Samples were prepared by dissolving the diacrylate monomer BAB-6
(4,4
[6-acyloyloxy)-hexyloxy]-1,1-biphenylene) and the
photoinitiator BME (benzoin
methyl ether) in a 20:1 ratio by weight in the nematic liquid crystal 5CB
(4-pentyl-4-cyanobiphenyl). The materials were mixed well and
then filled into cells
consisting of two parallel glass plates with a 10-
m separation. The
inner faces of the glass
plates were coated with polyimide in order to induce a homogeneous
alignment at these surfaces. After
the cells had been filled they were then irradiated for 30 minutes with UV
light at 18 mW/cm
. The
resulting polymer network was studied by means of birefringence
measurements and also directly by
scanning electron microscopy. The microscopy study appeared to show that a
polymer network
consisting of fibrils of submicron thickness had been formed with an
alignment more or less in the
direction of the LC director [2]. This result must be treated
cautiously, however, as it
is possible that the network may have compacted when the liquid crystal was
extracted. This
possibility is given credence by the fact that the birefringence studies indicated a much smallerfibril diameter, perhaps of the order of a few nanometers.
The nematic-isotropic transition was studied in this system by measurements
of the optical
birefringence. A 2mm-wide light beam from a 5 mW He-Ne laser was directed
at a cell contained
between crossed polarizers, with the liquid crystal director at an angle of
45
to the
polarizer transmission axis.
The birefringence 
was found from measurements of the intensity
I of transmitted light
through the relationship 
with
the wavelength of
light and d the cell thickness. The temperature was varied at a rate less
then 0.1 mK/minute, and
was controlled with a precision of 5 mK.
The results are shown in Fig. 1. In the absence of polymer, the transition
from the nematic to the
isotropic phase was seen as an abrupt vanishing of the birefringence when
the temperature was raised
above 35C. As
the polymer concentration was increased, the birefringence, which is
proportional to the LC order
parameter, decreased in the nematic phase, and the transition was shifted
to lower temperatures. At
temperatures above the transition temperature some residual order remained
in the liquid crystal,
presumably as a result of the anisotropic environment caused by the
presence of the polymer.
When the concentration of polymer reached a point somewhat above 10%, the first-order phase transition was no longer apparent. Instead, the birefringence decreased continually from its low-temperature value as the temperature was raised. The most striking aspect of these results was that there was an appreciable range over which the birefringence decreased linearly with temperature. The possibility of there being two continuous phase transitions at the boundaries of the linear region could not be excluded by our measurements.