Lead Investigator: Kimberly Jones (Lead, Howard University)
Additional Investigators: Sanjib Sharma (Howard University)

Project Objectives:

• Develop an understanding of fouling by high-activity tank waste in a crossflow filtration system based on operational parameters;
• Develop appropriate parameters for a predictive AI membrane performance simulation model; and
• Recommend appropriate fouling management strategies for the filtration process that would reduce fouling and increase overall process (separation) efficiency.

Significance/Impact:
Vitrification is a critical treatment process for the immobilization of radioactive wastes and eliminates corresponding environmental threats for the tank wastes at the Hanford Site. Waste volume reduction is an important step to ensure the success of vitrification, which will remove excess water, concentrate solids, and optimize waste composition for reaction with glass-forming materials. Therefore, there is a critical need to develop the most suitable solid-liquid separation method

Crossflow filtration, specifically microfiltration, is one of the principal separation technologies that has been proven to be an effective method of dewatering waste by previous work at PNNL (Daniel et al., 2010). However, due to the unique properties of tank wastes (i.e., high solid content, high pH, and high ionic strength), persistent membrane fouling hinders the treatment process. In addition, the proposed Low Activity Waste Pretreatment System (LAWPS) poses the potential to have significant depth fouling, which is little understood at this time. Therefore, fouling will remain a serious problem during crossflow membrane filtration of tank waste and will significantly impact the overall timeline and costs associated with the remediation of the Hanford site.

Prior studies have shown that typical fouling mechanisms and conventional wisdom involved in planning and operating crossflow membrane systems may not apply to the unique feed streams from waste tanks. When challenged with a synthetic feed, the membrane performance continued to decay at long timeframes and did not reach a steady state. To address this issue and attempt to explain fouling behavior, a new model was proposed to more reasonably match theory, and experimental results (Schonewillet al., 2015), but further efforts are needed to determine the mechanistic causes of this behavior in order to recommend modifications to the process to improve filtration performance.

Public Benefits:
This research improves treatment efficiency via real-time monitoring and predictive maintenance with applicability to waste streams outside of DOE. The treatment techniques developed here are scalable to different wastewater treatment facilities and flexible to fit changing waste streams and treatment requirements. The result will be increased transparency to the affected community and cost savings to the public.

References:

Asghari, M., Dashti, A., Rezakazemi, M., Jokar, E., & Halakoei, H. (2020). Application of neural networks in membrane separation. Reviews in Chemical Engineering, 36(2), 265–310. https://doi.org/10.1515/revce-2018-0011

Daniel, R., Schonewill, P., Shimskey, R., & Peterson, R. (2010). Brief review of filtration studies for waste treatment at the Hanford site. PNNL-20023, Prepared for the U.S. DOE under Contract DE-AC05-76RL01830.

Daniel, R. C., Billing, J. M., Bontha, J. R., Brown, C. F., Eslinger, P. W., Hanson, B. D., Huckaby, J. L., Karri, N. K., Kimura, M. L., Kurath, D. E., & Minette, M. J. (2010). EFRT M-12 issue resolution: Comparison of filter performance at PEP and CUF scale. PNNL-18498 Rev 1; WTP-RPT-185 Rev 1, Pacific Northwest National Laboratory, Richland, WA.

Daniel, R. C., Billing, J. M., Burns, C. A., Peterson, R. A., Russell, R. L., Schonewill, P. P., & Shimskey, R. W. (2011). Filtration understanding: FY10 testing results and filtration model update. PNNL-20299, Pacific Northwest National Laboratory, Richland, WA.

Eren, B., Ileri, R., Dogan, E., Caglar, N., & Koyuncu, I. (2012). Development of artificial neural network for prediction of salt recovery by nanofiltration from textile industry wastewaters. Desalination and Water Treatment, 50(1-3), 317–328. https://doi.org/10.1080/19443994.2012.719743

Ghandehari, S., Montazer-Rahmati, M. M., & Asghari, M. (2013). Modeling the flux decline during protein microfiltration: A comparison between feed-forward back propagation and radial basis function neural networks. Separation Science and Technology, 48(9), 1324–1330. https://doi.org/10.1080/01496395.2012.736914

Liu, Q., Kim, S. H., & Lee, S. (2009). Prediction of microfiltration membrane fouling using artificial neural network models. Separation and Purification Technology, 70(1), 96–102. https://doi.org/10.1016/j.seppur.2009.08.017

Roehl, E., et al. (2018). Modeling fouling in a large RO system with artificial neural networks. Journal of Membrane Science, 552, 95–106. https://doi.org/10.1016/j.memsci.2018.01.064

*Rollock, R., Liu, Y., Ramamoorthy, M., & Jones, K. (2014). Understanding the fouling mechanisms of inorganic particulates in a microfiltration system. Presented at the 24th annual NAMS meeting, May 31 – June 4, 2014, in Houston, TX.

Schmitt, F., & Do, K. U. (2017). Prediction of membrane fouling using artificial neural networks for wastewater treated by membrane bioreactor technologies: Bottlenecks and possibilities. Environmental Science and Pollution Research, 24(29), 22885–22913. https://doi.org/10.1007/s11356-017-0046-7

Schmitt, F., Banu, R., Yeom, I.-T., & Do, K.-U. (2018). Development of artificial neural networks to predict membrane fouling in an anoxic-aerobic membrane bioreactor treating domestic wastewater. Biochemical Engineering Journal, 133, 47–58. https://doi.org/10.1016/j.bej.2018.02.001

Schonewill, P. P., Daniel, R. C., Russell, R. L., Shimskey, R. W., Burns, C. A., Billing, J. M., Rapko, B. M., & Peterson, R. A. (2012). Development of an S-Saltcake simulant using crossflow filtration as a validation technique. Separation Science and Technology, 47(14-15), 2098–2107.

Srivastava, A., et al. (2021). Response surface methodology and artificial neural network modeling for the performance evaluation of pilot-scale hybrid nanofiltration (NF) & reverse osmosis (RO) membrane system for the treatment of brackish groundwater. Journal of Environmental Management, 278, 111497. https://doi.org/10.1016/j.jenvman.2020.111497