Anisotropic particles from LC polymers for optical manipulation

Article abstract of DOI:10.1021/ma0613279

We have developed a method to convert 10 different LC acrylate monomers into colloids by dispersion polymerization. This yields nine different types of anisotropic colloids with nematic and different smectic phases. The diameter of these colloids mostly varied between 0.5 and 3.5 μm; it can be adjusted by variation of the solvent mixture and it can be systematically increased by seed polymerization. The polydispersity of the anisotropic colloids is thereby often below 10%. Polarizing microscopy shows that colloids of a size between about 2 to 4 μm appear to have a bipolar director configuration. Smaller colloids appear uniaxially oriented, the resolution does, however, not allow a more refined investigation of the director pattern. These anisotropic spheres (diameter between 0.7 and 3.7 μm) can be trapped with an optical tweezers. Circularly polarized light transfers a torque to the particles, enabling one to rotate them clockwise and anticlockwise, which makes these spheres attractive as actuators. The size dependence of their rotational frequency makes it additionally possible to determine changes of the director configuration with size. For a nematic colloid (P 6). it could be shown that the anisotropy stays constant from 1.6 to 3.4 μm.

Full text of DOI:10.1021/ma0613279

8326
Macromolecules 2006, 39, 8326-8333
Anisotropic Particles from LC Polymers for Optical Manipulation
Melanie Vennes,† Stephen Martin,‡,§ Thomas Gisler,‡ and Rudolf Zentel*,†
Institute of Organic Chemistry, UniVersity of Mainz, Duesbergweg 10-14, D-55099 Mainz, and
Department of Physics-Fach M621, UniVersity of Konstanz, D-78457 Konstanz, Germany
ReceiVed June 14, 2006; ReVised Manuscript ReceiVed September 14, 2006
ABSTRACT: We have developed a method to convert 10 different LC acrylate monomers into colloids by
dispersion polymerization. This yields nine different types of anisotropic colloids with nematic and different
smectic phases. The diameter of these colloids mostly varied between 0.5 and 3.5 µm; it can be adjusted by
variation of the solvent mixture and it can be systematically increased by seed polymerization. The polydispersity
of the anisotropic colloids is thereby often below 10%. Polarizing microscopy shows that colloids of a size between
about 2 to 4 µm appear to have a bipolar director configuration. Smaller colloids appear uniaxially oriented, the
resolution does, however, not allow a more refined investigation of the director pattern. These anisotropic spheres
(diameter between 0.7 and 3.7 µm) can be trapped with an optical tweezers. Circularly polarized light transfers
a torque to the particles, enabling one to rotate them clockwise and anticlockwise, which makes these spheres
attractive as actuators. The size dependence of their rotational frequency makes it additionally possible to determine
changes of the director configuration with size. For a nematic colloid (P 6), it could be shown that the anisotropy
stays constant from 1.6 to 3.4 µm.
Introduction
knowledge there are no systematic studies about the preparation
of anisotropic colloids from different LC materials with this
method, and of the influence of the LC material and its phases
on the properties of the resulting colloids.
Another possible method to produce colloids of LC polymers
is to transfer the prepared polymer into spherical objects via
the miniemulsion process, as reported recently.10,11 Different
side-chain10 and main-chain11 LC polymers were used, but in
this case the sizes of the colloids were only in the range of
about 50-300 nm, which is too small for a direct observation
with a polarizing microscope and too small for manipulation
with optical tweezers.
In this work, we report about the preparation of various
anisotropic colloids of different smectic and nematic acrylate
polymers by a special modification of a precipitation poly-
merization called dispersion polymerization,12 which we recently
described for the first time for this purpose.13 These colloids
have diameters of a few micrometers and can therefore be used
for experiments with optical tweezers, of which we show some
examples.
In recent years colloids, and especially optically anisotropic
colloidal particles, have been of growing interest because they
can be manipulated with optical tweezers. They have, for
example, been used to demonstrate the orbital and spin angular
momentum of light1 and to quantitatively measure on a
micrometer-size scale viscosity in liquid media.2 Birefringent
particles are of interest as potential optical switches3 and
scattering polarizers4 and may be useful for microfluidic
applications as stirrers and valves.5
There are different types of birefringent colloidal particles
available such as fluoropolymer latex particles, inorganic
particles (e.g., calcium carbonate vaterite) and particles out of
liquid crystals.6 One disadvantage of most fluoropolymers and
inorganic particles is their anisotropic shape. Birefringent
colloids from LC polymers, on the contrary, are usually spherical
at particle sizes useful for optical tweezers (around 1 µm). Liquid
crystal (LC) emulsions were mostly prepared by either extrusion
of a liquid-crystalline material into a co-flowing liquid stream,7
leading to highly monodisperse particles with controllable sizes
ranging from some tens to hundreds of micrometers, or by
dropping a solution of the liquid crystal into the dispersion
agent,6,8 leading to particle sizes from about 100 nm to some
micrometers. However both methods have their flaws. The co-
flowing stream procedure has a very low yield of particles and
becomes impractical when the viscosity of the particle material
becomes too high, whereas the dispersion agent method has very
high polydispersity for particles over a few hundred nanometers.
For the stabilization of the LC colloids, mixtures of photopo-
lymerizable diacrylates and photoinitiator were used and po-
lymerized after the droplet formation by UV-irradiation.3,6,8,9
However, in each case only one commercially available
monomer (RM257 or RMM14) was investigated. To our
Results and Discussion
The starting point of this work was the attempt (1) to broaden
the materials basis of LC colloids prepared via dispersion
polymerization, (2) to correlate properties of the starting
monomers (solubility, phase) with the LC colloids obtained, and
(3) to control size and size distribution. Of particular importance
was the attempt to prepare colloids of larger size, as larger
colloids are more suitable for experiments with optical tweezers.
Therefore, we synthesized a large number of different acrylate
and methacrylate monomers with different spacer lengths,
mesogenic units, and polarities (see Scheme 1). The monomers
with ethylene oxide spacer (M 1 and M 2) are, for example,
more hydrophilic than the other acrylate monomers. To compare
similar polymers with different spacer lengths (2, 6, or 11 CH2-
groups) and different mesogenic units (two- or three-ring
mesogens), the monomers M 3 to M 13 were synthesized. The
chiral monomers M 11 to M 13 were synthesized in analogy to
ferroelectric polysiloxanes and may lead to colloids with an
* Corresponding author. Fax: (+49) 6131/39-24778. E-mail:
zentel@uni-mainz.de.
† Institute of Organic Chemistry, University of Mainz.
‡ Department of Physics-Fach M621, University of Konstanz.
§ Current address: Australian National University, Research School of
Chemistry, Canberra, ACT 0200, Australia.
10.1021/ma0613279 CCC: $33.50 © 2006 American Chemical Society
Published on Web 10/31/2006

Macromolecules, Vol. 39, No. 24, 2006
Anisotropic Particles from LC Polymers 8327
Scheme 1. Synthesized Monomers Used in This Work
Table 1. Phase Behavior of the Monomers and Polymers
P 4 is claimed to havesin addition to a smectic and a nematic
phasesa reentrant nematic phase, but without giving any
indication on how this was proven.19,20 We could not find any
evidence for the reentrant phase and assume that the polymer
is just smectic and nematic. The polymers P 3, P 5, P 6, and
P 8 have nematic phases, whereas the polymers P 9 and P 10
show additionally smectic phases. These phases could be
identified by texture observation in agreement with literature
data. For the chiral polymers P 11 to P 13 smectic LC phases
were found, but due to the high viscosity of the samples they
could not be aligned well enough to differentiate between
different phases by texture observation.
We prepared colloids from the synthesized monomers by
dispersion polymerization, which is a special case of a “pre-
cipitation polymerization”.12-15 At the beginning of this process,
the monomer, a steric stabilizer like hydroxypropyl cellulose
(HPC) and a radical initiator like dibenzoyl peroxide (DBPO)
are dissolved in a solvent, in which the LC polymer itself is
insoluble. The radicals formed during thermal decomposition
of the initiator do not only start the polymerization of the
monomer, but also abstract hydrogen from the hydroxylpropyl
cellulose chain, leading to a graft polymerization of LC polymer
onto HPC. This happens until the polymer including the graft-
polymer precipitates and forms a sterically stabilized micelle.
After the formation of the polymer micelles, further polymer-
ization proceeded mostly in the micelles, which grow with time,
leading to sterically stabilized polymer spheres.
The choice of a suitable solvent/nonsolvent system is very
important for this process.12 For most of the monomers we used,
mixtures of ethanol (EtOH) and 2-methoxyethanol (ME) proved
to be suitable. Previous experiments have shown that the size
of the colloids can be tuned best by choice of the solvent,
whereas the amount of initiator and stabilizer only had minor
effects on the particle size. Thus, for these studies we varied
mostly the mixing ratio of the solvents in order to get colloids
of different size and polydispersity. Since the polydispersity of
the obtained colloids may depend on to the polydispersity of
the HPC used, we fractionated HPC (see Experimental Part)
for these experiments. The results of the experiments are
summarized in Table 2 and Figure 1. The size of the colloids
was determined with an optical microscope. Optical microscope
images of the colloids are presented in Figure 2.
monomer phase behavior, °C polymer phase behavior, °C
M 1
M 2
M 3
M 4
M 5
M 6
M 7
M 8
M 9
M 10
M 11
M 12
M 13
c 87 i
c 59 i
c 85 i
c 69 i
c 111 i
c 72 i
P 1
g 32 i
P 217,18 sE,glass 55 sE 113 i
P 319
g 50 n 112 i
P 419,20 g 38 sA 115 n 123 i
P 521,22 g 75 n 110 i
P 621,22 g 33 n 133 i
c 106 n (<180#)
c 92 i
c 53 i
c 96 n 258 i
c 56 i
c 74 i
c1 30 c2 59 s 65 i
P 723
g 90 s 220 i
P 821,22 g 62 n 116 i
P 921,22 g 22 sA 90 n 109 i
P 1021
P 1113
P 12
g 56 k1 110 k2 118 s 138 n 257 i
sX 76 sA/C* 122 i
c 92 lc 107 i
P 1324
sX 56 sC* 158 i
a Key: c ) crystalline, i ) isotropic, g ) glassy, n ) nematic, s )
smectic, sE ) smectic E, sA ) smectic A, sX ) higher-ordered smectic, sC*
) chiral smectic C, lc ) liquid crystalline, and # thermal polymerization
occurred).
especially low symmetry. Such liquid-crystalline colloids from
ferroelectric LC materials with a chiral smectic C* phase are
interesting because they must have a polar axis, if the particle
gets small enough (below the pitch of the superstructure).16
Most of the monomers and their corresponding LC polymers
have already been described in the literature. Therefore, the
phase behavior of the polymers is also already known (see Table
1). As we only had small amounts of the polymers available,
we only checked by DSC measurements and polarizing mi-
croscopy for indications of the same phases. For that, we used
the bulk polymer that we obtained from larger agglomerates
remaining as residue after the centrifugation of the dispersion
polymerization batches. Generally our phase assignment was
consistent with the literature. The phase transition temperatures
were usually slightly reduced, which may be due to the lower
molecular weights and the presence of the steric stabilizer
(graftet cellulose chains) necessary for the dispersion poly-
merization.
Polymer P 1 (not described so far) turned out to be not liquid-
crystalline. In contrast, polymer P 2 is one of the few literature-
known liquid-crystalline polymers with ethylene oxide spacer
that has moderate phase transition temperatures.17,18 According
to literature, it possesses a higher ordered smectic E phase. This
smectic phase could not be verified with our methods. Polymer

8328 Vennes et al.
Macromolecules, Vol. 39, No. 24, 2006
Table 2. Colloid Diameters and Standard Deviations for Dispersion Polymerization with Different Solvent Mixtures and Monomers
d (∆d)/µm
solvent
EtOH
M 1
M 2
M 3
M 4
M 5
M 6
M 7
M 8
M 9
M 11
0.83
(0.28)
a
0.75
(0.12)
0.85
(0.26)
a
0.66
(0.06)
0.76
(0.07)
a
0.74
(0.08)
0.89
(0.18)
1.11
(0.36)
a
a
0.69
(0.05)
0.75
(0.05)
a
a
a
0.78
(0.08)
1.10
(0.18)
a
0.99
(0.09)
0.84
(0.13)
1.13
(0.17)
1.00d
(0.50)
a
EtOH/ME 10:1
EtOH/ME 5:1
EtOH/ME 4:1
EtOH/ME 2:1
EtOH/ME 1:1
EtOH/ME 2:3
EtOH/ME 1:2
EtOH/ME 1:5
EtOH/ME 1:10
ME
0.71
(0.08)
a
a
a
a
0.90
(0.09)
a
a
a
a
a
a
a
a
a
a
a
a
a
a
0.83
(0.06)
1.01
(0.08)
1.16
(0.07)
1.74
(0.75)
3.65
(2.95)
2.4
a
a
a
0.87
(0.15)
1.82
(0.62)
a
1.16
(0.33)
1.52
(0.61)
a
0.76
(0.10)
1.03
(0.29)
a
0.70
(0.08)
0.78
(0.08)
0.83
(0.08)
0.95
(0.10)
0.63
1.21
(0.15)
3.3
(2.7)
a
1.37
(0.30)
a
3.3
(2.7)
a
a
a
a
a
a
a
a
a
a
a
4.5
(3.7)
a
1.98
(0.56)
2.1
(1.1)
a
1.44
(0.58)
a
12
(11)
a
b
a
a
a
(1.1)
a
(0.10)
a
11.5
(10.5)c
a
4.4
(3.6)c
b
b
a
3.6e
f
0.81
(3.1)
(0.11)
f
a Has not been performed. b No colloids. c Slightly turbid. d Additionally agglomerates. e Additionally larger nonspherical objects. Only a few colloids
up to 30 µm.
As a result it was possible to polymerize the monomers M 1
to M 9 and M 11 with EtOH/ME mixtures into sterically
stabilized LC colloids of different size (mostly between 0.5 and
3.5 µm). The graphs in Figure 1 show for all systems a strong
increase in diameter and polydispersity with increasing amount
of ME. This is in agreement with literature.12 ME is thereby
the better solvent as evidenced by the fact that some polymers
(P3, P 5, P 8, and P 9) are soluble in it and do not precipitate
at all. The increase of diameter is about exponential. Concerning
differences between the monomers Figure 1 and Table 2 show
that the monomers containing cyano groups still show colloid
formation at large ME concentrations. The monomers featuring
a C6 spacer can even be polymerized to colloids in pure ME.
For the monomers containing a methoxy group, however, at
least a third of the solvent mixture must be ethanol to yield
colloids.
These results can be rationalized in the following way:12,14,15
Solvent mixtures with a large amount of ethanol are poor
solvents for the newly formed oligomers and polymers. As a
result of the early precipitation of the polymers more particle
nuclei occur and thus more and smaller particles are formed
finally. At the same time secondary nucleation happens rarely.
Thus, a narrow size distribution is obtained for solvent mixtures
rich in ethanol.
In contrast, for better solvent mixtures (more ME) the
polymers precipitate later. This leads to less particle nuclei which
grow to bigger lattices. Additionally, the polydispersity in-
creases. This is likely due to secondary nucleations resulting
from an oversaturated solution and the smaller amounts of
colloids present.
Concerning the different monomers the success of colloid
formation depends on the molecular structure of the monomers
as follows: Large apolar hydrocarbon segments seem to be
hardly compatible with this type of dispersion polymerization.
From the monomers with C11 spacers (M 11-M 13) only one
can be converted into colloids; from the three monomers with
three benzene rings in the mesogenic structure (M 7, M 10,
and M 13), again only one (M 7) yields colloids. This one has
a highly polar cyano group directly linked at the mesogen. In
this context it must be noted that for the monomers M 10,
M 12, and M 13 that did not lead to colloids (or only to few
Figure 1. Particle diameter vs mole fraction (ME) for M 3 to M 9.
colloids and agglomerates) for EtOH/ME mixtures, other

Macromolecules, Vol. 39, No. 24, 2006
Anisotropic Particles from LC Polymers 8329
Figure 2. Optical microscope images of colloids of different polymers (top, broader size distribution; top left, P 4 (EtOH/ME 1:5); top middle,
P 6 (EtOH/ME 1:2); top right, P 9 (EtOH/ME 1:1); bottom, narrow size distribution; bottom left, P 7 (EtOH/ME 1:1); bottom middle, P 3 (EtOH/
ME 2:1); bottom right, P 6 (EtOHME 2:1). The pictures were taken either on sedimented mono- and multilayers (bottom right and middle) or on
floating colloids (bottom left and top).
Figure 3. Optical microscope images of colloids prepared by “normal” dispersion polymerization and seed polymerization (left, P 6 (EtOH/ME
1:1; diameter ) 1.16 ( 0.07); middle left, P 6 (seed polymerization in EtOH/ME 1:1; diameter ) 1.57 ( 0.12 µm); middle right, P 11 (EtOH;
diameter ) 0.99 ( 0.09 µm); right, P 11 (seed polymerization in EtOH; diameter ) 1.25 ( 0.12 µm).
solvents like mixtures of ME or EtOH and THF were also tried.
However, again no or only a few single colloids and other solid
pieces, agglomerates etc. were obtained. Concerning the size
of the colloids, the comparison of similar monomers with C2
spacers and C6 spacers shows that for a given solvent mixture
(at least for solvent mixtures where the colloids still have a
narrow distribution) the colloids with C6 spacers are a bit bigger.
However, in most cases no colloids were obtained (M 10 and
M 13).
As a result of these experiments it is possible to make colloids
with diameters mostly between about 0.5 to 3.5 µm, whereas
bigger colloids were only obtained in limited numbers together
with smaller colloids in samples with a very broad size
distribution. In addition the large colloids are only accessible
for some monomers. To get bigger colloids with a narrow
distribution, a seed polymerization with the monomers M 6 and
M 11 was developed. For this purpose solvent mixtures that
still lead to colloids with a narrow size distribution (EtOH/ME
1:1 for M 6, pure EtOH for M 11) were used. The seed
polymerization was started like the other dispersion polymeriza-
tions, but additionally every 30 min a solution of monomer in
the used solvent mixture was added (in total nine times, see
Experimental Part). As a result it can be shown that seed
polymerization is a general method to increase the diameter of
the LC colloids, but a lot of monomer is needed. In theory
doubling the amount of monomer should increase the radius
by approximately 25% (assuming a doubling of mass). The
experiments showed a radius increase less than expected. For
monomer M 6 a quadruplication of the amount of monomer
increases the diameter from 1.16 ( 0.07 to 1.57 ( 0.12 µm
(theroretically 1.84 µm), whereas for monomer M 11 the
quadruplication leads to a diameter of 1.25 ( 0.12 µm
(theoretically 1.57 µm) instead of 0.99 ( 0.09 µm. In both cases
the big colloidssnot accessible by direct polymerizationsstill
have a narrow size distribution of about or less than 10% (see
Figure 3).
In a next step, the colloid suspensions were investigated by
polarizing microscopy. All colloids except the ones from
monomer M 1 show evidence of birefringence when viewed
between crossed polarizers, whereupon the colloids of the
monomers with C2 spacers (M 3, M 5, M 8) seem a bit less
bright/birefringent than the others. This may be a result of their
higher Tg-values, which make it more difficult to establish the
LC phase and to reach the stable director configuration. For
small diameters up to about 1.8 µm the colloids show a uniform
birefringence; they appear as spots whose brightness fluctuate

8330 Vennes et al.
Macromolecules, Vol. 39, No. 24, 2006
Figure 4. Polarizing microscope images of colloids of P 6 (EtOH/ME 2:3) (sample holder was rotated).
when they rotate due to Brownian motion (see movie 1 in the
Supporting Information). The optical resolution of the micro-
scope is, however, too low to characterize their director pattern
in more detail. As soon as the particles get big enough (diameter
approximately 2-4 µm), maltese cross and “baseball” type
patterns can be found. Figure 4 shows polarizing microscope
images of a dried sample of colloids of P 6. The sample holder
was rotated; thereby the texture changes from a maltese cross
to a “baseball” type pattern. This indicates that the colloids of
intermediate size have a bipolar director configuration (see
Figure 3). Larger colloids (> 4 µm) show usually a more
complex director pattern.
The prepared colloids are generally suitable for use in optical
tweezers. For our experiments we used a standard optical
tweezers setup with a circularly polarized trapping beam. While
it was very difficult to trap the small colloids (below 1.6 µm)
for any significant amount of time, colloids of an intermediate
size (1.8-2.5 µm) could be trapped with ease and held for longer
times. However, much larger colloids present a problem in that
they sediment rather quickly, resulting in undesired effects such
as sticking to the sample surface. When trapped, the spheres
could be typically held in the tweezer for several minutes before
escaping due to thermal fluctuations. This allows various
manipulations and investigations. When the birefringent colloids
are trapped under circularly polarized light, a torque T is
transferred to the anisotropic particle and it rotates. This rotation
force is countered by the Stokes drag, resulting in a constant
rotation frequency f given by
the Supporting Information). This shows the potential of the
anisotropic colloids as actuators.
Rotation of the colloids in the laser field allows it to study
the quality of the LC ordering (director configuration) and its
dependence on the size of the colloids. Generally, the rotation
speed of the colloids decreases as the size increases as predicted
by eq I for a constant torque T. The efficiency of the torque
transfer is dependent on the anisotropy of the colloids and thus
the director configuration. To investigate if the anisotropy/
birefringence of the colloids changes with size and to determine
the size limit at which the colloids get polycrystalline, we
measured the rotational frequency for colloids of different
diameter and plotted log(f) vs log(r) (see Figure 5). If the
f ) T/(8πnr3)
(I)
Figure 5. Dependence of log(f) on log(r) for colloids of P 4, P 6, and
P 9. Also displayed as a guide are lines of exponent γ ) -3 relating
to the constant torque assumption of eq I.
where r ) radius and n ) solvent viscosity, in this case for
water.
Viewed between crossed polarizers the particles director field
depolarizes the light, and, provided there is a bipolar director
field in the particle, appears as a spot whose brightness is
dependent on the angle at which it is aligned between the crossed
polarizers, resulting in an intensity fluctuation, I ) sin2(2θ).
For particles that are big enough (diameter above 2 µm) a change
of the optical texture during rotation can also be observed.
However slightly defocusing the camera lens can blur out the
detail somewhat, and we again see an intensity relationship
I ) sin2(2θ).
For our experiments we trapped and rotated anisotropic
colloids of P 4, P 6, P 9, and P 11 with different sizes between
0.7 and 3.7 µm at a constant laser power (see movie 2 in the
Supporting Information). By changing the position of the
quarter-waveplate (rotation of 90° or turning around 180°) and
thus the rotational sense of the polarized light, the direction of
the rotation of the colloids could be changed between clockwise
and anticlockwise. One of the films shows a trapped colloid
that picks-up another colloid to orbit around it (see movie 3 in
anisotropy does not change, this plot should give a straight line
with a slope of -3 and a y intercept of log(T/(8πn). In contrast,
if there is any change in the torque with particle size, there is
a deviation from a straight line on the graph. Thus, it is possible
to detect changes in the director pattern below the limit of the
optical resolution.
For various colloids of P 6 ranging in size from 1.6 to 3.5
µm we found a slope of -3 over nearly the full size range
studied (see Figure 5). Thus, torque and anisotropy/birefringence
remain constant suggesting the particles do not change their
director pattern in the size range measured. Only colloids of
diameters above 3.4 µm start to deviate a little bit from the
graph, indicating that the birefringence starts to decrease. Small
colloids of P 6 with 0.7 µm diameter are beginning to approach
the size of the trapping laser focus (∼0.5 µm) at which point
assumptions made in the formulation of eq I begin to become
invalid. As a result, rotation frequencies of these smaller particles
are not included for this particular analysis.

Macromolecules, Vol. 39, No. 24, 2006
Anisotropic Particles from LC Polymers 8331
4-[2-(2-Acryloyloxyethoxy)ethoxy]benzoic Acid. A mixture of
5.677 g (25.1 mmol) of 4-[2-(2-hydroxyethoxy)ethoxy]benzoic acid,
17.1 mL (249 mmol) of acrylic acid, 1.1 g of p-toluenesulfonic
acid, 1.01 g of 2,6-di-tert-butyl-4-methylphenol, and 40 mL of
chloroform was refluxed at the water separator until no more water
was separated. After cooling, the reaction mixture was diluted with
chloroform. A part of the product was insoluble and could be
isolated by filtration. The solution was extracted with water and
dried over sodium sulfate, and the solvent was removed under
reduced pressure. The crude product was purified by column
chromatography (chloroform).
For colloids of P 4 or P 9, however, the measured points do
not all fit on a straight line (see Figure 5). For colloids of P 4
starting from small spheres of about 1.8 µm the slope of the
graph begins to decrease more sharply with particle size,
suggesting for eq I that the absorbed torque starts to decrease
greater than predicted with increasing size. It is apparent from
this that the anisotropy/birefringence decreases with increasing
size of the colloids. For colloids of P 9 a similar trend is
observed to a much greater extent. Colloids of P 11 generally
rotate very slowly, probably because they are less birefringent
than the other colloids.
Finally it is interesting to note that colloids of the polymer
P 6 are nematic and follow the simple theory for size
dependence, which indicates that their anisotropy remains
constant. In contrast, for colloids of the polymers P 4 and P 9,
which are smectic, the anisotropy changes with size. This may
be a result of competing orientational tendencies in smectic
colloids with curved or straight smectic layers, as observed for
much smaller smectic colloids by TEM measurements10 (see
Supporting Information). This effect would of course not arise
with a nematic liquid crystal structure.
Yield: 3.97 g (57%) of a colorless solid, mp 129 °C.
FT-IR (ATR): ν˜ ) 2956, 2894 (C-H), 2658, 2561 (O-H),
1724, 1685 (CdO), 1605 (CdC), 1409, 1255, 1169, 1133, 1057,
809, 772.
1H NMR (DMSO-d6): δ ) 12.61 (bs, 1H, -COOH); 7.86 (d,
2H, Ar-H, ortho to -COOH); 7.01 (d, 2H, Ar-H, meta to
2
3
-COOH); 6.32 (dd, 1H, CH2dCH- (trans), J ) 1.9 Hz, J )
3
3
17.3 Hz); 6.17 (dd, 1H, CH2dCH-, J ) 17.3 Hz, J ) 9.9 Hz);
5.93 (dd, 1H, CH2dCH- (cis), 2J ) 1.9 Hz, 3J ) 9.9 Hz);4.24 (t,
2H, -CH2O-); 4.16 (t, 2H, -CH2O-); 3.77 (t, 2H, -CH2O-);
3.70 (t, 2H, -CH2O-).
4-[2-(2-Acryloyloxyethoxy)ethoxy]benzoic Acid 4-Cyanophe-
nyl Ester (M 1). A solution of 1.497 g (5.341 mmol) of 4-[2-(2-
acryloyloxyethoxy)ethoxy]benzoic acid, a pinch of 2,6-di-tert-butyl-
4-methylphenol, 5 mL of thionyl chloride and a droplet of DMF
was stirred at room temperature for 40 min. Afterwards the excess
thionyl chloride was removed under reduced pressure. The residue
was suspended in 6 mL of THF and the suspension was cooled to
0 °C. A solution of 640.8 mg (5.379 mmol) of 4-cyanophenol and
1.1 mL of triethylamine in 12 mL of THF was added dropwise at
0 °C. The mixture was stirred for 20 h at room temperature. Then
the solvent was evaporated and the residue was dissolved in 12
mL of methylene chloride. The organic phase was extracted several
times with water and dried over sodium sulfate, and the solvent
was removed under reduced pressure. The crude product was
purified by column chromatography (hexane/ethyl acetate 2:1).
Yield: 1.05 g (52%) of colorless needles, mp 87 °C.
MS (m/z): 381.3 (M+).
Conclusion
Dispersion polymerization allows the conversion of many LC
acrylate monomers directly into spherical, but optically aniso-
tropic colloids possessing nematic or various smectic phases at
ambient temperature. These colloids in most instances have a
bipolar director configuration. It is additionally possible with
this method to vary the size of the colloids and to obtain a
relatively low size distribution. Because of their optical aniso-
tropy the particles can be rotated with circularly polarized light.
This makes them attractive as actuators to rotate other colloids.
Experimental Part
Commercial chemicals were used without further purification.
The hydroxypropyl cellulose was fractionated according to the
literature25 and the fraction with Mn ) 82 020 and Mw ) 138 000
g/mol was used for the dispersion polymerization. NMR spectra
were recorded on a Bruker 300 MHz spectrometer. FT-IR spectra
were measured on a Bruker Vector 22 FTIR spectrometer. Mass
spectroscopy was performed on a Finnigan MAT 90 spectrometer.
Polarizing microscopy was performed with a Zeiss Jenapol SL 100-
microscope. DSC measurements were performed with a DSC 7 of
Perkin-Elmer.
(C21H19NO6)n (381.4)n Calcd C 66.13, H 5.02, N 3.67; Found C
66.17, H 4.94, N 3.70.
FT-IR (ATR): ν˜ ) 2963, 2956, 2892 (C-H), 2230 (CtN), 1739,
1726 (CdO), 1603 (CdC), 1498, 1252, 1200, 1162, 1130, 1051,
844, 757.
1H NMR (CDCl3): δ ) 8.11 (d, 2H, Ar-H, ortho to -COO-);
7.72 (d, 2H, Ar-H, ortho to -CN); 7.33 (d, 2H, Ar-H, meta to
-CN); 7.00 (d, 2H, Ar-H, meta to -COO-); 6.42 (dd, 1H, CH2d
2
3
CH- (trans), J ) 1.5 Hz, J ) 17.3 Hz); 6.14 (dd, 1H, CH2d
3
3
Synthesis of the Monomers. The monomers were synthesized
according to literature or standard procedures for etherifications
and esterifications (for references, see Table 1).
CH-, J ) 17.3 Hz, J ) 10.3 Hz); 5.83 (dd, 1H, CH2dCH-
(cis), 2J ) 1.5 Hz, 3J ) 10.3 Hz); 4.34 (t, 2H, -CH2O-); 4.21 (t,
2H, -CH2O-); 3.89 (t, 2H, -CH2O-); 3.81 (t, 2H, -CH2O-).
13C NMR (CDCl3): δ ) 166.1 (1C, CH2CHCOO-); 163.9 (1C,
-COO-); 163.5 (1C, Ar-C, ipso to -OCH2-); 154.4 (1C, Ar-
C, para to CN); 133.7 + 132.5 (4C, Ar-C, ortho to CN, meta to
OCH2); 131.2 (1C, CH2CH-); 128.2 (1C, CH2CH-); 123.0 +
121.0 (3C, Ar-C, para to OCH2, meta to CN); 118.3 (1C, -CN);
114.6 (2C, Ar-C, ortho to OCH2); 109.6 (1C, Ar-C, ipso to CN);
69.4 + 67.7 + 63.5 (4C, -CH2-).
4-[2-(2-Hydroxyethoxy)ethoxy]benzoic Acid. A solution of
44.55 g (794 mmol) of potassium hydroxide, 50.09 g (363 mmol)
of 4-hydroxybenzoic acid, a pinch of potassium iodide, 60 mL of
water, and 140 mL of ethanol was refluxed, and 42 mL (398 mmol)
of 2-(2-chloroethoxy)ethanol was slowly added dropwise. After that
the solution was refluxed for 2 days. The solvent was removed
under reduced pressure and the residue was dissolved in 250 mL
of water. Impurities were removed by extracting the aqueous phase
with diethyl ether. Then 6 N hydrochloric acid was added to the
aqueous phase at 50 °C. The precipitate formed during cooling to
0 °C was isolated by filtration and recrystallized from diethyl ether/
acetone 3:1 or ethyl acetate.
Yield: 33.23 g (40.5%) of colorless needles, mp 118 °C.
FT-IR (ATR): ν˜ ) 3338 (O-H), 2939, 2887 (C-H), 2663, 2552
(O-H), 1677 (CdO), 1605 (CdC), 1579, 1428, 1254, 1171, 1130,
1051, 939, 772, 649.
1H NMR (DMSO-d6): δ ) 12.54 (bs, 1H, -COOH); 7.87 (d,
2H, Ar-H, ortho to -COOH); 7.02 (d, 2H, Ar-H, meta to -COOH);
4.63 (bs, 1H, -OH); 4.16 (t, 2H, -CH2O-); 3.74 (t, 2H,
-CH2O-); 3.49 (m, 4H, -CH2O-).
(S)-2-Chloro-4-methylpentanoic Acid 4′-(11-Acryloyloxyun-
decyloxy)biphenyl-4-yl Ester (M 12). Monomer M 12 was
synthesized in analogy to monomer M 11.13
Yield: 72% of a colorless solid, mp 74 °C.
20
[R]D ) - 12.03° in CHCl3 (c ) 52.35 g L-1).
MS (m/z): 542.3 (M+).
Anal. (C32H43ClO5)n (543.1)n Calcd: C, 70.76; H, 7.98. Found:
C, 70.90; H, 7.97.
FT-IR (ATR): ν˜ ) 2959, 2933, 2921, 2852 (C-H), 1752, 1724
(CdO), 1606 (CdC), 1497, 1473, 1410, 1282, 1205, 1170, 1039,
836, 805 cm-1
.
1H NMR (CDCl3): δ ) 7.54 (d, 2H, Ar-H, meta to OOC, 3J )
8.8 Hz); 7.46 (d, 2H, Ar-H, meta to OCH2, 3J ) 8.5 Hz); 7.14 (d,

8332 Vennes et al.
Macromolecules, Vol. 39, No. 24, 2006
2H, Ar-H, ortho to OOC, 3J ) 8.8 Hz); 6.94 (d, 2H, Ar-H, ortho
to OCH2, 3J ) 8.5 Hz); 6.38 (dd, 1H, CH2dCH- (trans), 2J ) 1.5
Hz, 3J ) 17.5 Hz); 6.10 (dd, 1H, CH2dCH-, 3J ) 17.5 Hz, 3J )
10.3 Hz); 5.79 (dd, 1H, CH2dCH- (cis), J ) 1.5 Hz, J ) 10.3
Hz); 4.53 (t, 1H, -CHCl-); 4.13 (t, 2H, -COOCH2-); 3.97 (t,
2H, -OCH2-); 1.98 (m, 3H, -CHCH3-, -CH2CH3-); 1.81-
1.28 (m, 18H, -CH2-); 1.01 (m, 6H, -CH3).
Table 3. Diameters, Frequencies, log(r) and log(f) for Colloids of
P 4, P 6, and P 9
colloid
d [µm]
f [Hz]
log(r)
log(f)
2
3
C 4-1
C 4-2
C 4-3
C 4-4
C 4-5
C 4-6
C 4-7
C 4-8
C 4-9
C 4-10
C 4-11
C 6-1
C 6-2
C 6-3
C 6-4
C 6-5
C 6-6
C 6-7
C 6-8
C 6-9
C 6-10
C 6-11
C 9-1
C 9-2
C 9-3
C 9-4
C 9-5
C 9-6
C 9-7
C 9-8
C 9-9
C 9-10
C 9-11
1.8
1.8
1.9
2.0
2.1
2.1
2.2
2.3
2.3
2.7
3.4
0.7
1.6
2.1
2.6
2.8
2.8
3.3
3.3
3.4
3.5
3.5
0.9
1.4
1.6
2.2
2.5
2.8
2.8
2.9
3.0
3.0
3.3
1.52
1.63
1.43
1.30
1.12
1.33
1.15
0.93
1.08
0.48
0.16
3.75
0.92
0.28
0.18
0.14
0.13
0.10
0.10
0.06
0.06
0.05
1.23
0.81
0.83
1.18
1.02
0.59
0.59
0.27
0.21
0.21
0.12
-0.046
-0.046
-0.022
0
0.180
0.211
0.154
0.114
0.047
0.125
0.060
-0.034
0.033
0.021
0.021
0.041
0.061
0.061
0.130
0.230
-0.456
-0.097
0.021
0.114
0.146
0.146
0.217
0.217
0.230
0.243
0.243
-0.347
-0.155
-0.097
0.041
0.097
0.146
0.146
0.161
0.176
0.176
0.217
Synthesis of the Colloids. The dispersion polymerization of the
monomers was always performed by the same procedure.
In a typical process 50 mg of monomer and 5 mg of the
fractionated hydroxypropyl cellulose were degassed for about 20
min with nitrogen in a 10 mL test tube that was closed with a
septum. Then 0.8 mL of the degassed solvent mixture was added,
and the reaction mixture was heated under stirring (700 rpm) to 75
°C. About 30 min later a degassed solution of 5 mg of dibenzoyl
peroxide in 0.3 mL of the solution mixture that had been also heated
to 75 °C before was added to start the polymerization. After a few
minutes, the solution turned turbid. The reaction was allowed to
proceed for 24 h and then was ended by opening the test tube and
cooling.
One drop of the suspension was used without further purification
for the determination of the colloidal particle size with an optical
microscope.
For purification the suspension was shortly (approximately 30
s) centrifuged with 2000 rpm and the sediment was isolated. The
suspension was further centrifuged, the supernatant was decanted
and the colloidal particles were redispersed in ethanol (independent
of the used solvent mixture). This procedure was repeated one to
two times.
-0.321
-0.805
0.574
-0.036
-0.547
-0.756
-0.859
-0.899
-1.010
-1.003
-1.208
-1.247
-1.292
0.090
-0.091
-0.083
0.072
0.010
-0.229
-0.229
-0.572
-0.675
-0.670
-0.936
The used solvent mixtures for the different polymerization
batches are shown in Table 2.
Synthesis of Colloids by Seed Polymerization. First, 75 mg of
monomer M 6 and 7.5 mg of HPC were degassed for about 30
min with nitrogen. Then 1.0 mL of a degassed mixture of EtOH/
ME 1:1 was added, and the reaction mixture was heated under
stirring (700 rpm) to 75 °C. After 45 min, a degassed solution of
7 mg dibenzoyl peroxide in 0.6 mL of the solution mixture that
had been also heated to 75 °C before was added to start the
polymerization. Every 30 min, 0.3 mL of a solution of 250 mg of
M 6 in 3 mL of solution mixture was added (in total nine times).
The reaction was ended after 24 h and purified by centrifugation
as described above.
Supporting Information Available: Movie 1 showing a
dispersion of colloids of P 6 between crossed polarizers. The
colloids appear as spots whose brightness fluctuates when they
rotate due to Brownian motion, movie 2 showing a 3.3 µm colloid
of P 6 (C 6-8) that was trapped and rotated with an optical tweezers
with circularly polarized light, and movie 3 showing a trapped
colloid of P 9 that picks up another colloid to orbit around it, which
shows the potential of the anisotropic colloids as actuators, and a
TEM image showing small smectic colloids (about 50-150 nm)
of a “diluted” polysiloxane (see ref 10) with curved and straight
smectic layers, where the contrast is due to the silicon in the polymer
backbone. This material is available free of charge via the Internet
at http://pubs.acs.org.
The seed polymerization of monomer M 11 was performed the
same way, but pure ethanol was used.
Experiments with Optical Tweezers. We used the same optical
tweezers setup as in Sandomirski et al.6 and Martin et al.26 with a
trapping produced by a λ ) 532 nm Verdi-V2 frequency-doubled
Nd:YAG-Laser (Verdi-V2, Coherent Inc.) and a imaging with a
CCD camera (HighSpeedStar1, LaVision GmbH). The samples
were prepared as follows: The colloid dispersion was highly diluted
(volume fraction <10-2) with a 10 mM aqueous SDS salt solution
to prevent particle aggregation. A few drops of this solution were
placed in an approximately 170 µm thick sample cell, which was
hermetically sealed with UV curing glue. The sample was then
placed in the optical tweezers and the colloids were trapped at a
distance of about 10-20 µm above the bottom cell surface.
The trapping beam was passed through a quarter-wave plate to
produce a circularly polarized beam which produced the torque
which was transferred to the particle. When a colloid was trapped,
a focused unpolarized image was first taken to determine its
diameter. Then the imaging polarizers were crossed, and ap-
proximately 10 000 images were taken at a speed of 30 frames per
second. The average time it took for the colloid to rotate 360° was
determined by measuring the intensity fluctuations of the particle
in the images which follows I ) sin2(2θ). The results for colloids
of the polymers P 4, P 6, and P 9 are shown in Table 3.
References and Notes
(1) Garces-Chavez, V.; McGloin, D.; Summers, M. D.; Fernandez-Nieves,
A.; Spalding, G. C.; Cristobal, G.; Dholakia, K. J. Opt. A: Pure Appl.
Opt. 2004, 6, S235-238.
(2) Bishop, A. I.; Nieminen, T. A.; Heckenberg, N. R.; Rubinsztein-
Dunlop, H. Phys. ReV. Lett. 2004, 92, 198104-1-198104-4.
(3) Cairns, D. R.; Sibulkin, M.; Crawford, G. P. Appl. Phys. Lett. 2001,
78, 2643-2645.
(4) Amimori, I.; Priezjev, N. V.; Pelcovits, R. A.; Crawford, G. P. J. Appl.
Phys. 2003, 93, 3248-3252.
(5) Ferna´ndez-Nieves, A. Soft Matter 2006, 2, 105-108.
(6) Sandomirski, K.; Martin, S.; Maret, G.; Stark, H.; Gisler, T. J. Phys.:
Condens. Matter 2004, 16, S4137-S4144.
(7) Umbanhowar, B. P.; Prasad, V.; Weitz, D. A. Langmuir 2000, 16,
347-351.
(8) D´ıaz-Leyva, P.; Pe´rez, E.; Arauz-Lara, J. L. J. Chem. Phys. 2004,
121 (18), 9103-9110.
(9) Ferna´ndez-Nieves, A.; Cristobal, G.; Garce´s-Cha´vez, V.; Spalding,
G. C.; Dholakia, K.; Weitz, D. A. AdV. Mater. 2005, 17, 680-684.
(10) Vennes, M.; Zentel, R.; Ro¨ssle, M.; Stepputat, M.; Kolb, U. AdV.
Mater. 2005, 17, 2123-2127.
(11) Yang, Z.; Huck, W. T. S.; Clarke, S. M.; Tajbakhsh, A. R.; Terentjev,
E. M. Nature mater. 2005, 4, 486-490.
(12) Lok, K. P.; Ober, C. K. Can. J. Chem. 1985, 63, 209-216.
(13) Vennes, M.; Zentel, R. Macromol. Chem. Phys. 2004, 205, 2303-
2311.
Acknowledgment. The authors thank the DFG for the
support of the work.

Macromolecules, Vol. 39, No. 24, 2006
Anisotropic Particles from LC Polymers 8333
(14) Arshady, R. Colloid Polym. Sci. 1992, 270, 717-732.
(15) Kawaguchi, S.; Ito, K. AdV. Polym. Sci. 2005, 175, 299-328.
(16) Goodby, J. W. Ferroelectric Liquid Crystals; Gordon and Breach
Science Publishers: Philadelphia, PA, 1991.
(17) Chovino, C.; Guillon, D.; Gramain, P. Polymer 1998, 39, 6385-6390.
(18) Duran, R.: Gramain, P. Makromol. Chem. 1987, 188, 2001-2009.
(19) Shibaev, V. P.; Kostromin, S. G.; Plate, N. A. Eur. Polym. J. 1982,
18, 651-659.
(20) Je´roˆme, B.; Commandeur, J.; De Jeu, W. H. Liq. Cryst. 1997, 22,
685-692.
(21) Portugall, M.; Ringsdorf, H.; Zentel, R. Makromol. Chem. 1982, 183,
2311-2321.
(22) Zentel, R.; Strobl, G. R.; Ringsdorf, H. Macromolecules 1985, 18, 8,
960-965.
(23) Imanishi, Y.; Kondo, K.; Kitamura, T. Jpn. J. Appl. Phys. 1990, 29,
L1477-L1479.
(24) Skarp, K.; Andersson, G.; Lagerwall, S. T.; Kapitza, H.; Poths, H.;
Zentel, R. Ferroelectrics 1991, 122, 127-137.
(25) Wirick, M. G.; Waldman, M. H. J. Appl. Polym. Sci. 1970, 14, 579-
597.
(26) Martin, S.; Reichert, M.; Stark, H.; Gisler, T. Phys. ReV. Lett. 2006,
submitted.
MA0613279