Apr 23, 2024
4:00pm - 4:15pm
Room 439, Level 4, Summit
Nurila Kambar1,Cecilia Leal1
University of Illinois at Urbana-Champaign1
Nurila Kambar1,Cecilia Leal1
University of Illinois at Urbana-Champaign1
Membranes play an essential role in diverse engineering fields such as biomedicine, energy generation and water treatment. Thus, development of new types of membranes with tailored properties has been an active area of research. Two–component hybrid materials where phospholipid (PL) membrane stacks are hybridized with synthetic block copolymers (BCP) are of great interest due to their ability to combine the unique properties of two different materials. Hybrid membranes can demonstrate a wide range of self-assembled structures and properties that can be tailored to specific applications. The nanoscale arrangement of hybrid membranes into distinct lateral domains with varying structures and compositions is considered crucial for membrane functionality. The precise understanding of this organization has been challenging, primarily due to the absence of direct methods for probing nanoscale membrane features. Utilizing machine learning for the domain analysis of hybrid multilamellar nano-sized vesicles in cryo-TEM images, we are able to precisely generate detailed high-resolution 2D thickness maps of the membranes. These maps provide insights into the distribution of polymer-rich, lipid-rich, and well-mixed domains within the vesicles. Despite the optical microscopy indicating homogeneous mixing in the hybrid membranes, we reveal the coexistence of two distinct membrane structures within homogeneously mixed lipid-polymer hybrid vesicles. In this study, we highlight the significance of our semi-automated technique for directly iamging nanodomains in both biomimetic and biological membranes.