Studies of the influence of polymer networks on devices which employ nematic materials include twisted nematic (TN) and supertwisted nematic (STN) liquid crystal devices (LCD). One such investigation by Bos, et al, is reported in Chap 13, Crawford & Zumer, eds., 1996. They found that, in TN cells, a significant reduction in operational voltage can be achieved at low polymer concentration and, in STN cells, the undesirable striping texture can be eliminated and a reduction in driving voltage achieved.
In these and other applications of devices based on polymer stabilization, it is important to understand the role of the polymer network and its morphology as well as the factors controlling it. Another group of ALCOM researchers, Hudson, Rajaram, and Chien (Rajaram, 1996) made use of nematic liquid crystal materials in an important study of the effect of monomer structure and temperature of photopolymerization on network morphology. They also developed a kinetic model of the network formation as described and illustrated below.
PSLC cells are generally prepared by dissolving and photopolymerizing monomers (typically less than 5 wt%) in a liquid crystals matrix to form a polymer network. Hudson, et al. (Rajaram, 1996), chose to study network formation in a nematic liquid crystal matrix (solvent) of planar (homogeneous) aligned nematic liquid crystal as well as in an isotropic environment. The homogeneous alignment was provided by pretreatment--coating and rubbing--of the inner side of the bounding glass cell faces. The particular sample solutions of interest in this brief summary contain 3 wt% of either the diacrylate monomer BAB or BAB6 with 0.3 wt% photoinitiator BME dissolved in E48, a eutectic mix of several similar low-molar mass liquid crystals.
After being sandwiched between the treated glass cell faces, the solution was photopolymerized under an UV light source. Here, polymer phase separation and network formation take place. In some studies the liquid crystal matrix was then dissolved by placing the cell in hexane (2 days). Finally, the cell was carefully split open to permit study of the bare polymer network using the scanning electron microscope (SEM). It has recently been discovered that normal evaporation of hexane or ethanol leads to tearing of the network, while a more faithful representation of both small scale (nodular clusters) and large scale (network) structure is preserved by super critical drying in CO2.
This intriguing large scale structure reveals regions of the specimen that are rich and poor in polymer concentration after phase separation has occurred. Fig(a) illustrates the regularity of this large scale structure which has a characteristic dimension between 5 and 30 micrometers, depending on polymerization condition and monomer concentration. Fig(b)—a high magnification SEM photo—illustrates the presence of nodular clusters in the varying density network where it seems apparent from the inhomogeneity of the distribution of polymer particles that the large scale structure is related to the aggregation process in these systems.
|
Figure (a) : SEM micrograph showing the large scale variation in network density. Low network density areas appear near the bottom and again near the top of the image with much higher density in between. The molecular director of the nematic solvent during polymerization was in the vertical direction. |
 |
Figure (b) : A high magnification SEM image showing the fine scale structure. Note the nodular clusters that compose the aggregate network structure. This structure is unoriented but other polymerization conditions can yield oriented fibrils. |
Returning to the small scale polymer structure, Hudson, et al., attributed the beadlike morphology formed by the photopolymerization of BAB to rigidity due to the lack of flexible spacers between the rigid core of the monomer and the polymerizable moiety. This created a loose network with nodular and aggregated beads having pores ranging in size and shape. The primary cluster size depended significantly on polymerization temperature. A fiber-like morphology, such as that illustrated in Fig (c), is also frequently observed.
 |
Figure (c) : A high magnification SEM image showing the fibrous network morphology |
The changes in the network can be clarified and explained based on upon a kinetic model proposed by Hudson, et al., in which the photopolymerization of uniformly distributed monomers in a liquid crystal solution initiates a reaction-limited process which eventually becomes diffusion-limited in later stages of network formation. This is illustrated by the following movie:
The movie begins with a close-up look at the reaction-limited process in a small region where, in the early stages of the UV photopolymerization, monomers in the vicinity migrate to feed the polymerization reaction. Small primary particles are formed as the polymer molecules reach a sufficient molecular weight and/or branch content to phase separate. It is these particles which are first illustrated, moving with random motion and direction.
Because the primary particles contain very few active free radical species on their surfaces, their sticking probability is relatively low so that they frequently bounce off each other. As these primary clusters grow in size, cluster diffusion eventually becomes rate limiting. In the video, the nodular clusters are now replaced by larger circular symbols, for ease in subsequent representation. Next, the video zooms out to show a section of the entire network. Larger clusters diffuse more slowly. More importantly, as the neighboring regions are depleted, the clusters must diffuse farther to contact others. Under these conditions and because they contain more active surface sites their sticking probability increases to essentially one. Thus, their reactions become diffusion limited as they randomly and slowly aggregate to form the resultant network morphology.
Based on the above model, the size of the nodular primary clusters depends on the condition of crossover from reaction-limited to diffusion-limited growth and, thus, on the polymerization conditions which depend on UV exposure and photoinitiator concentration. Experimental results are consistent with this expectation.
Further studies by Hudson have illustrated the phase separation and aggregation process more clearly, as seen in the following series of images, obtained with phase contrast optical microscopy (which is sensitive to difference in refractive index). The effects of varying initiator concentration and UV photopolymerization time for the process are displayed for a fixed monomer concentration of 2 wt% BAB in toluene, beginning with a uniform solution, moving through phase separation of tiny mobile reaction units and finally aggregation.
 |
Four phase-contrast optical micrograph images of the same region at various stages of polymerization of a 2 wt% BAB in toluene after A) 8, B) 26, C) 28, and D) 46 s of UV irradiation, respectively from top to bottom. The aggregating polymer beads appear dark. A gradient in the initiator concentration was established, so that different stages of polymerization could be examined simultaneously. Where the concentration of initiator was greatest (at the left side of the figures), the polymerization was completed first and the polymer aggregates are fixed. At the right side of the figures, the initial uniform appearance of the solution is shown. At any given point of the specimen, its appearance proceeds from uniform gray, to small mobile aggregates, and to larger aggregates that ultimately are fixed in a gel network (The height of each figure is 200 microns). |
| The formation of the network from a uniform solution (i.e. without an initiator concentration gradient) can be examined in situ as a function of UV photopolymerization time, as illustrated by the animation to the left where a solution of 2 wt% BAB in toluene with 2 wt% initiator at 20° C is examined by optical phase contrast microscopy at 4 different exposure times: 0, 7, 8.5, and 11 seconds, and in the same area. The height of each image is 200 microns. The first frame shows the sample that has received no radiation. Click on the arrow in the lower right corner to advance to the next frame, showing the effect of 7 seconds of photopolymerization, then to 8.5 seconds and 11 seconds. The polymerization must proceed for some time before the small mobile clusters become visible at 7 seconds. The aggregation process then proceeds rapidly, and in the last frame the polymer network is evident. |
Another informative study by Hudson provides insight on the control of the size of the large scale structure of the polymer network yielded by the polymerization of 3 wt% BAB in 5CB. The phase contrast optical micrographs below illustrate the results that the structure becomes finer as either the initiator concentration or the UV intensity increases, because the concentration of free radical species increases. This, in turn, increases the number of polymer aggregates (at fixed monomer concentration) and therefore reduces the size of the aggregate at the point of gelation.
 |
||||
| 0.03 wt% BME 0.06 mW/cm2 |
0.03 wt% BME 1.5 mW/cm2 |
0.3 wt% BME 1.5 mW/cm2 |
||
We are grateful to Professor Steven Hudson of the Macromolecular Science Department, CWRU, for providing all the microscope images in this section and for many helpful discussions.
 |
Introduction to Polymer Stabilized Liquid Crystals |  |
Polymer Stabilized Cholesteric Liquid Crystals |  |