Current Projects
Overview of Reactive Distillation
We are currently studying reactive distillation in collaboration with Professor Dennis Miller. Reactive distillation is ideal for reactions that are difficult to drive to completion without separation of one of the products. Such reactions are called 'equilibrium limited'. As an example, consider the formation of ethyl lactate.
Ethyl lactate is an environmentally friendly solvent that can be made from lactic acid and ethanol:
Ethyl lactate and water have different volatilities. By running the reaction in a distillation column, the water goes up the column permitting the remaining lactic acid to be driven to form ethyl lactate. In a batch or flow reactor, the precence of water prevents the complete conversion of lactic acid. The principle of reaction distillation is shown in the schematic below. A reaction catalyst is in the center of the column. The colors are intended to help illustrate where the components are distributed.
VLE for Reactive Distillation
|
Example P-x-y data
|
The lab's custom P-x-y apparatus provides very stable and reproducible sampling of VLE data. Frequently P-x-y experiments do not provide vapor samples. As shown at left, the Lira group is able to provide reproducible samples of the vapor phase. The data shown at left are from multiple runs starting from both pure compositions, which demonstrates the stability of the apparatus. The need for these data are for upgrading of organic acids derived from bioprocessing. These acids are valuable feedstocks for chemicals from biomass. Reactive distillation is an attractive way to obtain products when the equilibrium constant for the reaction is near unity, because it combines separation with the reactor. However, for reliable process modeling, data of this type are needed. This project is in collaboration with D.J. Miller, Department of Chemical Engineering and Materials Science at MSU. |
Molecular Simulation for Property Prediction
| Thermodynamic properties such as vapor pressure are needed to design reactive distillation columns. Vapor pressures are not available for many bio-derived chemicals. We are applying a molecular simulation technique SPEAD (Step Potentials for Equilibrium and Dynamics) largely developed by Professor Richard Elliott of the University of Akron. The technique uses hard molecule simulations upon which the attractive potentials are superimposed. For simulations the molecules are divided into 'united atoms' to represent functional groups as illustrated on the right for butane. The united atoms sites are assiged a raduis and potential energy parameters. The purpose of this work is to parametized united atom functional groups that occur in bio-dervied chemials. The figure on the right shows the predictions of vapor pressure for some secondary alcohols. |
|
Probing sequence dependence of siRNA-protein interactions by molecular simulation
RNA interference (RNAi) is a natural phenomenon resulting in potent and primarily specific gene silencing, potentially facilitating disease diagnosis and treatment, and improving our understanding of biological processes. RNAi is initiated in cells by the presence of short interfering RNAs (siRNAs). Unfortunately, these siRNAs can elicit immune responses in mammalian cells, limiting their functionality. The two known pathways responsible for this immune response are through Protein Kinase R (PKR) and Toll-like Receptor 3 (TLR-3). Both PKR and TLR-3 bind dsRNA and signal an immune response that can result in cell death. As both proteins possess similar RNA-binding domains, a clear understanding of the interactions that occur between the RNA and protein domain would be helpful in designing sequences that avoid being bound by the proteins. Here, we describe ongoing efforts in computational molecular modeling of the RNA-protein domain interactions.
Five configurations have been selected to investigate differences in siRNA binding to the Xenopus laevis RNA-binding Protein A, which possesses a similar binding domain to PKR and TLR-3. The basis for comparison is a simulation using the RNA sequence and binding domain crystallized by Ryter and Schultz (1). Next, the RNA sequence is modified to reflect the sequence and structure of a siRNA used experimentally for targeting the green fluorescent protein. Additional simulations are being performed with and without overhangs, as the overhangs that are typical of siRNAs are hypothesized to contribute greatly to recognition by the binding domains in the proteins. We have also rotated the siRNA 180 degrees longitudinally with and without overhangs to investigate energetic differences dependent upon the siRNA sequence.
Simulations are performed using CHARMM (version 27 parameter set), in an aqueous environment with sodium counterions. Simulations were started in September, 2006 and will proceed until 30 ns of real time has been simulated to allow for sufficient relaxation and sampling of conformational dynamics on the nanosecond time scale by the protein-RNA complex. This project is in collaboration with Professors S. Patrick Walton of CHEMS and Michael Fieg of Biochemistry and Mocular Biology.
Below is a short movie of a simulation of the interaction over a very short period of 70 ps. The RNA is the double-stranded helix and the two protein binding sites are the other ribbons. Water and sodium are hidden for the movie. The RNA and binding site structures are simplified for the purposes of display, but the simulations are carried out with representation of all atoms and bonds.
(1) Ryter, J.M.; Schultz, S.C. "Molecular basis of double-stranded RNA-protein interactions: structure of a dsRNA-binding domain complexed with dsRNA," EMBO J. 17, 7505-7513 (1998).
Improved Biodiesel Processing
Biodiesel is of substantial interest to supplement petroleum fuels. It is methyl esters of vegetable oils. It has the advantage of having zero sulfur emissions and has good lubrication properties for engine components. Currently biodiesel is blended with petroleum diesel and as little as 5% biodiesel is called 'biodiesel'. In collaboration with Professor Dennis Miller we are developing technology to apply reactive distillation to biofuels.
A significant portion of this effort is the conversion of the normal by product glycerol to acetals. Glycerol acetals have boiling points in the same range as biodiesel, and have been evaluated at concentrations up to about 5 wt% as additives. Currently the biodiesel producers get little value for the glycerol because it is contaiminated with catalyst. By incorporating acetals into biodiesel the total mass produced per kg of feed can be increased by 16%. To make this a reality, we are investigating catalysis and the options for continuous processing of biodiesel using heterogeneous catalysts. As an interesting note, glycerol acetals are present in wine, so are not expected to create health concerns.
Recovery of Succinic and Acetic Acid from Fermentation
Succinic acid is expected to be a major platform chemcial of the future bioeconomy. Succinic acid can be converted to the ethyl ester, and then used to create polybutylsuccinate (PBS). PBS is a flexible polymer much like polyethylene (milk containers) and polypropylene (door handles, grips, etc.). Diethyl succinate is also useful as a solvent.
This project is in collaboration with Diversified Natural Products, a Michigan company, and Dennis Miller of the CHEMS department. We are developing technology to recover succinic acid and acetic acid simultanously from fermentation broths. We expect to use reactive distillation as well as other purification/separation technologies.