This paper focuses on several dispersions that are of active scientific and/or technological interest in the Research Area at Bell Laboratories/Lucent Technologies. There are numerous other important dispersions, many of which are discussed by other WTEC workshop participants. These are photographic emulsions, inks, paper coatings, magnetic recording materials, pigments, and fillers.
During sol-gel processing a concentrated dispersion of colloids is chemically converted into a gel body. Subsequent drying and sintering leads to the final glass or ceramic product. This method has the potential of producing controlled structures of a variety of shapes, e.g., small particles, fibers, thin films, tubes, and plates. This method can be economical, and the processing is done at low temperatures.
A state-of-the-art colloidal silica casting process has been developed at Lucent Technologies that reproducibly yields tubes of pure silica with lengths of more than 1 meter. These tubes are used as overcladding material for optical fiber preforms. Fiber drawn from these preforms meets the standards for loss, strength, and dimensions.
There were several major technical challenges: achieving sufficient purity of starting materials; removal of refractory particles that lead to breakage during the fiber drawing process; and achieving very tight dimensional tolerances. A particularly difficult problem was the development of procedures for drying the gel bodies that avoid cracking due to stresses.
A successful sol casting, drying, purification, and sintering process has been developed that produces large overcladding tubes for fabrication of optical fiber preforms. The process uses a low-surface-area fumed silica dispersed in a basic aqueous solution. The sol is stable against gelation at high pH and can be centrifuged to remove large impurities. The sol is then cast into a mold after a hydrolyzable ester is added to reduce the pH slowly and start the gelation. The gel body is removed from the mold and dried over several days. Final heat treatment involves several steps: (1) pyrolysis of organics; (2) treatment in reactive halogen atmosphere to purify the body; and (3) sintering in He at temperatures approaching 1500oC.
The above technique allows production of large tubes equivalent to those made by drawing from synthetic silica boules and other shapes, with significant process and cost advantages. Sol-gel processing is also being evaluated for thin film processing in silicon VLSI and "silicon optical bench" (waveguide) applications.
Polymer and silica colloidal spheres suspended in solvents have been used as model systems to study the interactions between the colloids and their phase diagrams. Of interest are two-dimensional (2D) and three-dimensional (3D) suspensions and the influence that flat or modulated planes have on the structures that form.
The structure and properties of 2D and 3D assemblies of colloid particles are of intrinsic interest; in addition, the detailed particle real-space and time information obtained from these model systems is valuable for the fundamental understanding of phase transitions of ordinary "atomic" scale matter. In many instances a colloidal suspension can be used as a surrogate for studying the equilibrium properties of an atomic system because a suspension of submicron monodisperse colloid spheres in a solvent has a well-defined temperature by virtue of collisions between the colloid spheres and the solvent molecules. The interaction potential between the colloid spheres can be changed by adjusting their charge.
Recent scientific achievements include the elucidation of the liquid-hexatic-solid phase transition in 2D, the phases of charged spheres confined between walls, phases of binary systems (two different size spheres), the measurement of the interaction potential between colloids, the structure of hard-sphere glasses, and template-directed colloidal crystallization.
Colloidal crystals have been proposed as optical limiters and switches. Further interesting optical materials that could be made by self-assembly of colloidal particles include filters, waveguides, amplifiers, and modulators or photonic band gap components.
There are several groups in the United States working on all aspects of this field: University of Colorado, Boulder; University of Pennsylvania, University of Chicago, Princeton University, Clarkson University, University of Delaware, University of Pittsburgh, and Bell Labs.
Reflective displays are needed for portable products such as cellular phones, communicators, personal digital assistants, and hand-held computers. Incorporating displays that are reflective enough to be used without a backlight greatly reduces both the weight and power consumption of these products. The reflective displays currently used in portable devices are usually twisted or supertwisted nematic liquid crystal. One drawback of these displays is that the reflected light is attenuated by several passages through polarizers, resulting in a dull "black-on-gray" display rather than a bright "black-on-white" display.
A promising new technology for producing brighter displays is based on polymer dispersed liquid crystals (PDLCs), which are made of micron-sized liquid crystal droplets trapped in a polymer matrix sandwiched between glass plates. If no electric field is applied, the orientation of the liquid crystal is determined by the droplet shape and is random between neighboring droplets. In this state, the PDLC material will appear milky white and opaque, since light is strongly scattered as it passes through droplets of different orientation. By applying an electric field, one can orient all of the liquid crystals in the same direction, causing the material to become transparent. Thus, if a black backing is used in the display, the regions where the electric field is applied will appear dark. Since this technology requires no polarizers, PDLCs can be used to make reflective displays with a brightness and contrast similar to black ink on white paper. Moreover, these displays are simpler to assemble than twisted nematic displays, since they do not require alignment layers.
Several groups in the United States (Kent State, University of Pennsylvania, Bell Labs) have recently made advances in understanding the dependence of the electro-optical performance on droplet structure and size and how to reproducibly achieve the desired microstructure through better processing control. It was also discovered that surface interactions at the droplet walls have a profound effect on contrast and brightness and can be manipulated to yield PDLCs that require very low switching voltages.
A technologically related to PDLCs are nematic curvilinear aligned phase (NCAP) materials, where the liquid crystal is encapsulated in a polymer film by encapsulation or emulsification. Raychem Corporation has products that use NCAP technology.
In chemical mechanical polishing (CMP), colloidal dispersions are not part of the end product, but rather play a crucial role during processing. Tight control and thorough understanding of the process is intimately linked to the yield and quality of the product.
As integrated circuits become more and more complex with increasing layers of metallization and interlayer dielectrics, VLSI manufacturing needs tools to planarize the wafers between the deposition and processing steps. Over the past few years it has become clear that the method of choice will be CMP. During this process the wafer is pressed against a polishing pad, and the presence of silica, alumina, ceria, or zirconia colloidal slurries leads to the desired planarization. Factors that influence the effectiveness are, among others, the particle size and shape of the colloids, and their surface chemistry and stability.
Currently, vapor deposited silica is used as the interlayer dielectric material; in the near future, materials with lower dielectric constants will be implemented. Choices are fluorinated SiO2, porous SiO2, and polymers. How to planarize these materials is the subject of current investigation.
CMP is an integral part of the semiconductor industry roadmaps and is expected to play a crucial role in back-end wafer processing. All the major integrated circuit manufacturers have efforts in developing CMP. There has been strong publication and patent activity over the past three years (> 180 U.S. patents in CMP since 1984, with more than half of all patents issued since 1993). Several conferences, workshops, and symposia per year are devoted to the subject. There has also been a significant increase in patents by Japanese companies in the past year.
McKay, R.B., ed. 1994. Technological applications of dispersions. Surfactant science series, vol. 52. New York: Marcel Dekker.
Wedlock, D.J., ed. 1994. Controlled particle, droplet, and bubble formation. Oxford: Butterworth-Heinemann.
Brinker, C.J., and G.W. Scherer. 1990. Sol-gel science. San Diego: Academic Press.
MacChesney, J.B., D.W. Johnson, S. Bhandarkar, M.P. Bohrer, J.W. Fleming, E.M. Monberg, and D.J. Trevor. N.d. Electronic Letters. Submitted.
Colloids as Model Systems
Ackerson, B.J., and K. Schatzel. 1995. Phys. Rev. E. 52: 6448.
Chang, S., L. Liu, and S.A. Asher. 1994. J. Am. Chem. Soc. 116: 6739.
Murray, C.A., and D.G. Grier. 1995. Am. Sci. 83: 238.
_____. 1996. Ann. Rev. Phys. Chem. 47.
Pusey, P.N., et al. 1989. Phys. Rev. Lett. 63: 2753.
Russel, W.B., D.A. Saville, and W.R. Schowalter. 1989. Colloidal dispersions. Cambridge: Cambridge Univ. Press.
van Blaaderen, A., R. Ruel, and P. Wiltzius. 1997. Nature 385: 321.
Polymer-Dispersed Liquid Crystals
Amundson, K., A. van Blaaderen, and P. Wiltzius. 1997. Phys. Rev. E. 55: 1646.
Doane, J.W. 1994. MRS Bulletin 22.
Drzaic, P.S. 1995. Liquid crystal dispersions. New Jersey: World Scientific.
Chemical Mechanical Polishing
CMP-MIC. 1997. Proceedings, Second International Chemical-Mechanical Polish for ULSI Multilevel Interconnection Conference in Santa Clara, CA. Library of Congress No. 89-644090.
_____. 1997. Planarization for ULSI Multilevel Interconnection Conference, Workshop Visuals Book.