Thrust Area C:  Rate Processes

 

Accomplishments for 11/1/99-7/31/00

 

Thrust Area C, our largest, deals primarily with solving engineering problems. Our first year has focused on equipment design and installation. Here we have divided Thrust Area C into sections: (1) membranes, (2) surfactant-based cleaning processes and (3) polymer systems.

 

Membranes

The objective of the membrane separation subgroup is to develop the ability to separate CO2 from industrial gas streams, organic molecules and colloidal particles to reduce re-compression costs and aid in CO2 recycling. The membrane subgroup in has cooperated to establish system designs for the three locations to allow testing membranes with diverse structures and properties under relevant supercritical CO2 conditions.  Sketches of the proposed system at NCSU and the recently completed prototype at UT Austin for testing high performance polymer membranes are shown in Appendix G Figures II.C.1 and II.C.2, respectively.

CO2 Separation Properties of Novel Substituted Polyacetylenes (12)   In the conventional method of measuring gas permeability through polymeric membrane, pressure is applied across the membrane while the low-pressure-side is at vacuum or ambient state. This condition, however, is not appropriate for the current system because large pressure difference between high- and low-pressure-side can destroy the membrane and/or supporting components. The idea of this research is to keep the downstream pressure high (much higher than the ambient pressure). In order to do this we modified the conventional gas permeation system to include a gas booster to compress gas up to ~300 bar; a valve connect/disconnect up- and downstream; and flow controller to vent a small portion of downstream flux.   See Figure II.C.1 in Appendix G.

 

Separation and Recovery of SCF CO2 Using Microporous UF/NF Membranes (13)  Microporous ceramic and metallic membranes in crossflow configuration (as ultrafiltration – UF, and nano-filtration – NF) are the focus of research at NC A&T. To date we have conducted a thorough literature search on membrane-based SCF CO2 separation.  For our exploratory research, we identified U.S. Filters’ Membralox® membranes as candidates for UF and NF experiments. The Membralox® membrane and module housing are designed for a maximum operating pressure of 115 psi while our SCF CO2 filtration system will require operating it at about 1100 psi. We acquired Membralox® membranes and a module housing from US Filter. Currently we are working on retrofitting the module housing to handle our required operating pressure. We have the preliminary design of the experimental high-pressure crossflow membrane filtration system and are working on identifying appropriate vendors for high-pressure pumps and valves.

 

Robust Sorption-Diffusion Polymeric Membranes for Service in High Pressure CO2 (14)  A prototype high-pressure system for testing membranes under aggressive feed conditions has been built.  The system is rated to feed pressures of 4000 psi with controllable transmembrane pressure drops.  The pressure of the incoming gas stream from a standard source of CO2 is being achieved with a Haskell booster pump capable of providing 4000 psi feeds independent of incoming pressure.  Supporting the membrane properly for separations utilizing high feed pressures with low permeate pressures was challenging but has been accomplished.  Under extreme conditions, the support provided by Millipore was severely deformed.  Although the cell is rated to 10,000 psi, the support required modification to suit our testing needs.  The membranes are now supported with a sintered metal disk resting in the modified Millipore cell.  Integrating the Haskell booster pump with the modified Millipore cell has allowed successful testing of transport properties of membranes with upstream pressures as high as 3000 psi for a range of downstream conditions.  A system schematic is shown in Appendix G Figure II.C.2.

Mass Transfer at Interfaces

 

The goal of the following studies is to utilize experimental systems for continuous, on-line monitoring of decontamination processes to evaluate key parameters in the cleaning domain.

An Image Sensing System for CO2-Based Surfactant Decontamination Processes (16)  This program has three parts: development of techniques to monitor decontamination rates, on-line analysis of surface contaminants and phase equilibrium of polymer systems.

I. Real-time Rate Monitoring of Decontamination of Solid Surfaces in SCCO2: Development of reconfigurable computing hardware has been stopped in favor of the Quartz Crystal Microbalance (QCM) technique.  QCM can be used to monitor in real time decontamination rates of solid surfaces.  This technique is very accurate and reproducible and gives rate and thickness resolutions to Angstroms per second. We are designing the experimental set-up, using those components of the original design we can.  Some significant redesign will be required as QCMs are traditionally designed for vacuum systems. Selection and sizing of components is underway.

II. Precision Surface Cleaning with Liquid/SCCO2: This project treats the removal of cutting and machine oils, silicone oils, and hydraulic fluids from variety of industrial substrates with liquid / SCCO2 to <10 ug/cm2.  The equipment (Appendix G Figure II.C.3) will provide: (1) the rate of the on-line removal of contaminants from the substrates, (2) a spectroscopic surface analysis to determine the contaminants left on the substrate after cleaning and (3) solubility of the contaminants in the liquid / SCCO2. Most of this system was purchases using EPA funding.

III.  Phase Equilibrium of Polymer Systems: Determination of phase boundaries and compositions are required for the safe and efficient operation of polymer processes. We will determine the solubility of a polymer by visualizing its dew point.  The advantage of this technique is that neither sampling nor extraction is required.  A complete system purchased will be installed at NCSU by Fall 2000.  The major system components are variable volume cell, mechanical stirrer, syringe pump, sapphire window to video camera, VCR & TV, and PC interface.

 

Surfactant/Supercritical CO2 Cleaning of Contaminated Substrates (20)  Research to date has examined the solubility of several commercial surfactants in carbon dioxide. A class of fluorinated surfactants has been found that dissolve in CO2. The cleaning performance of this surfactant/supercritical fluid (SSCF) mixture w/ and w/o water has been examined for the removal of model contaminants from metal surfaces.  The effect of oily, polar and non-polar contaminants has also been examined for various CO2-based systems on metallic substrate cleaning. Additional funding has been secured from the Department of Energy and 3M Company.

 
Polymer Systems

 

Formation of Reversible Networks (21)  A “sticky reptation” model has been developed for the dynamics of entangled solutions of associating polymers with many stickers per chain. The process of making and breaking reversible bonds between associating groups (stickers) controls the dynamics of associating polymers. At a high degree of association there are very few unassociated stickers. It is very difficult for a sticker to find a new partner to associate with after breaking the bond with an old one. Typically a sticker returns to its old partner following an unsuccessful search for a new one, prolonging the effective lifetime of reversible bonds. In the “sticky reptation” model the search for a new partner is restricted to a part of the tube confining the entangled chain. Another important effect is the increase of the fraction of the inter-chain associations at the expense of the intra-chain ones with increasing polymer concentration. The “sticky reptation” model predicts very strong concentration dependence of viscosity.

 

Rheology of Polymer Melts and Solutions in CO2 (17)  A high-pressure extrusion slit die rheometer was constructed to measure the viscosity of polymer melts plasticized by liquid and supercritical CO2. Experimental measurements of viscosity as a function of shear rate, pressure, temperature and CO2 concentration were conducted for three commercial polystyrene melts.  CO2 was shown to be an effective plasticizer for polystyrene, lowering the viscosity of the polymer melt by as much as 80% depending of the process conditions and CO2 concentration.  Existing theories were used to develop a free volume model for predicting the effects of CO2 concentration and pressure on polymer melt rheology, dependent only on material parameters of the polymer melt. This model provides a simple predictive equations to quantify CO2 plasticization effects.

The High-Pressure Magnetically Levitated Sphere Rheometer project focuses on the development a novel rheological device to measure the physical properties of polymer melts with dissolved CO2 and polymer solutions in CO2.  Unlike the extruder system, which enables only viscosity measurements at processing conditions and of very high viscosity systems, this system will enable measurement of lower viscosity materials. The device operates by measuring the difference in the magnetic intensity required to maintain the position of a levitated sphere while the fluid is at rest. Rheometer design has been completed and construction is underway.

Coatings from CO2 (C6)  Significant progress has been made in the area of coatings using liquid CO2.  Novel negative photoresists and photoacid generators that are soluble in CO2 have been synthesized.   Carbonell and students have developed a high-pressure spin-coating apparatus and succeeded in creating uniform one-micron thick layers on five-inch silicon wafers using liquid CO2 as a solvent.  These are being exposed and developed in CO2 to test their resolution in photolithography. A free meniscus coating method has been developed to create sub-micron lubricant films on magnetic disk drives.  Both of these tools are extremely promising for many applications, especially when combined with new CO2-soluble polymers being developed at UNC.

 

Plans for 8/1/00-7/31/01

 

Next year, the main focus of Thrust Area C will be to better coordinate individual research programs along the lines of this report.  For example, research by C. Grant and K. Roberts utilizes different approaches and systems to study cleaning; increased collaboration should enhance the overall productivity. Natural collaboration with other Thrust Areas in the Center will be encouraged.  For example, the need for both solubility and phase behavior information from Area B is a critical component to understand decontamination, polymer and other processes.

 

Membranes Over the next 12 months, construction of systems for ceramic and metallic membranes will be completed at NC A& T and at NC State (Appendix G Fig. II.C.1).  The prototype at UT Austin  (Appendix G Fig. II.C.2) will be optimized using ideas suggested from both of the North Carolina groups.  Performance data for the various membrane materials and membrane types in the presence of pure supercritical CO2 will be collected and analyzed over the next 12 months.  Selectivity of the various membranes under realistic supercritical feed conditions will be studied as a function of model solutes in supercritical feeds.

 

Mass Transfer at Interfaces  Construction of equipment will be complete and experimentation begun. The set-up will be able to control variables and define flow field or static environment to provide data allowing for development of model equations for decontamination in liquid and SCCO2. These variables include shear, pressure, temperature, amount and concentration of contaminant and surfactant, co-solvents, etc.  Initially, naphthalene will used as the contaminant and the QCM surface will be coated with simulated stainless steel.   Identification of reverse micelle mixtures with H2O and relevant phase equilibria for SSCF mixtures will be determined. The effects of P, T, surfactant molecule-type, contaminants and composition on cleaning performance of the SSCF systems will be determined. Mass transfer effects will also be examined. Preliminary phase equilibria models will be developed.  SSCF mixtures will be evaluated to decontaminate fibrous and polymeric substrates. The performance of the mixtures will be analyzed using the predictive models of surfactant/supercritical CO2 equilibria. Analysis of reverse micelle cleaning systems will be conducted using techniques of Carbonell and Grant (NCSU).  F and Si based surfactants from Desimone (UNC) will be tested.

 

Polymer Systems The next stage in the study of reversible networks is to develop a Monte Carlo program to study the phase behavior of associating polymers in compressible solvents. The result of this study will be a phase diagram of polymer/compressible solvent binary system as a function of polymer concentration, polymer molecular weight, strength and number of the reversible bonds per chain, CO2 pressure and temperature. Construction of high-pressure magnetically levitated sphere rheometer will be completed and studies undertaken on steady, low shear properties of various systems. The effects of CO2 concentration and swelling on rheology will be investigated.