Multimodal Characterization of Separation Membranes in Three Dimensions using Scanning Transmission Electron Microscopy (STEM) Tomography and Electron Energy Loss Spectroscopy (EELS)

For decades, thin film composite membranes, in particular those employing an aromatic polyamide active layer, have become the dominant technology for the purification of water by nanofiltration and reverse-osmosis [1]. The extreme physical and chemical requirements demanded of these membranes (selectivity, permeability, fouling resistance, mechanical strength) combined with unique difficulties in studying their structure-property relationships on the nanoscale, have largely frustrated efforts to develop viable membranes based off new technologies. After more than 40 years, synthetic polymer membranes, many of which were discovered serendipitously, remain state-of-the-art. As the world confronts a changing climate, growing demand for water, and a transition away from fossil energy it is increasingly important that new effective, energy-efficient separation technologies be developed [2]. This will require advances in computational modeling and synthesis, informed by new and improved methods of material characterization.<br/><br/>TEM has long been a powerful tool for the nanoscale characterization of membranes; however, all TEM methods fundamentally produce some sort of two-dimensional projection of the sample. Although there can be great value in 2D top-down and cross-sectional imaging of membranes [3], the structure of a membrane must be studied in more than two dimensions to fully describe its function. Unlike conventional TEM-tomography methods, which have been successfully employed to study 3D morphological properties of separation membranes [4] high angle annular darkfield scanning transmission microscopy (HA-ADF STEM) scans a focused probe over the sample and is capable of producing images at the highest achievable resolutions whose contrast is proportional to sample density and thickness. Further, each scan line can be acquired with a different defocus, allowing large areas of the sample to be kept in-focus even at high tilt angles [5]. Taken together, these qualities allow for three-dimensional images that accurately reflect the sample’s scattering density to be computed from the HA-ADF tilt series data.<br/><br/>In this presentation I will discuss how HA-ADF STEM tomograms can be collected, calculated, and combined with electron energy loss spectroscopy data to produce three-dimensional images of chemistry and sample density to inform water transport simulations [6] and membrane synthesis. Particular emphasis will be given to the experimental and geometrical limitations on quantitativeness, such as the limited range of angles from which a membrane can be viewed in TEM, and how optimization and machine-learning techniques can be employed improve quantitativeness beyond the current state-of-the-art. Last, prospects for a ‘closed-loop’ unifying synthesis, characterization, and computational design will be discussed.<br/><br/>[1] Lee, K.P., Arnot, T., Mattia, D. A review of reverse osmosis membrane materials for desalination – Development to date and future potential. Journal of Membrane Science 2011, 370, 1, 1-22.<br/><br/>[2] Park, H., <i>et al.</i> Maximizing the right stuff: The trade-off between membrane permeability and selectivity. <i>Science 2017</i>, 356, 6343.<br/><br/>[3] Pacheco, F., <i>et al. </i>Characterization of isolated polyamide thin films of RO and NF membranes using novel TEM techniques. <i>Journal of Membrane Science </i>2010, 258, 51-59.<br/><br/>[4] An, H., <i>et al.</i> Mechanism and performance relevance of nanomorphogenesis in polyamide films revealed by quantitative 3D imaging and machine learning. <i>Science Advances</i> 2022, 8, eabk1888.<br/><br/>[5] Culp, T., <i>et al. </i>Electron tomography reveals details of the internal microstructure of desalination membranes. <i>Proceedings of the National Academy of Sciences </i>2018, 115, 35, 8695.<br/><br/>[6] Culp, T., <i>et al. </i>Nanoscale control of internal inhomogeneity enhances water transport in desalination membranes. <i>Science </i>2021, 371, 72-75.