Progress 06/01/14 to 01/31/16
Outputs
Target Audience:This technology is aimed at commercial producers of edible oil. The large players in agriculture, like Cargill and Bunge, are positioned to continue using conventional solvent-based methods of oilseeds processing. Due to this pattern, CMS plans to target smaller oil producers who are best in the growing niche organic foods market. CMS's membrane solution is attractive to an oil producer because it is a reliable dehydration method superior to evaporators and organic membranes in simplicity and durability. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?
Nothing Reported
How have the results been disseminated to communities of interest?We have been in contact with ADM, Cargill, TerViva. Reached out for Bunge and Williamette. Possible participation in Ag Innovation Showcase at St. Louis in Septemeber 2016. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Hydrophilic customized amorphous polymers (CAP) were synthesized and membranes were fabricated from them for permeation testing in a high pressure dead end cell. The hydrophilic CAPs that have been synthesized for this program are copolymers of Perfluoro-2,2-Dimethyl-1,3-Dioxole (PDD) and either EVEOH or PSEPVE, which are functionally -[PDD(-C5F8O2-)]m-[CF2CF-(OCF2CF(CF3)OCF2CF2X)]n- where X represents the functional groups of the co-monomers. That is where X = -SO2F in PSEPVE, and-CH2OH in EVEOH, In order to obtain optimal properties, the comonomer ratios of these two polymers were varied. The polymers that were synthesized were characterized by measuring the Tg with a differential scanning calorimeter. Polymers were further characterized for composition by FT-IR (Fourier Transform Infrared Spectroscopy) and molecular weight by solution viscosity. Flat sheet membranes were fabricated from each of the polymers characterized by casting them onto a porous polyacrylonitrile (PAN) support. The casting solutions used varied with the specific polymer, but typically were a mixture of a fluorinated solvent and an alcohol. The cast membranes were tested for water permeability in nanofiltration mode in the heated dead end cell. This cell is a model HP4750 stirred cell sold by Sterlitech. It was modified to provide heating capability specifically to meet the needs of this program. The membrane polymers were permeation tested with pure water to determine if they were suitable for use in nanofiltration of wheat germ oil/water mixtures. Of the seven polymers tested, two were found to exhibit water permeation rates that exceeded the pressure normalized flux goal of 0.35 liter/m2/hr/bar. The two polymers were PDD-EVEOH 100/50 and 100/100. The same copolymer in the 100/30 ratio also showed promise, but may have been too thick to provide the desired pressure normalized flux. Water permeation test data over a range of pressures for the PDD-EVEOH 100/100 polymer membrane was investigated. As expected the absolute water flux increases with pressure. However, the pressure normalized water flux decreases with pressure. This is an unusual result. The pressure normalized flux typically does not change significantly with pressure. Additional work is needed to verify these and the other test results with EVEOH copolymers and to improve permeation rates. The PSEPVE copolymer exhibited very low water flux. No additional work with these polymers is planned. To improve water permeation rates at lower water concentrations (30 to 40% water), pervaporation studies with CMS-3 membranes were initiated mid year 2015.Nanofiltration pressure normalized fluxes were found to not be sufficient at high water concentrations for the vegetable oil dewatering application. The concept of membrane pervaporation (PV) was integrated into the aqueous extraction process downstream from the three phase centrifugation step. In this step, the cream is separated from residual solids and a water rich phase. The cream, typically containing 30 to 40 weight % water, is fed to the proposed pervaporation membrane, after heating to a temperature in the range of 30° to 60° C, depending on the oil type. CMS glassy membranes separate by molecular sieving. Therefore water which is a small molecule, rapidly permeates through CMS membranes, while larger molecules permeate across much more slowly. The membraneis designed to reduce the water concentration sufficiently to destabilize the emulsion. The leaving water concentration required will vary with oil type (determined by what is required to break the emulsion), but will typically be 10 to 20 weight %. A pervaporation test rig was fabricated by personnel at Compact Membrane Systems (CMS). The membrane minimodule consists of hollow fibers in a PFA tube. Each hollow fiber consists of a composite CMS-3 membrane coated on the outside diameter of a cylindrical microporous support. The support is made of chemically and thermally resistant polymer. The membrane area used in this program ranged from 550 to 830 cm2. The pervaporation tests were done in recycle mode, with solution leaving the membrane being returned to the feed tank. During operation, water is removed selectively by permeation through the membrane while the oil are retained in the liquid. In addition to the work described above, design, construction, and delivery of the laboratory test apparatus to be used by the subcontractor, Oklahoma State University, for water permeation tests with wheat germ oil/water mixtures under flow conditions has been completed. Successfully demonstrated pervaporation-mode separation of water from emulsified oil/water mixture at 45° C. The water concentration was reduced from 50 wt% to 28 wt%. A conservatively low operating temperature (45° C) demonstrated membrane efficacy when low operating temperature is desired. The rate of water removal was compared to an oil/water mixture without emulsifier added and found to be better with the emulsifier present. This test clearly demonstrates that the proposed membrane process will be capable of breaking oil/protein/water emulsions. Demonstrated at the laboratory scale that the CMS pervaporation membrane can reduce water content in an oil water mixture from 50% down to as little as 1%, a much lower concentration than is necessary to break oil/protein/water emulsions. Separation of water was demonstrated to occur more rapidly at higher temperatures. For a canola oil/water mixture, water was removed 60% more rapidly at 60° C than at 45° C. Optimization of the operating conditions, such as sweep gas flow rate and temperatures up to 90° C, is expected to improve the temperature advantage further. It may be noted that temperatures up to 90° C are considered safe for soybean proteins. Separation of water from corn germ extract solution, wheat germ extract solution, and a canola oil water mixture of similar composition were demonstrated at 60° C. The rates of water removal from corn germ extract solution, wheat germ extract solution, and a canola oil water mixture were similar, as indicated by the similar slope of the water concentrations as a function of elapsed time. The cost of the proposed membrane pervaporation process for breaking the oil/protein/water is 5.0 cents per gallon of oil at a plant capacity of 1 ton per hour and only 3.6 cents per gallon at a plant capacity of 2 tons per hour, based on projected performance of the commercial membrane. For premium organic oil products which typically sell for about $7 per gallon, these are very low charges, representing less than 0.75% of the selling price. Fluorinated membrane constituents, such as perfluorooctanoic acid (PFOA, C8), were not found in the processed solvent systems up to the studied detectable limit of 0.5 ppm. Therefore, we are not anticipating any regulatory issues associated with the use of fluorinated polymers in vegetable oil emulsion breaking. CMS-7 (87%PDD w/TFE) membranes have much greater water permeance than CMS-3 (65%PDD w/TFE) but CMS-7 was not available in the Phase I program. The PDD-PPVE membranes may have even greater permeance than the CMS-7 membranes. (PPVE is CF2=CFOCF2CF2CF3). Therefore, the focus of the work in the proposed Phase II and the commercialization effort to follow will be with CMS-7 and PDD-PPVE membranes.
Publications
|
Progress 06/01/14 to 05/31/15
Outputs
Target Audience:
Nothing Reported
Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?
Nothing Reported
How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals? During the next reporting period,membrane material selection,verification of initial flux data, and optimization of the polymer membrane chemistry will be evaluated. In addition, Long term stability tests, membrane material leaching tests, and a preliminary economic analysis will be performed.
Impacts
What was accomplished under these goals?
Hydrophilic customized amorphous polymers (CAP) were synthesized and membranes were fabricated from them for permeation testing in a high pressure dead end cell. The hydrophilic CAPs that have been synthesized for this program are copolymers of Perfluoro Dimethyl Dioxole (PDD) and either EVEOH or PSEPVE.In order to obtain optimal properties, the co-monomer ratios of these two polymers were varied.The polymers that were synthesized were characterized by measuring the Tg with a differential scanning calorimeter. Polymers were further characterized for composition by FT-IR (fourier transform infrared spectroscopy) and molecular weight by solution viscosity. Flat sheet membranes were fabricated from each of the polymers characterized by casting them onto a porous polyacrylonitrile (PAN) support. The casting solutions used varied with the specific polymer, but typically were a mixture of a fluorinated solvent and an alcohol. The cast membranes were tested for water permeability in nanofiltration mode in the heated dead end cell. This cell is a model HP4750 stirred cell sold by Sterlitech. It was modified to provide heating capability specifically to meet the needs of this program. The membrane polymerswere permeation tested with pure water to determine if they were suitable for use in nanofiltration of wheat germ oil/water mixtures. Of the seven polymers tested, two were found to exhibit water permeation rates that exceeded the pressure normalized flux goal of 0.35 liter/m2/hr/bar. The two polymers were PDD-EVEOH 100/50 and 100/100. The same copolymer in the 100/30 ratio also showed promise, but may have been too thick to provide the desired pressure normalized flux. Water permeation test data over a range of pressures for the PDD-EVEOH 100/100 polymer membrane was investigated. As expected the absolute water flux increases with pressure. However, the pressure normalized water flux decreases with pressure. This is an unusual result. The pressure normalized flux typically does not change significantly with pressure. Additional work is needed to verify these and the other test results with EVEOH copolymers and improve permeation rates. The PSEPVE copolymer exhibited very low water flux. No additional work with these polymers is planned. In addition to the work described above, design, construction, and deliveryof the laboratory test apparatus to be used by the subcontractor, Oklahoma State University, for water permeation tests with wheat germ oil/water mixtures under flow conditions has been completed. The system is presentlyin use at OSU.
Publications
|