Progress 07/15/18 to 03/14/19

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? Nothing Reported

Impacts
What was accomplished under these goals? Discussed below is the progress made towards the goals listed. To date, Via Separations has successfully constructed a 1.8" spiral module with selected balance of system components. Additionally, much progress is being made in parallel to generate 6 square feet of membrane area (2 sheets of 3 square feet) using a semi-continuous manufacturing process. More details follow. Customer coupon To date, Via has shipped large area sheets to six customers. The size of the sheet is dependent on the test rig available to the customer to test. Customers cannot be named due to confidentiality, but each of these customers has paid Via for the samples to indicate their buy-in and interest. Data is forthcoming from most of these entities. Narrowing of required operating phase space Further customer interviews have revealed the maximum operating temperature and pH required for a spiral wound nanofiltration unit in the processing of whey. In the original proposal, Via notes the requirements of 100 C, oscillating pH regimes and substantial levels of oxidizing agents. With additional understanding of external factors such as OSHA constraints, Via has narrowed and quantified the focus of efforts to 85C and pH 11. Further, mjany food processors are moving away from using oxidizers to clean their components, thus the pH is defined by the quantity of caustic required to clean the membranes and other elements of the system. Selection of balance of system components Selection of the permeate tube, permeate collector, feed spacer, and glue are complete. These have been informed by market research and individual component testing. Accelerated lifetime testing will inform the iteration of these components. Specifically, the selections are below: Permeate carrier: Hornwood 1848 Feed spacer: Naltx/PP 45 mils, 06 strands, diamond Glue: HB Fuller UR3543 The permeate carrier was selected through extensive testing with the GO membrane/support construction in cross-flow configuration. A sample set of a typical 100-hour duration experiment follows. The graph on the left shows a permeate carrier that cannot support the membrane under pressure conditions (the number associated with the data point is the pressure in PSI), causing it to collapse, increase flow rate too much, and exhibit poor rejection. The graph on the left shows a linear increase in flow rate with pressure, and corresponds to consistent solute retention. Figure 2: Data for permeate flow rate reveals a poor (left) and good (right) permeate carrier that is able to support the graphene oxide membrane under pressure conditions up to 400 psi. Feed spacers were also determined through cross flow testing. In this case, fouling, pressure drop, and ability to clean are improved or exacerbated by the choice of thickness, density, and cross section of the feed spacer. Specifically, Via has selected a mid-thick feed spacer (around 45mL) because thinner feed spacers are more prone to fouling, and feed spacers larger than this are hard to roll and too large to fit in the channels of test cells. As for the geometry, a mid-packed density feed spacer was targeted as it slows the velocity of the fluid and decreases the incidence of fouling; the diamond shape was simply chosen due to its widespread production in industry as compared to the square orientation. Finally, glue trials were conducted by testing the adhesion between components before and after exposure to elevated temperature and pH. Samples were soaked in pH 12 NaOH solution, pH 1 HCl solution and 3% bleach solution at 95 C for one week. After being taken out of solution and allowed none of the samples showed a significant weight change which would indicate the adhesive was decomposing and when attempting to pull apart the two pieces of support material, the layers of the support material failed before the adhesive. An example is shown in Figure 3. Figure 3: A glue sample from the glue selection design of experiments. This sample was soaked for 3 weeks in NaOH. Construction of spiral wound element Each of the components was integrated into a spiral wound element in order to test the efficacy of construction, evaluate failure mechanisms, and inform the team about the process. However, Via intends to partner with third party contractors to roll elements in the future. Two entities have been identified and will proceed with rolling spirals in the future. The images below show a spiral element rolled without the active graphene oxide layer. Until development of the semi-continuous manufacturing process is complete (see next section), the sheets of GO produced are not in the form factor required for a membrane. Figure 4: Glue lines are administered (left) to the stack of elements required for a spiral element (center) before rolling into a tightly wound configuration (right). Development of semi-continuous manufacturing process In order to generate enough area for a 1.8" spiral element, both scale up and transition from batch process to roll-to-roll must occur. Funded by a complementary grant, Via has made considerable progress towards the development of a semi-continuous (easily translatable to roll-to-roll process). This has involved the purchase and customization of automated drawdown machine. Photographed below, the tool is capable of making 3 square feet sheets, two of which will be stacked up in order to construct a spiral wound membrane. Ongoing efforts are focused on integrating the necessary chemistry into the process to conduct cross-linking, the proprietary process Via has developed to produce the selectivity required for dairy markets. Figure 5: Via's automated drawdown machine capable of fabricating three square feet of membrane material. Further work is required to match process conditions for Via's proprietary chemistry.

Publications