3D measurement of fouling in membrane filtration: towards model based investigation and design

由于上游滴度和相关负担的增加，膜过滤系统的污染是一个日益严重的问题。这些复杂的进料中含有大量不需要的物质，这些物质会通过几种污染机制对性能产生不利影响，这些污染机制会降低膜结构的渗透性，并改变筛分系数[2]。在这里，我们的目标是通过测量和可视化方法来更好地理解这些机制，这将为CoE愿景的建模和仿真工作提供信息，从而改进膜的设计和操作。目标-推进成像方法来测量膜内污染物的位置和当前生物处理挑战的饲料类型-结合共聚焦显微镜和x射线CT技术，以新的分辨率确定膜内污染物的位置和种类。-在多个长度尺度上使用这些方法，从最初的小样本到折叠纸墨盒和单道TFF系统，以确定放大方法的影响。-将三维数据与建模方法相结合。项目描述过滤过程中的污垢是工业生物工艺中常见的问题[1,2]，需要考虑各种目标产品尺寸和必须去除的杂质。在伦敦大学学院和CoE内部，各种高分辨率成像技术已被应用于可视化和表征Pall分离介质，包括x射线CT、共聚焦显微镜、电子显微镜和聚焦离子束显微镜[3-5]。这表明，以互补的方式利用多种技术可以克服单个方法的缺点，并最大限度地收集有关样本的数据。在这个项目中，共聚焦显微镜和超荧光显微镜将分别对标记饲料进行光学切片和大视场成像，以确定不同特征（如孔径和不对称性）的Pall过滤膜内被捕获物质的位置和组成。该研究将建立在CoE MSc项目的基础上，该项目利用共聚焦显微镜应用于双层膜系统，研究了脂质体在Pall膜内的包裹位置。通过使用不同大小和组成的脂质体材料来模拟慢病毒等产品和潜在的污染物，例如可以在成像前单独标记的细胞外囊泡，可以进一步实现这一目标。这些模拟将被设计成与其他CoE EngD项目（如慢病毒制造）保持一致，生成的数据将支持CoE愿景Gronemeyer等人（2014），[2]，Fallahianbijan等人（2019）[3]，Johnson等人（2017）[4]Johnson等人（2018），[5]Jackson等人（2014）提议的时间表第1年：成像和分析培训（例如共聚焦），脂质体的初始膜实验第2年：扩展介质（包括双层）和配方（移动到慢病毒的脂质体模拟）第3年：将结果与CoE中的其他成像数据结合，研究多长度尺度成像第4年：完成研究，将透镜模拟与实际进行比较，CoE成像协作，提交论文

Bioprocess challenge the project seeks to address Fouling of membrane filtration systems is an ever-increasing issue due to upstream titre and associated burden increases [1]. These complex feeds contain a large array of unwanted material that can detrimentally affect performance through several fouling mechanisms that can reduce permeation through membrane structure and change sieving coefficients [2]. Here we aim to develop a better understanding of these mechanisms using a measurement and visualisation approach that will inform modelling and simulation efforts core the CoE vision of enabling improved membrane design and operation.Objectives- To advance imaging approaches to measure foulant locations within membrane and feed types representative of current bioprocessing challenges - To combine confocal microscopy and x-ray CT techniques to determine at new levels of resolution the location and species of foulant within membranes.- Use these methods at multiple length scales, from small samples initially up to pleated sheet cartridges and single pass TFF systems to determine the impact of scale-up approach.- Interface the three dimensional data acquired with modelling approaches.Project DescriptionFouling during filtration is a commonplace issue for industrial bioprocesses [1, 2], with a variety of target product sizes to consider and impurities that must be removed. At UCL and within the CoE various high-resolution imaging techniques have been applied to visualise and characterise Pall separation media, including x-ray CT, confocal microscopy, electron microscopy and focused ion beam microscopy [3-5]. This has shown that utilising multiple techniques in a complementary manner can overcome the shortcomings of individual methods and maximise the data gathered about a sample. In this project confocal and super-fluorescence microscopy, will enable optical slicing and large field of view imaging respectively on tagged feeds to identify the location and composition of entrapped material within Pall filtration membranes of varying characteristics such as pore size and asymmetry.The research would build upon a CoE MSc project that investigated the location liposome entrapment within Pall membranes using confocal microscopy applied to dual layer membrane systems. It would further this by using liposome material with differing sizes and compositions in order to mimic products such as lentivirus and potential foulants e.g. extracellular vesicles that could be individually labelled prior to imaging. These mimics will be designed to align with other CoE EngD projects such as in lentivirus manufacture, and the data generated will support the CoE vision.[1] Gronemeyer et al. (2014), [2], Fallahianbijan et al. (2019) [3], Johnson et al. (2017)[4] Johnson et al. (2018), [5] Jackson et al. (2014)Proposed timelineYear 1: Imaging and analysis training (e.g. confocal), initial membrane experiments with liposomesYear 2: Expand media (including dual layer) and formulation (move to liposome mimic of lentivirus) Year 3: Combine results with other imaging data in the CoE, investigate multi-length scale imagingYear 4: Complete studies, compare lenti mimic to actual, CoE imaging collaboration, submit thesis