] pR nN = (9)
with n being the average number of radicals per polymer particle
and Np the
number of polymer particles per cm3 of continuous phase.
11
-
Theoretical Part
Chain Tranfer Tkn nR T P + +T [ ][ ]T TdTr k Tdt R
= = (10) where T stands for the CTA responsible for the
production of a so-called dead
polymer chain Pn and a new radical T, capable of subsequently
reinitiating the
polymerization:
Reinitiation 1rkT M R + [ ][r rr k T M= ] (11)
Termination tckn m n mR R P
++ [ ][tc tc n mr k R R ] = (12) tdkn m n mR R P
+ + P [ ][td td n mr k R R ] = (13) where rtc and rtd are the
rates of termination by combination and
disproportionation respectively, while ktc and ktd stand for the
corresponding
rate constants.
Chain transfer is important in that it may alter the molecular
weight of the polymer
product in an undesirable manner. On the other hand, controlled
chain transfer may be
employed to advantage in the control of molecular weight at a
specified level. The most
widely used CTAs are compounds with only one relatively weak
bond like thiols, disulfides,
carbon tetrachloride or tetrabromide, but even hydrogen and
propane can be used with a very
reactive polyethylene radical.
A chain-transfer constant CT for a CTA is defined as the ratio
of the rate constant kT for
the chain transfer of a propagating radical with that agent over
the rate constant kp for the
propagation of the radical:
TT
p
kCk
= (14)
By dividing the rate expression for transfer by that for
propagation:
[ ] [ ][ ] [ ]T
d T TCd M M
= (15)
and subsequently expressing the monomer concentration [M] in
terms of the monomer
conversion XM we obtain:
[ ][ ] 1
MT
M
dXd T CT X
= (16)
which after integration results in an expression of the CTA
concentration [T] as a function of
the monomer conversion:
0[ ] [ ] (1 ) TC
MT T X= (17)
12
-
Theoretical Part
Via this equation it is possible to calculate the evolution of
the CTA conversion XT with
time if the monomer conversion is determined and the
chain-transfer constant value known.
Moreover, the ratio of the monomer to the CTA concentration may
be now calculated as
a function of the monomer conversion:
100
[ ][ ] (1 )[ ] [ ]
TCM
MM XT T
= (18)
The actual value of this ratio is particularly crucial for
describing the instantaneous Chain
Length Distribution of a macromolecular mixture, as will be
discussed later on.
1.6. Chain Length Distribution of a polymeric mixture
1.6.1. Average Molecular Weight
The average Molecular Weight (MW) of a macromolecular mixture,
is commonly
expressed by means of two statistical quantities, the number and
the weight average
molecular weight, defined respectively as:
n ii
iM x M
= (19) w i
iiM w M
= (20)
where xi and wi are the mole and weight fraction of the
macromolecular species with actual
molecular weight Mi contained in the polymeric mixture.
1.6.2. Average Degree of Polymerization
Another statistical parameter describing the Chain Length
Distribution (CLD) of a
polymeric mixture is the average Degree of Polymerization (DP),
which is defined as the
average number of structural units per polymeric chain. This may
equivalently to nM and
wM be expressed in terms of number or mass distribution, as the
number or weight average
degree of polymerization respectively:
11
nn
i
MiX x iM
== = (21)
11
ww
i
MiX w iM
== = (22)
13
-
Theoretical Part
where M1 is the molecular weight of one monomer unit, while xi
and wi stand for the mole
and weight fraction of the macromolecular species with i monomer
units contained in the
polymeric mixture.
1.6.3. Polydispersity Index
Yet another useful statistical parameter, which gives a feeling
of the distributions width,
is the Polydispersity Index (PDI):
1wn
XPDIX
= (23)
1.6.4. Instantaneous Properties
For a free-radical polymerization system the instantaneous CLD
properties, that is their
values at any time instant, may be written in terms of two
kinetic parameters, and as
following:
1
2
inst
nX =
+ (24)
2
2( 3 2 )( )
inst
wX += + (25)
2
2( 3 2 )( 1 2 )( )( )
instPDI + += + (26)
where:
[ ][ ]
tc
p
k Rk M
= (27)
[ ][ ][ ] [ ]
tdT
p p
k Rkk M k M
= + (28)
Nevertheless, in the presence of a CTA in the system the
inactivation of the growing
macroradicals becomes controlled by the chain-transfer reactions
rather than the termination
reactions themselves. Therefore and reduce to:
= 0 (29)
[ ] [ ][ ] [ ]
TT
p
k TCk M
= = (30)
thus allowing the above instantaneous properties to be
calculated as functions of the monomer
to CTA concentration ratio if the chain transfer constant value
of the system is known:
14
-
Devices and Characterization Techniques
1 [ ][ ]
inst
nT
MXC T
= (31)
2 [ ][ ]
inst
wT
MXC T
= (32)
( )instPDI = 2 (33) In fact the chain transfer constant value
may be calculated from the boundary condition:
000 0
[ ]1 [ ] 1[ ] [ ]
inst
n T instt o T t n
MMX CC T TX= =
= = (34)
2. Devices and Characterization Techniques
2.1. Reactor
The polymerization reactions are going to be carried out in 250
ml round glass flasks,
immersed in a thermostatic oil bath and equipped with a reflux
condenser using cool water.
The necessary stirring is achieved by means of a magnetic
beaker. This setup is equivalent to
a laboratory batch reactor. Assuming sufficient mixing, the
reactor may be regarded as
homogeneous, that is with concentration and temperature
independent of their location inside
the reactor vessel.
Vacuum may be applied to the system, which can also be purged
with nitrogen and left to
operate under inert nitrogen atmosphere, in a slight
overpressure. Finally, the flasks feature
two outlets that may be used for inserting a thermometer or
thermocouple or for sampling
purposes.
A full operation cycle of a batch reactor consists of the
following stages: loading of the
reactant components, running of the reaction, removal of the
reaction mixture and finally
cleaning and preparation of the reactor for the next cycle.
2.2. Sonication
Ultrasonication treatment possesses a special place among the
techniques for size
decreasing and droplets creation or rupture. A sonification
device creates high frequency
15
-
Devices and Characterization Techniques
sound waves, offers in other words high amounts of mechanical
energy to the system under
treatment, resulting for example in the formation of an
emulsion.
For the miniemulsification a UP 400S Dr Hielscher sonifier will
be used. The
ultrasounds are generated by a cylindrical metal probe, which
has to be well immersed into
the dispersion. Two are the important parameters that should be
controlled during sonication:
Time of treatment and power offered, expressed as amplitude
percentage. The ultrasounds
application may in addition be continuous or pulsed.
2.3. Particle size analysis
For determining the Particle Size Distribution (PSD) of the
latex prepared the Malvern
Instruments Ltd. Mastersized Hydro 2000 will be used, which
applies the laser diffraction
method.
2.3.1. Theory applied
There are several theories and models that the modern particle
size analyst can use. One
of the simplest is the Frauenhofer model that can predict the
scattering pattern created when a
solid, opaque disc is passed through a laser beam. This model is
satisfactory for some
particles but does not describe the scattering exactly, since
very few particles are disc shaped.
Moreover, as the particle size increases, the scattering pattern
becomes still more
complicated as the light waves from each scattering centre in
the particle increasingly
interfere with one another, so that the intensity measured shows
pronounced maxima and
minima at particular angles, determined by the particle size and
refractive index. The accepted
theory which accurately predicts the light scattering behaviour
of all materials under all
conditions is the Mie theory (1908). Mie theory was developed to
predict the scattering
pattern of spherical particles of any size and deals with the
way light passes through, or is
absorbed by the particle. This theory is more accurate, but it
does assume some specific
information about the particle is known, such as its refractive
index and absorption.
That result will not be described in detail, however the key
point of this theory is that if
the size of the particle and other details about its structure
are known, one can accurately
predict the way it will scatter light. Each particle size will
have its own characteristic
scattering pattern, like a fingerprint, that is unlike any other
particle size.
16
-
Devices and Characterization Techniques
2.3.2. Organology, measuring and data interpretation
The Mastersizer works backwards from the above mentioned
theories by using the optical
unit to capture the actual scattering pattern from a field of
particles. Subsequently, using the
theory briefly described above, it can calculate the size of the
particles that create this pattern.
The basic parts of the Mastersizer instrument are three: the
optical unit, a carrier for
dispersing the sample and a computer unit. There are three
distinct procedures involved in
measuring a sample:
First, the sample is prepared, dispersed in the correct
concentration and then delivered to the optical unit. This is the
purpose of the sample dispersion accessories. Sample
preparation
is the most important stage of making a measurement, so care
must be taken that the sample is
not poorly prepared (i.e. being unrepresentative or badly
dispersed).
Second comes the capturing of the scattering pattern from the
prepared sample, referred to as measurement, which is actually the
function of the optical unit. During the laser
diffraction measurement the particles are passing through a
focused laser beam, scattering the
light at an angle reversely proportional to their size. The
angular intensity of the scattered
light is measured by an array of individual photosensitive
detectors, each of which collects the
light scattered from a particular range of angles. The
correlation between the intensity of the
scattering and the scattering angle is the main source of
information for calculating the paricle
sizes. The scattering pattern generated by the particles may
also be predicted by Mies theory.
Figure 6. Operation principle of the laser diffraction particle
size analyzer.
a ) Main parts of the optical unit. b) Magnification of the
sample flow cell.
17
-
Devices and Characterization Techniques
The detector array within the optical unit takes a snap-shot of
the scattering pattern.
Obviously this snap-shot will only capture the scattering
pattern from the particles passing
through the analyzer beam at that particular time. Taking only
one snap-shot may not give a
representative reading of the scattering pattern. To overcome
this, the Mastersizer takes many
snap-shots and averages the result. Typically, over 2000 snaps
are made for each
measurement, each of them lasting 1ms.
Finally, once the measurement is complete, the raw data
contained in the measurement are automatically analysed by the
software, using the Mie theory. Once this analysis is done,
the information may be displayed in various ways.
For interpreting the results offered by the device, three
fundamental concepts have to be
clarified: First, the measurement results (the original PSD
derived) are volume-based. Second,
the result is expressed in terms of equivalent spheres, assuming
that each particle is perfectly
spherical. Third, the distribution parameters are geometrically
and not arithmetically
optimized, in order that the best optical analysis is
achieved.
2.4. Gel Permeation Chromatography
2.4.1. Principle of operation
Gel Permeation Chromatography (GPC) or more correctly termed
Size Exclusion
Chromatography (SEC), is a column fractionation method for
polymers providing a relative
molecular weight. In GPC, solvated polymer molecules are
separated according to their sizes,
such allowing the determination of their weight distribution and
their average molecular
weight. For the samples characterization an HP 1100 GPC
Helwet-Packard analyzer
equipped with the respective software is going to be used.
In GPC the mobile phase (a dilute solution of polymer in the GPC
solvent) is pumped
continuously through a column packed with the stationary phase
(a porous gel which is
wetted by the fluid). As the sample moves along the column with
the mobile phase, the larger
polymer molecules are excluded from many pores of the stationary
phase, traveling mainly in
the interstitial volume between the porous beads of the packing
material. On the other hand,
the smaller molecules find more pores of the stationary phase
accessible, diffusing in and out
of the actual pores of the packing material, a moving being
governed by Brownian motion (an
entropically governed phenomenon). The smaller the molecule, the
more of the stationary
phase volume is accessible to it and the longer it stays in the
column. Consequently, smaller
molecules are eluted later.
18
-
Devices and Characterization Techniques
Figure 7. a) GPC column separation of a polymer, b) Elution
peaks of a GPC chromatogram
The separation of a solute of a given size in solution is
determined by a distribution
coefficient, KGPC, which is the fraction of the internal pore
volume of the gel accessible to this
solute, Vi; in other words KGPC is the ratio of Vi over the
total pore volume. The value of
retention volume, Vr, for a certain solute is:
Vr = V0 + KGPC Vi (35)
where V0 is the interstitial or mobile phase volume. Very large
molecules, completely
excluded from the pores of the gel, are eluted at the
interstitial volume (a case where Vr = V0
and KGPC = 0). Very small molecules, having free access to both
the stationary and the mobile
phase, are eluted at the sum of the intersticial and the pore
volume (a case where Vr = V0 + Vi
and KGPC = 1). For all intermediate species KGPC is a separation
constant between 0 and 1.
2.4.2. Organology and data interpretation
A variety of column packings for GPC analysis may be used.
Essential conditions for the
effective fractionation of the polymers are that pore sizes in
the column packing should be
comparable to polymer sizes in the solution and that packings
should have a substantial pore
volume, typically 0.5 < Vi/V0 < 1.65 for macroporous
packing.
Regarding the eluent, this should ideally be a good solvent for
the polymer, permit high
detector response from the polymer and wet the packing surface.
Common eluents for
polymers that can be dissolved at room temperature are
tetrahydrofuran and chloroform.
The detector system in GPC is typically concentration sensitive,
measuring the change of
the refractive index or UV absorption of the effluent, which is
linearly related to the polymer
concentration. More precisely, a light beam crosses the sample
and the reference cell twice.
19
-
Devices and Characterization Techniques
The reference cell contains pure eluent and when the sample cell
is filled with eluent too, the
light beam is directed with a lens so as the two detectors
photodiodes D1 and D2 measure
the same light intensity. When however, a liquid with a
different refractive index enters the
sample cell, the beam will be geometrically dislocated,
proportionally to the relative change
of the refractive index, causing the measured light intensity at
the two diodes to change
respectively.
Figure 8. Schematic representation of the RI detector
Solvent flow is conveniently measured by means of elapsed time
since sample injection,
relying implicitly on a constant solvent pumping rate. The
column effluent is monitored by
the detector, whose response at each time instant (corresponding
to a certain elution volume)
is proportional to the amount of polymer in solution. This
response is transformed into a
weight fraction of the total polymer amount and the
corresponding elution volume can be
directly related to the molecular weight by means of suitable
calibration. The resulting
chromatogram is therefore a weight distribution of the polymer
as a function of the molecular
weight.
The calibration is performed injecting into the column some
commercially available PS
standard of known MW and recording the corresponding
chromatogram. The resulting graph
is narrow, allowing us to associate the elution volume to the
MW. Thus, using a series of such
samples a calibration curve for that particular polymer in that
GPC solvent and column set
may be produced.
20
-
Experimental Part
3. Experimental Part
3.1. Materials
The basic physical and chemical properties of the various
materials used, as well as their
role in the polymerization reaction mixture are summarized in
the following table.
3.2. Experimental procedure
3.2.1. Emulsion polymerization
Reaction The relative amounts of the components and the reaction
conditions for the emulsion
polymerization experiment are given at the corresponding
table.
The hydrophobic components (the monomer Styrene and the CTA
tert-dodecyl
mercaptan) should be added to the aqueous surfactant (SDS)
solution containing the water-
soluble initiator (KPS). This mixture should remain for a few
minutes under stirring in a
beaker at room temperature, so that an emulsion is formed.
Table 1. Recipe for the emulsion polymerization experiment.
Component Relative amount Actual Mass or Volume
H2O 120 ml
St 20% wt water 26.5 ml
SDS 1% wt monomer 0.24 g
KPS 1% wt monomer 0.24 g
CTA 2.5% wt monomer 0.70 ml
Reaction Temperature: constant at 50 oC
The reactor may now be loaded with the emulsion (the heating
medium in the jacket must
have reached an acceptable temperature, initially for heating
and later for maintaining the
reaction mixture at the desired temperature). Right after the
loading, extreme care should be
taken that the whole system is firstly air-tightly sealed and
then thoroughly purged with
nitrogen in several alternating vacuum-nitrogen cycles. The
reaction may then be left running
under inert nitrogen atmosphere. The assumption that the
reaction actually begins at the time
21
-
Experimental Part
Table 2. Physical and chemical properties of the materials used
and their role in the polymerization system.
Name Styrene Sodium Dodecyl Sulfate Hexadecane Potassium
Persulfate tert-Dodecanethiol
Abbreviation St SDS HD KPS CTA
Molecular formula
C16H34
CH3(CH2)8C(CH3)2SH
Role in the system Monomer surfactant hydrophobe polymerization
initiator Chain Transfer Agent
Physical state clear liquid powder clear liquid powder clear
liquid
Colour colourless white colourless white colourless
Smell pungent light light odourless repulsive
Molecular Weight (g/mol) 104.2 288.4 226.4 270.3 202.4
Density (g/ml) 0.906 0.370 0.773 2.477 0.860
Solubility in water (g/l) insoluble 10 insoluble complete
negligible
Boiling Point (oC) 145-146 N/A 287 N/A 227-248
Melting Point (oC) -31 206 18 100 N/A
Autoignition temperature (oC) 490 N/A 202 N/A 230
22
-
Experimental Part
when the reaction mixture reaches the desired reaction
temperature is consistent enough for
the purpose of this particular experiment.
Several samples have to be carefully withdrawn from the reactor
at regular time intervals
for subsequent analyses and immediately cooled down in an ice
bath. After reaction
completion the final latex will be obtained and all the
equipment should be efficiently
cleaned.
Characterization The samples collected will be dried over the
weekend in an oven under vacuum, in order
that the all the volatile compounds are perfectly evaporated.
Thus, the weight fraction of the
solids content in the reaction mixture at the specific time
instant ti that the sample was taken
will be determined gravimetrically, from which the corresponding
monomer conversion XM
can easily be calculated:
0
(% | %|% |
i
i
)solids t t SDS KPS CTA HDM t t
St t t
wt wtX
wt= + +
==
= + (36)
These dried latex samples will be dissolved in chloroform, which
is a particularly strong
solvent for polymers, and used for GPC analyses in order to
determine their MWD.
Samples of the intermediate and final latexes will be injected
into the particle size
analyzer in order to determine their PSD.
3.2.2. Miniemulsion polymerization
Reaction and characterization The relative amounts of the
components and the reaction conditions for the miniemulsion
polymerization experiment are given at the corresponding table.
For the sake of a fair
comparison between the two polymerization techniques, which is
the actual objective of this
exercise, the miniemulsion recipe is substantially the same as
the emulsion recipe, the only
differences being the introduction of the hydrophobe
(Hexadecane) in the system and the
intense shearing of the emulsion, induced by ultrasound
treatment.
Therefore, the oil phase consisting of the monomer (St), the
hydrophobe (HD) and the
CTA (tert-dodecyl mercaptan) has to be mixed with the aqueous
phase consisting of the
surfactant (SDS) and the water-soluble initiator (KPS) dissolved
in water. The resulting
dispersion should remain for a few minutes under mechanical
stirring at room temperature, in
order that a conventional emulsion is first formed and
subsequently sonicated at 0 oC, so that
23
-
Experimental Part
finally a miniemulsion is successfully created. Right after
miniemulsification it is
advantageous that an additional small aliquot of surfactant is
added to complete the surface
coverage of each minidroplet (post-stabilization) and thus
prevent maturation of the system (a
change in the droplet size with time).
Table 3. Recipe for the miniemulsion polymerization
experiment.
Component Relative amount Actual Mass or Volume
H2O 110 + 10 ml
St 20% wt water 26.5 ml
SDS 2% wt monomer 0.24 + 0.24 g
HD 4% wt monomer 1.24 ml
KPS 1% wt monomer 0.24 g
CTA 2.5% wt monomer 0.70 ml
Reaction Temperature: constant at 50 oC
Sonication: 2 min at 90% amplitude, continuous cycle (0 oC)
The rest of the reaction, sampling, cleaning and
characterization procedures to be
followed are exactly the same as in the emulsion polymerization
case. Needless to say
however, differences in the results are anticipated.
24
-
Tasks
4. Tasks
1. Consider two droplets of your miniemulsion, with radiuses r1
and r2.
a) Given that the total volume of the dispersed phase is anyhow
stable, what is the constrain
applying for the arithmetic values of r1 and r2?
b) Plot the chemical potential of the system consisting of these
two droplets as a function of
their radius ratio (X axis in logarithmic scale) and comment on
the shape of the graph bearing
in mind that miniemulsions are critically stabilized systems.
What does this practically mean
for miniemulsions?
c) What would be the mathematical condition if real
thermodynamic equilibrium was
achieved? What would this practically mean for a
miniemulsion?
Assume that the droplets radius may range in the typical
miniemulsion size area. The surface tension
for your system is reported to be 65 mN m-1.
2. Supply the definition of the monomer conversion and then
derive Eq. (16) from Eq. (15).
3. Give the definitions of the mole and weight fractions in the
cases of both the average MW
and DP and then use them to rewrite Eqs. (19) (22) as functions
of the number of moles, n.
Show as well that: . 1PDI
4. a) Calculate the monomer conversion of the samples you
collected for both emulsion and
miniemulsion polymerization. Plot the evolution of conversion
with reaction time (XM vs. t) in
the same diagramme for the two reactions. Comment on the
differences between the two
polymerization techniques.
b) For the miniemulsion polymerization case, calculate and
include additionally in the
previous graph the evolution of the CTA conversion (XT vs.
t).
5. For the miniemulsion polymerization plot the residual monomer
and CTA fractions vs.
monomer conversion ([M]/[M]0 and [T]/[T]0 vs. XM) in the same
diagramme and comment on
the graph.
6. a) Draw and compare the PSD of the polymer particles
resulting from the two
polymerizations (Laser Diffraction measurements obtained for
your final samples).
25
-
Tasks
b) Assuming an average particle size for each case calculate the
total number of polymer
particles in the continuous phase, Np,tot and the number of
polymer particles per cm3 of
continuous phase, Np. What other assumptions do you need to make
for this calculation?
7. Assuming that for 35% monomer conversion steady-state
conditions for the miniemulsion
reaction have been reached, calculate the consumption rates of
the monomer and the CTA (rp
and rT respectively) and compare them by calculating the ratio
of rT over rp. What does this
mean for the MW of the polymer being formed?
Use the Np value you have already calculated. Typically: 0.5n =
For 50oC: 1 1212pk l mol s = and 1 1318Tk l mol
= s
8. a) Using your GPC measurements plot the evolution of the
number average MW with
monomer conversion ( nM vs. XM) in the same diagramme for the
two reactions. Do the same
in a second diagramme for the weight average MW ( wM vs. XM).
Comment on the differences
between the two polymerization techniques.
b) Plot the evolution of the experimental number average DP with
monomer conversion ( nX
vs. XM) in the same diagramme for the two reactions. For the
miniemulsion polymerization
case, calculate and include additionally in the previous graph
the evolution of the theoretical
number average DP (inst
nX vs. XM). Comment on the differences between the
experimental
and the theoretical curve.
c) Repeat subtask (b) for the weight average DP ( wX and
inst
wX vs. XM) and for the PDI (PDI
and (PDI)inst vs. XM).
d) From the diagramme of subtask (b) calculate the experimental
value for the chain transfer
constant and compare it with the one reported ( 1.5TC ), by
calculating the %deviation between the two values. Comment on the
result.
,exp ,
,
% 100%T T theorT theor
C Cdeviation
C= (37)
9. Explain why the whole kinetic analysis of the paragraphs
1.5.2 and 1.6.4 does not apply in
the conventional emulsion polymerization. Suggest qualitatively
what phenomena should also
be taken into account so that it can be used for the emulsion
polymerization as well.
26
-
Tasks
10. Using the values that you calculated for the total number of
polymer particles from
subtask 6 (b) and for the corresponding number average DP from
subtask 8 (b), calculate the
corresponding average number of macromolecular chains contained
in each polymer particle
for the two reactions. (NAvogadro = 6.023 1023 molecules
mol-1)
27
-
Bibliography
5. Bibliography [1] Anonymous, Operators Guide for the Malvern
Mastersizer Hydro 2000S/G, Malvern
Instruments Ltd., United Kingdom (1998)
[2] Antonietti M., Landfester K., Polyreactions in
miniemulsions, Progress in Polymer
Science, 27 (2002), 689-757
[3] Brandrup J., Immergut E. H., Grulke E.A., Polymer Handbook,
John Willey & Sons,
New York (2005)
[4] Hunter R., Foundations of Colloid Science, 2nd Ed., Oxford
University Press, New
York (2004)
[5] Landfester K., Bechthold N., Tiarks F., Antonietti M.,
Formulation and Stability
Mechanisms of Polymerizable Miniemulsions, Macromolecules, 32
(1999), 5222-5228
[6] Morbidelli M., Polymer Reaction & Colloid Engineering,
Lecture Notes, ETH Zurich
(2006)
[7] Nomura M., Suzuki H., Tokunaga H., Fujita K., Mass Transfer
Effects in Emulsion
Polymerization Systems. I. Diffusional Behaviour of Chain
Transfer Agents in the
Emulsion Polymerization of Styrene, Mass Transfer Effects, 51
(1994), 21-31
[8] Odian G., Principles of Polymerization, 4th Ed., John Willey
& Sons, New York (2004)
[9] Panayiotou K., Interphacial Phenomena and Colloidal Systems,
2nd Ed., Ziti Editions,
Thessaloniki (1998)
[10] Panayiotou K., Polymer Science and Technology, 2nd Ed.,
Pegasus Editions,
Thessaloniki (2000)
[11] Rudin A., The Elements of Polymer Science and Engineering,
2nd Ed., Acedemic
Press, London (1998)
[12] Sandler S., Karo W., Bonesteel J., Pearce E., Polymer
Synthesis and Characterization, A
Laboratory Manual, Acedemic Press, London (1998)
[13] Tiarks F., Landfester K., Antonietti M., Silica
Nanoparticles as Surfactants and Fillers
for Latexes Made by Miniemulsion Polymerization, Langmuir, 17
(2001), 5775-5780
[14] Young R. J., Introduction to Polymers, Chapman and Hall
Ltd, London (1983)
28